This item is only available as the following downloads:
1 TALKING ABOUT PUBLIC HEALTH RISK: NEWS COVERAGE OF THE NDM 1 IN INDIA, UNITED KINGDOM, AND THE UNIT E D STATES By BIJIE BIE A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQ UIREMENTS FOR THE DEGREE OF MASTER OF ADVERTISING UNIVERSITY OF FLORIDA 2012
2 2012 Bijie Bie
3 ACKNOWLEDGMENTS I would like to thank first and foremost, my chair, Dr.Treise for her incredible support and motivation. I feel so lucky to hav e you as my chair. Every time I felt lost or depressed, your smile, patience and intelligence gave me the courage and confidence to continue. Also, my sincere thanks and appreciation go to my committee members, Dr. Walsh Childers and Dr. Goodman, for thei r patience and guidance. Dr. Walsh Childers taught me how to gather information and how to be a careful writer., and Dr. Goodman gave me a lot of good suggestions and interesting ideas. I was fortunate to have you as a source of support and knowledge. It i s my great honor to be your student. I feel so grateful to my wonderful friends, Jing Bai, Fangfang Gao and Jung A Kim, who were always there to talk with me and comfort me. This thesis would not have been completed without their suggestions and encourage ment. Thank you to Becky Meng Zhang, you were always there to listen to everything I had to complain and answer my strange questions. Thank you also to Dennis Frohlich for offering me your time and effort with the coding work. I would like to express my d eep gratitude to my best friend, Lei, for putting up with me, talking with me and encouraging me to do whatever I like. Thank you for being there for me through some of the most difficult times in my life. Thank you for everything that has been and will be Last but not least, I would like to extend my love and sincere thanks to my family
4 in China: my father, my mother and my grandfather. Thank you for your love, trust and support. You will always be my most precious possessions. I love you and thank you all.
5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ........................................................................................3 LIST OF TABLES................................................ ....................................................7 ABSTRACT..............................................................................................................8 CHAPTER 1 INTRODUCTION.... .............................. ................................................................9 The Goal of the Study ...............................................................................................9 Background I nformation A bout NDM 1....................................................................... 9 Treatment of NDM 1 ....................................... ................................................. .... .... .. 11 Significance of the S tudy ........................................................................................ .. 12 2 LITERATURE REVIEW................ ...... ....................... .......................................1 5 Media C overage of H ealth R isks ........................................... .............. .. ........... ...... .. 15 Social C ontext in India, the United King dom and the United States ................... .. .. 15 Society of India, the United Kingdom and the U.S. ..... ............................ ....... ... 15 Health C are S ystems in India, the United Kingdom and t he United States.... ..... 19 Use of A ntibiotics in India, the United Kingdom and the United States..... .... .... 21 Media S ystems in India, the United Kingdom and the United States .............. ... 22 Prevalence of NDM 1 in India, the United Kingdom and the United States ....... .. 23 Health L iteracy.................................................................................................... ....... 24 Definition of L iteracy ................ ............................................................... ...... .. 24 Definition of H ealth L iteracy ..... ......................... ...................................... ..... .. .. 24 Health O utcomes and B ehaviors A ssociated W ith H ealth L iteracy....... ... 25 Health L iteracy and M edia L iteracy ...................................... ........................... ... 2 6 Definition of M edia L iteracy ...................................................................... ... 26 Health L iteracy in India, the United Kingdom a nd the United States.... ... 27 Media L iteracy E ducation i n the United Kingdom, the United States and India .. 28 Training of Health Journalists i n Three Countries................................. .................. .. 30 United Kingdom ............................................................................................ .... 30 United States ................................................................. ................................ .... 3 2 India .............................................................................................................. .... 33 Differences of H ealth J ournalists T raining A mong T hree C ountries .... ..... ......... 34 Theoretical Framework .......................................................................................... .. 35 Psychometric P aradi gm............................................................................... ..... 35 Risk C haracteristics .......................................................................................... 37 Risk P erception and M edia C overage................................................................ 3 9 Hypotheses .................................................................................................. .............39 3 METHOD.............................................................................................. ... 45 Design ............................................................................. ........................... ..... 45 Sample.................................................................................................................. ..... 45 Procedure ................................ .............................................................................. ... 46
6 Measures......................................................................................................... .... 46 4 RESULTS ...................................................................................................... .... 50 Review of M ethod ......................................................................................... ......... ... 50 Descriptions of the S ample ......................................................................... ..... 50 Intercoder R eliability ........ ..... ................. .......................................................... .. 50 Statistical A nalysis............................................................................................ .. 51 Review of Fin dings .......................................................................................... .... 51 H1a: Presence of W orst C ase S cenarios......................................................... .. 52 H1b: Placeme nt of W orst C ase S cenari os .......................... ............................ .. 52 H2a: The P resence of L oaded W ords .............................................................. .. 53 H2b: Placement of L oaded W o rds ........... ................................................... ...... 5 4 H3a: Use of U ncertai n W ords...................... ................................................... ... 54 H3b: Placement of U ncertain W ords .................... ........................................... .. 54 H4a: Risk I nformation A bout H uman I nfection/ D eath............ ........................ .... 55 H4b: Risk I nformation A bout F inancial L oss........... .............. ........................... .. 55 H5a: Personal P rotection I nformation .............................. ............................. ... 5 6 H5b: Societal P rotection I nformation............................. .................................... 56 H6a: Information A bout R isk C omparison to K nown R isks P resence of O ther S uperbugs and I nformation T ype ............... ............................................. .. 57 H6b: Placement of R isk C omparison to K now n R isks ............ ...... .................. .. 58 H7: Information A bout R isk C omparison in O ther C ountries...................... ..... 58 5 DISCUSSION................ .......... ... ....................................................................... .. 67 Major Findings and Implications ............................................................................... 67 Dread E voking I nformation .............................................................................. .. 67 Uncertainty................................................................................... .................... .. 68 Controllability ................................................................................................. .. 69 Familiarity ....................................... ................................................................ ... 71 Summary ....................................................................................................... .. 72 Limitations..... ....................................................................................................... .... 72 Future R esearch R ecommendations.................... ... ............................................. .... 73 APPENDIX A CODING SHEET............. ............ ................................................................ .. 74 B CODING G UIDELINES...................... ....................................................... .. 7 8 C SCO .. 82 D LOADED WORDS ................................. ..................................................... .. 83 E UNCERTAIN WORD S .................. ............................................................ ... 84 REFERENCE LIST .............................. ......................... .......................................... .. 85 BIOGRAPHICAL SKETC H 9 8
7 LIST OF TABLES Table page 1 1 Known s uperbugs and h ealth p roblems.............................................. ......... 13 2 1 Health related d ata in India, United Kingdom and the United States....... ... 43 2 2 Eighteen r isk c haracteristics ...... .. ................ .............................................. .. 44 4 1 Numbers of newspaper articles on ND M 1 in each country ......... ............ ... 59 4 2 Frequencies of w orst c ase s cenarios in n ews c overage ........................... ... 59 4 3 Placement of w orst c ase s cenarios in n ews c overage ...............................59 4 4 Frequencies of l oaded words in n ews c overage ..........................................60 4 5 Placement of l o aded w ords in n ews c overage ............................................60 4 6 Most f requently u sed l oaded w ords in n ews c overage .............................61 4 7 Frequencies of u ncertain w ords in n ews c overage .....................................62 4 8 Placement of u ncertain w ords in n ews c overage ........................................62 4 9 Most f requently u sed u ncertain w ords in n ews c overage ... .......................62 4 10 Risk m agnitude of h uman i nfection/ d eath ..................................................63 4 11 Risk m agnitude of f inancial l oss .......................................... .......................63 4 12 Frequencies of p ersonal p rotection i nformation in n ews c overage .............63 4 13 Personal p rotection i nformation in n ews c 4 14 Frequencies of s ocietal p rotection i nformation in n ews c overage ............... 64 4 15 Comparison to o ther s uperbugs Presence ............................................. 64 4 16 Comparison to o t her s uperbugs Information t ype .................................. 65 4 17 Placement of r isk c omparison to o ther s uperbugs in n ews c overage ...... 65 4 18 Risk c omparison to o ther c ountries ............. ................................................ 66
8 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Advertising TALKING ABOUT PUBLIC HEALTH RISK: NE WS COVERAGE OF THE NDM 1 IN INDIA, UNITED KINGDOM, AND THE UNIT E D STATES By Bijie Bie August 2012 Chair: Debbie Treise Major: Advertising This thesis examined how Indian, UK and U S newspapers conveyed health risk information in coverage of the NDM 1 superbug. A quantitative content analysis of 266 news articles was conducted. Using the psychometric paradigm as a theoretical framework, the following dimensions of risk characteristics were examin ed in this study: dread evoking information, uncertainty, perceived controllability, and familiarity. Some clear differences emerged in the way NDM 1 related risks were covered and possible reasons a re discussed. The results showed that social factors may affect the way media cover health related issues. Moreove r, the study indicated the complexity of global health communication. Based on the findings the study suggested that decision makers should consider a s more social factors as possible to ensure the public receives the correct and transparent information.
9 CHAPTER 1 INTRODUCTION The Goal of the Study The objective of the current study is to evaluate NDM 1 superbug coverage in newspapers from India, the United Kingdom and the United States, from the perspective of public health and risk communicatio n. These countries were chosen due to the different development stages of NDM 1 in each country. This superbug was initiall y discovered in India, and has received a lot of attention in India and the United Kingdom, and now it is emerging in the United States. The s tudy builds upon existing research in health communication, risk communication, social and psychological studies, and the methods used in this study are based on an earlier study conducted by Fung, Namkoong, and Brossard (2011). The study takes into consid eration the social context of these three countries, including the stage of development of NDM 1 in each country health care system, and media system. An important model in the field of risk analysis, the psychometric paradigm, w as used as a theoretical framework in the study. Background information about NDM 1 Antimicrobial resistance has been defined as "resistance of a microorganism to an antimicrobial medicine to which it was previously sensitive" ("Combat Antimicrobial Resis tance ," n.d., para. 2). In other words, infections caused by resistant microorganisms can make standard treatments ineffective so that individuals suffering f rom many forms of antimicrobial resistant organizisms will experience "prolonged illness and grea ter risk of death" ("Combat Antimicrobial Resistance ," n.d., para. 1). Antimicrobial resistance is now perceiv ed as a serious global threat, is "reaching unprecedented levels" (World Health Organization [WHO], 2011d, para. 2). It
10 s control over infectious diseases and is causing tremendous health and economic consequences. For example, in the European Union alone, more than 25,000 people die each year from antibiotic resistant bacterial infections (WHO, 2011d), and the annual economic loss is estimated to reach at least 1.5 billion Euros (WHO, 2011d). Antibiotic resistant bacteria are commonly referred to as "superbugs." The past several decades have witnessed an increase in the number of cases of different drug resistant superbugs: methici llin resistant Staphylococcus aureus (MRSA), Enterococcus faecium, Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and enterobacter species, and others (Moellering, 2010). The drug resistant superbugs and the kinds of health problem s they cause are shown in Table 1 1. The NDM 1 superbug, an acronym for New Delhi metallo lactamase 1, is a pan resistant enzyme that makes bacteria resistant to a broad range of antibiotics (Moellering, 2010). It was first identified in New Delhi, India in 2008 (Moellering, 2010), and since early 2010, resistant bacteria containing the NDM 1 enzyme have begun to strike other areas of India, Pakistan and the United Kingdom (WHO, 2010a). In August 2010, the spread of the resistant bacteria in India, Pakis tan and the United Kingdom was described in a report published in a British journal, The Lancet Infectious Diseases (WHO, 2011e). In that report, the new superbug was described as resistant to all currently available antibiotics (Golikeri, 2011). In other words, the NDM In 2010, the spread of bacteria carrying NDM 1 quickly became a global issue: a number of other coun tries, such as the United States, Canada, Australia, Belgium,
11 Japan, Sweden and Viet Nam all reported cases in 2010. For example, by early September 2010, three NDM 1 cases had been identified in three U.S. states: California, Massachusetts and Illinois ( WHO, 2010a). Consequently, NDM 1 has become a potentially serious public health issue and global attention. In a 2010 report, Professor John Conly from the University of Calgar y said he believed the spread of NDM1 containing ] a Day 2011, the World Health Organization launched a worldwide campaign to call on gove rnments to combat the global threat posed by antimicrobial resistance (WHO, 2011b). Treatment of NDM 1 According to the World Health Organization (2010a), NDM 1 is extremely hard to treat for many reasons. First, NDM 1 exists in a number of different types of bacteria. Most of these NDM 1 containing strains are resistant to the most powerful antibiotics, and at least 10% of these strains are incurable so far. Second, very little progress has been made in the development of new drugs for antimicrobials. So w hen a person is infected with bacteria that harbor this new resistance mechanism, there are few antibiotics to treat them. Moreover, the infection mechanism of NDM 1 is controlled by a set of genes that are transmissible and can move easily among bacterium Furthermore, E. coli NDM 1 has been found in E.coli bacteria, which is the most common cause of bladder and kidney infections. According to the Centers for Disease Control and Prevention (2012), some certain species of E. coli can cause diarrhea, while o thers can cause "urinary tract infections, respiratory illness and pneumonia, and other illnesses" (para. 1). The most recent outbreak of E. Coli was in
12 Europe, 2011 (WHO, n.d.). According to statistics available, there were 3,999 cases reported in Germany and 122 in 16 other countries during the 2011 European E. coli outbreak, and a total of 50 people died ("E.coli death toll increases to 50", 2011, July 1). Finally, although there are two antibiotics that may be effective against NDM 1, one of them, at th e highest dosage, has the potential to cause kidney damage in one third of people taking the drug To sum up, it is difficult to treat patients in fected with bacteria with the NDM 1 enzyme. The current study is designed to investigate Indian, UK and US new s coverage related to risks posed by the NDM 1 superbug to determine if there are significant differences in coverage in the three countries Significance of the Study This study is significant for the following reasons. First, because NDM 1 is becoming a highly important topic across many areas of the world this paper attempts to investigate what specific risk related information is covered when journalists are reporting about NDM 1. To date, no empirical study has been done regarding coverage of the NDM 1. T imely and continuously updated information is crucial for infection control, especially when people have only limited knowledge about a new risk (Ahmad, Krumkamp, & Reintjes, 2009). This s tudy may provide some insight for future research on health a nd risk communication. The theoretical framework of the study is the psychometric paradigm, an important and useful approach in research on public perceptions of risk (Bronfman, Cifuentes, & Willis, 2007; Renn & Rohrmann, 2000; Slovic, 1987). However, t here is limited literature using the psychometric paradigm to examine health coverage in the news media (for recent exceptions, see Fung et al., 2011). Thus, more research is needed in this area, as it can add to existing knowledge of health communication and
13 r isk communication by expanding the application of psychometric theory to the domain of global health communication, where a very limited number of studies have been conducted so far.
14 Table 1 1. Known superbugs and health problems Medical name of superb ugs Health problems Methicillin resistant Staphylococcus aureus (MRSA) Can affect various body sites and staph infections can be found in "the bloodstream, heart, lungs, blood, bone, urine, or at the site of a recent surgery" (PubMed Health, 2011, para. 9) Enterococcus species Can cause infections "anywhere in the body" ( WebMD, 2011, para. 2 ) Klebsiella pneumoniae Can cause "different types of healthcare associated infections, including pneumonia, bloodstream infections, wound or surgical site infection s, and meningitis" (Centers for Disease Control and Prevention [CDC], 2010, para. 1) Healthcare associated infections refer to infections that people acquire while they are receiving treatment for another condition in a healthcare setting" ( Department of Health and Human Services, n.d. para. 1 ). Acinetobacter baumannii Can cause bloodstream infections (CDC, 2004) and "various types of human infections, including pneumonia, wound infections, urinary tract infections, bacteremia, and meningitis" ( Choi, Le e, Lee, Park, & Lee, 2008, para. 1 ) Bacteremia is "the presence of viable bacteria in the circulating blood" (Bennett, 2012, para. 1). Pseudomonas aeruginosa Can cause pneumonia, extensive tissue damage or even septic shock (BBC News, 2012) Septic shoc an overwhelming infection leads to life Medical Encyclopedia, 2010, para. 1). Enterobacter species Can cause many different types of infections, including "bacteremia, lo wer respiratory tract infections, skin and soft tissue infections, urinary tract infections (UTIs), endocarditis [ an infection of the inner layer of the heart ] intra abdominal infections [ infections in the cavity of the abdomen ] septic arthritis [also kn own as infectious arthritis ] osteomyelitis [ bone infection ] and ophthalmic infections [eye infections] (Fraser, 2010, para. 1 )
1 5 CHAPTER 2 LITERATURE REVIEW Media C overage of Health R isks Mass media play a vital role in influencing public perception o f reality (Scheufele 1999). In particular, the media serve as a major source of health information for the general public (Bomlitz & Brezis, 2008) and play a pivotal role in communicating health related risk information to the public, especially to non exp ert audiences (Courtney, 2004; Sjberg, Kolarova, Rucai, & Bernstrm, 2000). Health related media content has been discussed in past research in sociology (Courtney, 2004; Wenham, Harris, & Sebar, 2009), psychology (Lyons, 2000; Renn & Rohrmann, 2000), sci ence communication (DeSilva, Muskavitch, & Roche, 2004), and health communication (Bomlitz & Brezis, 2008; Fung et al., 2011; Y. Liu, M. Liu, Xiao, Cai, & Xu, 2010; Stryker, 2003; Yanovitzky & Blitz, 2000). These scholarly studies have investigated how new s media convey health information (Bomlitz & Brezis, 2008; DeSilva, et al., 2004; Fung et al., 2011; Y. Liu et al., 2010; Wenham et al., 2009), how media content can shape individuals' perceptions of health risk s (Courtney, 2004; Lyons, 2000) and how the m & Blitz, 2000). Social C ontext in India, the United Kingdom and the United States In the current study, news articles collected from three countries (India, the UK and the U.S.) w ill be analyzed for cross national comparisons. The three countries not only represent vast differences in social proximity, health care systems, media systems, and people's health literacy, but also represent different development stages of NDM 1. Society of India, the United Kingdom and the U.S. Research has shown health communication is related to many social factors
16 including economic factors (Bloom & Canning, 2008; Parker, Woelfel, Hart, & Brown, 2009 ), cultural factors ( Kreuter & McClure, 2004) and r eligious factors (Ellison & Levin, 1998). Comparisons of above mentioned factors in India, the United Kingdom and the United States are presented below. Economy. The United States has the largest and most technologically powerful economy in the world (CIA, 2011c). The United Kingdom, another developed country, is the world's eighth largest economy by purchasing power parity and the third largest economy in Europe (CIA, 2011b). India is the only developing country among the three; however, as one of the majo r emerging economies in the world, it has become the world's fifth largest economy by purchasing power parity (CIA, 2011a) with the world's fifth fastest GDP real growth rate. In a word, each of these three countries plays an important and influential role in the international arena. However, as a developing country, poverty remains a major challenge in India, which is located in South Asia. It is second only to China in terms of population. According to the Central Intelligence Agency, the poverty rate (p ercentage of population belo w the poverty line) is 14% in the United Kingdom 15.1% in the U nited States but as high as 25% in India (CIA, 2011a, b, c). Culture Previous research also linked cultural factors to the quality of health promotion and health communication ( Kreuter & McClure, 2004). According to Hofstede's national culture index, the dimensions of power distance, individualism, and long term orientation vary widely among India, the United Kingd om, and the United States (see Geert H ofstede.com, n.d.) ; this is especially true for individualism and long term orientation. Americans and British people score high (about 90) on the dimension of individualism, which means the American and British culture tend to put great emphasis on personal achievemen t, personal opinions, personal needs and
17 privacy; however, Indians only score 48, which means people in Indian society tend to develop close ties with other people and have strong loyalty towards the group and the relationship. India scores hig h (61) on th e dimension of long term orientation, while western societies including the United Kingdom (scores 25) and the United States (scores 29), usually score low on this dimension. These scores reflect the fact that India has a more pragmatic future oriented cu lture, "rather than a conventional historical short term point of view" ( Geert Hofstede.com India, n.d., para. 13). Cultural factors have a great impact on population health and wellbeing. Researchers argued that individualism is disadvantageous to perso nal health care ( Eckersley 2006; Shiell & Hawe 1996) because many diseases (including heart diseases) are associated with lack of control over one's life, and lack of social support, but also Religion. Religious beliefs in these three countries also differ considerably. Religion in India has been dominated by Hinduism for centuries, while the dominant religion in the United Kingd om and the United States is Christianity. More than 80% of Indians identify themselves as Hindus, and only 2.3% of Indians were Christians in 2001 (Office of the Registrar General & Census Commissioner, 2001). In contrast, 71.6% of UK adults ( United Kingdo m Census 2001 Religion 2001) and 78.4% of U.S. adults ( U.S. Religious Landscape Survey 2008) were Christians. Hofstede's research into national and organizational culture suggest s that Hinduism has had considerable influence on the development of Indi an culture. According to the Geert Hofstede's website, which uses copyrighted information from on national cultures: Countries like India have a great tolerance for religious views from all over the
18 world Hinduism is often considered a philosophy more than even a religion; an amalgamation of ideas, views, practices and esoteric beliefs. ( Geert Hofstede.com India, n.d., para. 14). In some sense, the differences between Eastern and Western thought are reflected in the diver gence of Hinduism and Christianity. For example, Christians believe that the nature of man is sinful, while Hindus believe that man's problem is ignorance; Christian doctrine stresses "belief," while Hinduism stresses "duty"; Christians believe that Jesus Christ can save people from sin, while Hindus believe that nobody can have the assurance of salvation ( Comparison of Christian and Hindu Beliefs n.d.; Kreeft, 1987). Christianity believes that there is a strong link between suffering and evil, but that pe ople's suffering can be transformed into something beneficial with the help of God; however, central beliefs of Hinduism include a belief in Karma, the law of cause and effect, which implies that any form of suffering is not a punishment but a consequence of negative behavior in past life (Whitman, 2007). The religion health connection is an emerging and fast growing research area (Cadge, Ecklund, & Short, 2009; Ellison & Levin, 1998; Miller & Thoresen, 2003). Although some researchers argue d that the scie ntific evidence of the relationship between religion and medical views is "weak and inconsistent" (Sloan, Bagiella & Powell, 1999, p.667), recent research suggested that religious beliefs may have complex effects on health beliefs or health outcomes (Koeni g, 2007). The Health Care Providers' Handbook on Hindu Patients (The Queensland Government, 2011) mentioned that Hindu beliefs might affect health care in many respects. For example, most Hindus don't eat beef and some Hindus are entirely vegetarian; there fore, medications made using animals are not suitable for vegetarian Hindus. Furthermore, as mentioned before, Karma implies that people's thoughts and deeds in past lives
19 will affect their current life, including their health, and the best way to cope wit h suffering is to accept it and understand that the suffering is natural; therefore, Hindu patients' decision making regarding health care may be affected by the belief of Karma. Health C are S ystems in India, the United Kingdom and the United States Based on reports from the World Health Organization and statistics from the Central Intelligence Agency (2011a, b & c), it is clear that the healthcare system s and government investment in healthcare in these three countries are quite different. As shown in Tab le 2 1, there are observable differences between India and the other two developed countries in terms of health expenditure and health outcomes. For instance, regarding government expenditure on health as a percentage of total government expenditure, India spends 4.4% of its government budget on health, wh ile United Kingdom spends 15.1% and the United States spends 18.7% ( World Health Statistics 2011, 2011, see Table 2 1). United Kingdom. Health services in the United Kingdom are mainly financed by the gov ernment, although there is a growing private health care industry (Roe & Liberman, 2007). The National Health Service (NHS) provides preventive medicine, primary care and hospital services to all legal residents and citizens of the United Kingdom (Boyle, 2 011, p.21). The NHS is "one system, one organization" that "guarantees the right to health care access to all citizens" (Roe & Liberman, 2007, p. 193). According to World Health Statistics 2011 82.6% of the total expenditure on health in the United Kingdo m was paid by the government, while the percentage was 47.8% for the United States and 32.4% for India. Apart from public coverage, about 13% of the British population is covered by voluntary private health insurance (Boyle, 2011).
20 United States. Cutler ( 2008) states that although the United States has "the most technologically advanced medical system citizens do not automatically have medical coverage" (para. 1). Private business plays a significant role in the American health insurance sys tem, and a lot of people are uninsured across the nation. According to the Census Bureau, the number of uninsured persons in the United States was 49.9 million in 2010, constituting about 16.3% of the population (Christie, 2011). However, in 2010, The Heal th Care and Education Reconciliation Act of 2010 was signed into law. This act "maintains the private insurance system but makes it work for everybody" (Phillips, 2010, para. 2). New health insurance provisions in this act can possibly increase the availab ility of tax credits to help people cover the cost of insurance, and can possibly ensure that everyone has access to the health care when they need it (Phillips, 2010). Therefore, it is likely the condition of health insurance coverage in America will chan ge for the foreseeable future. India. The health care system in India is a "system of consumer paid health care" (Mehr, 2008, p.8). Although India spent 4.6% of its GDP on health, their government share of that expenditure is quite low, accounting for only 0.9% of its G DP (WHO, 2006b); meanwhile, out of pocket expenditure is high, although India has a large impoverished population (WHO, 2006a). In terms of health insurance, according to estimates, "only 3% to 5% of Indians are covered under any form of he alth insurance" (Rao, 2005: p.5). Quality control in public hospitals is almost absent in India; Mehr (2008) asserted that "medical care is largely unmeasured in public hospitals and physician offices" (p.6). Therefore, India's private hospital sector, whi ch provides high quality and high standard medical services, has become the most promising for "improving the
21 healthcare access and quality of care in India" (Mehr, 2008, p.6). However, those private practitioners cost much more than government hospitals a nd can only serve particular groups of people, including international medical tourists. Most of the medical tourists are from Southeast Asia, North America and Europe (Mehr, 2008) Health inequal ity between urban and rural areas is another big problem in India (Mehr, 2008). Health care in India's rural regions is poor. Many regions face a critical shortage of health care providers; many regions lack funds and access to primary care. Because of the poor sanitation and water supply difficulties in its rural area, India faces a great many of difficulties in dealing with diseases (Mehr, 2008). Use of antibiotics in India, the United Kingdom and the United States The rational use of antibiotics is key of infection prevention and control. As "infectious diseases remain a major cause of illness, disability, and death" (U.S. Department of Health and Human Services, n.d., para. 2), the control of infectious diseases relates closely to people's health status and is one of the major components of a country's health sy stem (U.S. Department of Health and Human Services, n.d.). However, inappropriate use of antibiotics can lead to severe consequences. According to the World Health Or ntimicrobial resistance is the inevitable consequence o found that inappropriate use of antimicrobial medicines creates an ideal environment 2011a). However, some developing countries such as India have limited regulation of the use of antibiotics in their healthcare system, inevitably leading to irrational use of antibiot ics in hospitals (WHO, 2010c). Moreover, a 2011 WHO study revealed that 53% of Indians take antibiotics without a doctor's p rescription (WHO Regional Office for South East Asia, 2011). A standardized measure of antibiotic consumption is DDD
22 (defined daily doses), which is recommended by the WHO Collaborating Centre for Drug Statistics Methodology (2009) Basically, the DDD is d efined as "the assumed average maintenance dose per day for a drug used for its main indication in adults" (WHO Collaborating Centre for Drug Statistics Methodology, 2009, para. 2). A popular DDD index is DDDs per 1000 inhabitants per day, which can "provi de a rough estimate of the proportion of the study population treated daily with a particular drug or group of drugs" ( The concept of the defined daily dose, para. 8). For example, when the antibiotic consumption in a certain population i s 10 DDDs per 1000 inhabitants per day, this indicates that 1% of the population on average might receive antibiotics ( The concept of the defined daily dose ). In terms of DDDs/1000 inhabitants/day, the antibiotic receiving rate for India was 39% 43% (Kotwani & Holloway, 2 011), while the corresponding rate was 24% for the United States and 15% for the United Kingdom (The Center for Disease Dynamics, Economics & Policy, 2007: p.8). Moreover, it is reported that in India, the antibiotics sector constitutes the largest market segment of the pharmaceutical industry with a share of 15.7% ( MIHR report to CIPIH, 2005). In summary it can be seen that the use of antibiotics in India is heavy and unregulated compared with use in the United Kingdom and the United States. Media system s in India, the United Kingdom and the United States Historically, the Western media system s, including those in Western Europe and North America, have been based on market oriented private ownership. In India, although there were some state owned media ag encies in the past, its private media sector has grown rapidly o ver the past decade (Business Knowledge Resource Online, 2012). In 2001, all of the top five daily newspapers and three of the top five television stations in India were private owned (Djankov McLiesh, Nenova, & Shleifer, 2001). In 2010, all of the top five Indian English language daily newspapers, including
23 The Times Of India Hindustan Times The Hindu The Telegraph and Deccan Chronicle ( Manocha 2010) were owned or controlled by large private companies such as Bennett & Coleman & Co. Ltd. On the other hand, India still ranked poorly on press freedom. According to a press freedom rankings release d by the organization Reporters Without Borders (2012), the United Kingdom was ranked 28th out of 179 countries, the United States was ranked 47th, while India was ranked 131st. Prevalence of NDM 1 in India, the United Kingdom and the United States The ND M 1 was initially identified in India. By September 2011, there were 143 NDM 1 cases in India, 88 cases in the United Kingdom (Health Protection Agency, 2011), and three cases in the United States. Moreover, a research paper publish ed in The Lancet Infecti ous Diseases reported that bacterial strains with the NDM 1 w ere found in the public water supply in New Delhi, India (Walsh, Weeks, Livermore, & Toleman). The water sources that tested positive were used for drinking, food preparation, bathing and persona l hygiene and laundry indicating bacteria carrying the NDM 1 ha ve been "widely disseminated in New Delhi and has spread into key enteric pathogens" (Walsh et al., p.360). Thus, the NDM 1 can be viewed as prevalent in India and the United Kingdom, and it is emerging in the United States. Shih, Wijaya and Brossard (2008) investigated media coverage of three types of public health epidemics, including mad cow disease, West Nile virus, and avian flu, and found that different development stag es of a specific risk influence media coverage of health related risk issues. Thus, it seems reasonable to deduce that news reporting of NDM 1 in India, the United Kingdom, and the United States would be influenced by the prevalence of NDM 1 in each countr y.
24 Health L iteracy Definition of L iteracy Because general literacy is the necessary prerequisite to processing and understanding health information, it is necessary to understand what literacy constitutes and how it may vary from one country to the next As the most comprehensive measure of adult literacy in the United States since 1992, the 2003 National Assessment of Adult Literacy (NAAL) assessed seven types of literacy skills: (1) basic skills of recognizing and understanding words; (2) basic skills o f understanding sentences; (3) text search capability; (4) ability to identify necessary calculations in text; (5) ability to calculate mentally or with the help of calculators to solve quantitative problems; (6) ability to make reasonable inferences while reading; (7) capability to use information obtained in above processes for decision making. Previous research found that l evel of education is closely associated with the usage pattern of health information in newspapers ( Hay, Coups, Ford, & DiBonaventura 2009 ). Compared to people with lower education levels, those with higher education levels are more likely to be exposed to health information in newspapers ( Hay, Coups, Ford, & DiBonaventura, 2009 ) and trust health information in newspapers ( Hesse, et al ., 2005 ). Definition of Health L iteracy In the report Healthy People 2010 health literacy was defined by the U S Department of Health and Human Services as "the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions" (2000, p. 11 20). In a broader definition of health literacy, the World Health Organization (1998) not only stressed citizens' personal ability to deal with health information, but also emphasized citizens' intrinsic
25 motivation for better health statu s : "Health literacy represents the cognitive and social skills which determine the motivation and ability of individuals to gain access to, understand, and use information in ways that promo te and maintain good health" (p. 10). Apart from that, the American Medical Association (1999) has defined health literacy in a more practical and functional way: "[Health literacy is] the ability to read and comprehend prescription bottles, appointment sl ips, and the other essential health related materials required to successfully function as a patient" (p. 552). Health Outcomes and Behaviors Associated with Health Literacy Research in the past years has documented that health literacy not only affects pe also affects people's health outcomes and behaviors (Parker, Ratzan, & Lurie, 2003; Speros, 2005). It is a strong predictor of health status that "empowers people to act appropriately in new and changing health related circumstances" (Speros, 2005, 33). The benefits of health literacy improvement include better health status, reduc ed health care costs, enhanced health knowledge and a more efficient health care system (Sper os, 2005). Therefore, improving health literacy has become a national health goal in the United States (Department of Health and Human Services, 2000). There are many negative consequences of inadequate health literacy. For instance, inadequate health lit eracy is associated with inappropriate treatments (Persell, Osborn, Richard, Skripkauskas &Wolf, 2007), relatively higher mortality (Sudore, Yaffe, Satterfield et al., 2006) and other undesirable health outcomes. At present, researchers believe inadequate health literacy is still common, affects people from all facets of life (Speros, 2005) and is a common challenge faced by health organizations all over the world (Stableford & Mettger, 2007).
26 Health L iteracy and M edia L iteracy H ealth news coverage in po pular media has been increased during the past decades (Gilk, 2003). I nterest in media literacy as a facet of health education has been growing because media literacy can have an impact on how people consume and utilize health content in the news media (Gi lk, 2003). Also, based on a national telephone survey, a report ("The Social Life of Health Information, 2011") released recently by the Pew Research Center indicated that people's media literacy and health literacy are "inextricably linked" (Long, 2011, p ara. 4). Therefore, it is necessary to take media literacy into consideration when discussing health literacy. Definition of M edia L iteracy Media literacy has become increasingly important to society (Livingstone, 2004), and has become a fast growing rese arch area during the past decades (Potter, 2010). To date, there is little consensus over the standard definition of media literacy; a number of scholars and organizations have made their own attempts and proposed various definitio ns (Brown, 1998; Potter, 2010). Briefly, media literacy refers to the ability or skill to use media information. For example, according to the National Communication Association, "A media literate person understands how words, images, and sounds influence the way meanings are crea ted and shared in contemporary society in ways that are both subtle and profound" (as cited in Potter, 2010, p.677). From a skills based approach, the skills of access, analysis, evaluation and content creation are the most recognized components of media l iteracy (Livingstone, 2003). For example, Schwarz (2005) defined media literacy as "the ability to access, analyze, evaluate, and create media in a variety of forms" (p. 11). Many scholars also have stressed the ability to think critically when defining me dia literacy (Adams & Hamm, 2001; Brown, 1998;
27 Siverblatt & Eliceiri, 1997). For instance, to Siverblatt and Eliceiri (1997), media literacy is "a critical thinking skill that enables audiences to decipher the information they receive through the channels of mass communications and empowers them to develop independent judgments about media content" (p. 48). In sum, although there are diverse definitions of media literacy, all definitions emphasize understanding "the selection, interpretation, and impact" of media messages (Rubin, 1998: p.3). Health L iteracy in India, the U nited Kingdom and the U nited States There is widespread agreement, including by researchers and healthcare practitioners, that health literacy is an important factor influencing the prevent ion and control of emerging diseases ( Gargano Schreier & Hughes 2010; Pappas, et al., 2007). Therefore, it seems reasonable to deduce that health literacy would be an important factor that affects people's perception and behaviors towards NDM 1. India. According to the most recent data from the 2011 census for India, the absolute literacy rate in India was only 74%, which means 26% of the Indian population was illiterate. Here, the criteria for "literate" is the ability to read and write with understandi ng (Park, 2009). No research has examined the health literacy status in India but given that general literacy is a required precursor to using health information effectively, it is clear that at least 26% of the Indian population may lack health literacy skills. The United States According to the U.S. Department of Health and Human Service (2008), more than one third of U.S. adults have difficulty with common health tasks, and only 12% of U.S. adults had proficient health literacy skills. The United King dom In comparison with India and the U.S., health literacy condition in the UK is relatively better. The 2011 Skills for Life Survey (SLS) showed that 14.9% of British adults were at or below the entry level for health literacy (UK
28 Department for Business Innovation and Skills, 2011). According to the UK Department of Health (2009), the 2011 Skills for Life Survey (SLS) indicated that less than 60 % of British adults had the full potential to deal with health information and perform complex tasks. Media literacy education in the United Kingdom the United States and India As noted by Livingstone and Thumim (2003), "there is little consensus over the appropriate means to measure media literacy" (p.1). Although scholars and educators have not yet agreed on how to measure media literacy, we can still draw useful lessons from the development of media literacy education in different countries. In the following paragraphs, the state of media literacy education in India, the United Kingdom and the United States w ill be reviewed and discussed. Generally speaking, media literacy education is developing rapidly in the United Kingdom and the United States, while it has only just begun to grow in India. The United Kingdom The United Kingdom has been widely acknowledg ed as a lea der in media literacy education because media literacy education has long been established in the national school curriculum (Rubin, 1998). UK media education not only emphasizes understanding the content and social influence of media messages, but also pays attention to how to engage citizens in the creation and production of media content (Brown, 1998). In 2002, the British government proposed a communications bill (draft) to 'promote media literacy' among the UK population. This was the first time that the United Kingdom had addressed this issue at the policy level (Livingstone & Thumim, 2003). Later in 2003, for the first time in UK legal history, the Communication Act of 2003 gave the duty to promote media literacy to the Office of Communicat ions (Ofcom). After that, media literacy level s in the United Kingdom generally have improved. For example, according to the recent UK Adults's Media
29 Literacy R eport (Ofcom, 2011), compared to the past, UK adults are more likely to understand that media co ntent is regulated across media platforms. The U nited States Media literacy is emerging in the United States (MacDonald, 2008). U.S. scholars and media literacy educators generally agree that citizens should be equipped with adequate skills to utilize and benefit from media. But unlike the media education in United Kingdom media literacy projects in the United States tend to pay more attention to the negative nature of mass media (Brown, 1998). In this way, media audiences are trained to be aware of the n egative influences of advertising and media entertainment. India. While media literacy education is fast growing in dev eloped countries such as the United Kingdom, the United States and Canada, it is just beginning to gain interest from educators and scho lars in developing countries such as India (Arul & Suresh, 2009). Although the Indian media industry is rapidly developing, media literacy in India is making very slow progress (Arul & Suresh, 2009). On the one hand, Indian media education "enjoys a low pr iority in educational institutions and communicatio n centers" (Arul & Suresh, 2009, p.121). For example, media literacy programs or media education programs are provided by a number of British universities (e.g., University of East Anglia, University of Lo ndon) and U.S. universities (e.g., the University of Texas Austin, Temple University); but in India, there are no similar college programs at all and research on media literacy education is very hard to find (Arul & Suresh, 2009). In addition "insufficien t training and lack of teaching materials available to media educators" (Arul & Suresh, 2009, p.127) make it very difficult for educators to keep up with the development of information and technology M edia literacy and health literacy are important for he alth communication
30 research. B oth media literacy and health literacy levels h ave an impact on people's attitudes toward health related risk information and how the health related risk information is delivered. Thus, in this study, media literacy and health literacy levels in India, United Kingdom and the United States are considered as important factors influencing the media messages about the NDM 1. Training of Health Journalists in Three Countries Today's health news industry environment is ever changing and challenging; thus, vocational training is crucially important for health journalists (Schwitzer, 2009). The primary objec knowledge and upgrade skills, to raise the quality of health journalism and to ensure t hat health messages are "delivered effectively by seasoned reporters who perform thoughtfully even in the face of breaking news and tight deadlines" (Dentzer, 2009, p. 1). Opportunities for training in health reporting and the extent of specialized trainin g among health journalists varies significantly among the three countries in this study. United Kingdom Organizations. In the United Kingdom, there are some professional organizations for health reporters, such as the Medical Journalists' Association (Lond on, UK) and the Guild of Health Writers (Hampton, UK). The Medical Journalists' Association (MJA) was founded in 1967 and has established itself to "support and encourage its members and enable them to work efficiently and at high levels of accuracy" ("Med ical journalism at its best", n.d.). Its membership consists of "over 400 of the UK's leading medical and health journalists" ("Medical journalism at its best", n.d.), as well as "doctors, nurses, therapists and academics who contribute to national and pro vincial newspapers, professional journals, business and consumer magazines, radio, television and the web, and/or who write books" ("About Us", n.d.,
31 para. 2). The MJA organize s and manage s various events for those who create health content in the media, such as the Are you at sea with statistics? (December, 2011), "Protecting patients, checking doctors: Have we got the balance right?" (October, 2011) and "Coventry health journalism" conference (June 2011). Health journalists in the United Kingdom also have the opportunity to benefit from some other European organizations and projects, such as the Association of Health Care Journalists (AHCJ), an offshoot of the American AHCJ organization that was launched for all European health journalists and the Hea RT (Health Reporter Training) project. The HeaRT project, which was launched in November 2010, was designed to "develop vocational training on health reporting for journalists" ("Health Reporter Training Summer 2012", 2012, p. 2). The HeaRT 2012 program is designed for "health journalists, working journalists, journalism students, editors and publishers" ("Health Reporter Training Summer 2012", 2012, p. 1). This 2012 program has seven partner countries in Europe, including Estonia, Finland, Germany, Greece, Romania, Spain, and the United Kingdom. In the United Kingdom, training sessions are available at Coventry University. Seventy seven seminars and courses were designed to provide participants with not only specialist knowledge, but also useful skills such as interpreting medical research reports and understanding statistics ("Health Reporter Training Summer 2012", March 11, 2012). University Education. In the United Kingdom, several universities offer health journalism programs designed to provide appropr iate and thorough training for future health reporters. For exa mple, established in 2009, the m aster j ournalism at City University London provides "a thorough grounding in the best practices in health, science and environmental journal ism, whilst teaching you to be a critical consumer of scientific information" ("Science Journalism: Overview", n.d., para.
32 1). The graduates have been working for newspapers, magazines, broadcast media and the World Health Organization ("Science Journalism: Career Prospects", n.d.). In addition, Cov entry University also offers a health j ournalism MA program. United States Organizations. In the United S tates, professional organizations also play a vital role in the training of health journalists. T he most well known organization is the Association of Health Care Journalists (AHJC). Founded in 1998 and now located at the University of Missouri Columbia School of Journalism, the AHCJ has more than 1,000 members from 15 nations ("History of AHCJ", n.d. ) and provides various types of training to its members, including conferences, workshops and seminars, as well as online interactive courses. In his research report, The State of Health Journalism in the U.S. Schwitzer (2009) described AHCJ as "one of th e most positive forces in the [Health Journalism] industry" (p. 5): Its annual (and more recently introduced regional) conferences have become popular and successful training opportunities for hundreds of journalists. The group has also begun offering peri odic workshops on urban and rural health topics. AHCJ has published comprehensive guides on covering the quality of health care, on covering hospitals, on multicultural health issues, and on covering obesity. Its online tip sheets address dozens of health care topics. (p. 5) University Education. The first health journalism program in the United States was launched at the University of Minnesota in 2003 (Johnson, 2006). This masters of arts program in health journalism, which combines knowledge and training in journalism and public health, was designed to bring together students from two disciplines and teach them how to communicate public health issues accurately and effectively (Johnson, 2006). At present, this program has two different training directions : health journalism and health communication. According to the official
33 website of the University of Minnesota, the health journalism emphasis prepares students to "gain advanced knowledge about public health and the evaluation of claims from health, medic al, and scientific sources, as well as advanced training on reporting health stories for different media" (Program curriculum section, n.d., para. 1). Also, the University of California Berkeley launched a three year M.P.H./M.J. (Master of Public Health an d Master of Journalism) concurrent degree program, which "allows students to combine their interests in public health, journalism, communications and media" (University of California Berkeley, n.d., para. 1). The program prepares students to be future publ ic health professionals who are also successful media practitioners, communicators, and journalists. But still there is much room for improvement. Schwitzer (2009) analyzed the data from a 2008 survey of members of the AHCJ, which was conducted jointly by AHCJ and the Kaiser Family Foundation (KFF). The survey showed that although 20% of respondents (staff journalists) said their training opportunities had increased in recent years, 43% of respondents reported a reduction of training opportunities (Schwitz er, 2009). This point is consistent with Dentzer's (2009) argument that many health journalists thought they didn't receive enough training in medical knowledge. Schwitzer's (2009) research suggested that more training opportunities should be created for y oung health journalists and editors. India Training provided by NGOs. People have long failed to pay proper attention to the significance of medical journalism, especially in developing countries like India ( 2011). Compare with the U nited States and United Kingdom, there seem to be very few training programs for health journalists in India. However, in summer 2006, an India based non profit organization, Communication for
34 Development and Learning, organized a training course for exper ienced Indian health journalists (International Journalists' network, 2006). Educational institutions. Several educational institutions in India offer medical journalism education to students. For example, the James Lind Institute offers two types of medic al journalism programs: a Professional Diploma in Medical Journalism and an Advanced Postgraduate Diploma in Scientific Writing and Medical Journalism, both of which are short term programs. The Professional Diploma in Medical Journalism program, which tak es only 4 6 months, prepares students with a "strong foundation in medical journalism and narrative techniques" (Professional Diploma in Medical Journ alism, n.d., para. 4). Another advanced postgraduate d iploma program, which takes 6 8 months, aims to provide high level of understanding of core writing and analytical skills" ( Advanced Post Graduate Diploma in Scientific Writing and Medical Journalism n.d., para. 3). In the Indian definition of medical journalism, medical news "helps inform readers abou t how to live longer and healthier lives, how to avoid unnecessary suffering and how to use resources as wisely as possible" ( Medical Journalism 2011, para. 5). Differing from health journalism education in the United Kingdom and United States, medical journalism education in India attaches great importance to audience interest and narrative skills. For instance, the James Lind Institute elaborated this point as follows: "A journalist should always keep in mind that he/she is narrating a story and not me rely setting down facts" ( 2011, para. 7). Differences of health journalists training among three countries To sum up, the current status of health journalism training in India is worse than that in the U nited Kingdom and the U nited States. In the United Kingdom the development of health journalists training depends not only on formal college
35 education and domestic organizations, but also on other European organizations. In the United States health journalists not only have the oppo rtunity to receive training from international organizations like AHCJ (Association of Health Care Journalists), but also have the opportunity to use resources from world class communication schools like the University of Missouri School of Journalism. In comparison, it seems that health journalists in the India do have fewer resources and opportunities. Given the discussions above, it is reasonable to expect that there would be some different patterns of reporting among newspaper c overage of NDM 1 among th ese three countries. The research question is presented as follows. RQ: Will the presentation of risk related information differ in Indian, UK and U S newspaper coverage of NDM 1? Theoretical Framework Psychometric paradigm People's perception of risk has an enormous impact on their decision making process and hazard management (Slovic, 2000). The psychometric paradigm, which was proposed by Slovic and his colleagues, is the predominant and most widely used approach in research about public attitudes towar d risks (Bronfman, Cifuentes, & Willis, 2007; Renn & Rohrmann, 2000; Slovic, 1987). According to Slovic et al. (1982), the basic idea of the psychometric paradigm is scaling methods and multivariate analysis techniques, researchers have developed "quantitative measures of perceived risk, perceived benefit, and other aspects of perceptions" (Slovic, 2000: p.xxii xxiii); hypothesized a set of risk characteristics; and te sted how these risk characteristics can influence people's risk perception and acceptance (Slovic et al., 1982; Slovic, 2000).
36 Typically, within the psychometric paradigm, researchers have used questionnaires to measure people's risk attitudes and per cept ions about various hazard s including technological risks and natural hazards, such as nuclear power, mad cow disease, avian flu, and so on (Slovic et al., 1982). Fischhoff et al. (1978) used the psy chometric paradigm to study perceived risk for 30 activit ies and technologies, including alcohol, fire fighting, handguns, and so forth. A total of 76 individuals completed anonymous questionnaires. Four measures of risk and benefit were examined: the perceived benefit of each activity, the perceived risk of eac h activity, acceptability of the current risk level, and characterization of risk based on risk dimensions such as dread, knowledge and controllability. The questionnaire data suggested that these risk dimensions might be "effective predictors of the trade off between acceptable risk and perceived benefit" (Fischhoff et al., 1978, p.127). As the study by Fischhoff et al. (1978) demonstrated perceived risk is quantifiable and predictable (Slovic et al., 1982; Slovic, 1987). The psychometric techniques utili zed by Fischhoff et al. (1978) were also found to be effective in risk perception research, numerous similar studies were carried out the following years (Slovic, 1987). Slovic, Fischhoff, and Lichtenstein (1980) conducted a study to investigate how expert s and the public perceive risk. Four different groups of people were involved in the study: college students, members of the League of Women Voters, members from the Active Club, and experts. Participants were asked to rate 30 activities with regard to the fatality and the relationship between perceived risk and annual fatalities, along with nine risk characteristics: voluntariness, immediacy of the effect, knowledge about risk to people, knowledge about risk to science, controllability, newness, potential threat to future generations, dread, and severity of consequences. The researchers found that public perception of risk is a multidimensional task and a
37 wide range of characteristics drive public perception of risk, including dread, severity, newness, pote ntial threat to future generations, and others; in contrast, the number of expected fatalities info rm experts' perception of risk. Thus, researchers within the psychometric paradigm were able to produce Lichtenstein, 1982: p.84), use psychometric techniques such as psychophysical perception (Renn & Rohrmann, 2000), construct cognitive m aps (Slovic, 1987), and investigate the roles of those psychometric variables in influencing people's risk perception and decision making. The utilization of psychometric techniques involves understanding people's behavior, predicting the potential social impact of an event and protecting potential same for everyone (Slovic et al., 1982; Slovic, 1987), research on people's risk perceptions show s that remarkable simila rities can be identified among diverse public groups (Slovic, 2000). In short, the psychometric paradigm has some implications for risk communication and risk management (Slovic, 1987). Risk characteristics Psychometric research suggests that there is a broad domain of risk characteristics that determine the acceptance of risk for the general public. However, there are several different versions of these "general concepts." Slovic, Fischhoff, and Lichetenstein (1995) proposed that risk perception is most strongly determined by the following five risk characteristics: (1) Knowledge about risk: the degree to which a risk is understood; (2) Severity of consequences: the likelihood of the risk causing death ;
38 (3) Dread: the degree to which the event evokes a feeling of dread; (4) Global catastrophic potential: the degree to which a risk may cause catastrophic consequences such as death across the world; (5) Controllability of the risk: the degree to which a risk can be prevented or controlled (such as a decrease in mortality). Some other studies have extended early research and investigated 18 risk characteristics (Slovic et al, 1980; Slovic, Fischhoff, and Lichetenstein, 2000). Using a factor analysis, Slovic et al. (1980) proposed three dimensions of perceived risk (see Tabl e 2 17), and "population" (number 18). The "dread" dimension means that risks are regarded as hard to control, hard to prevent, potentially fatal, having an inequitable distribution of risks and b enefits, catastrophic, globally catastrophic, threatening to future generations, irreducible, increasing in impact, involuntary, and affecting everyone. The "familiarity" dimension refers to difficulty in observing the risk, the immediacy of consequences, newne ss, the lack of understanding among those exposed, and outcomes unknown by science. The "population" dimension means the number of people exposed to the risk. Using the above mentioned risk characteristics, Slovic, Fischhoff, and Lichetenstein (2000) asked people to rate 90 different types of hazards, such as nuclear power, terrorism, DDT, antibiotics, and so on. According to the factor analysis, three "underlying dimensions of factors" were confirmed (see Table 2 2) (Slovic, Fischhoff, & Lichtenstein, 2000: p.141). These finding were also consistent with the research of Slovic, Fischhoff, and Lichtenstein (1982). To summarize, prior literature about psychometric paradigm research holds that there are three main risk characteristics in general: (1) the degree to which a risk is
39 known precisely; (2) the degree to which it evokes feelings of dread; and (3) the population involved in the risk (Slovic, Fischhoff, & Lichtenstein, 1982; Rohrmann & Renn, 2000; Slovic, Fischhoff, & Lichtenstein, 2000). Risk per ception and media coverage Risk perception studies also have pointed out that the media coverage of some unfortunate events (e.g., accidents, environmental pollution ) may greatly influence risk perceptions. Early work with the psyc hometric paradigm emphasi zed one question: What role do risk perceptions play in influencing the impact of unfortunate events? As a result, risk perceptions studies have shown that in addition to the nature of the particular event (e.g., how many people died in an accident) peopl e's risk perceptions also can be influenced by how the event is described in the media, especially when the coverage is biased. In this case, the influence of media can be significant (Slovic, 2000). Thus, in this particular study, the psychometric paradig m is used as a theoretical framework to investigate how newspapers construct and convey health related risk information. When the psychometric paradigm was introduced to communication research for the first time, Fung et al. (2011) studied news coverage of avian flu, using five dimensions of risk characteristics: (1) catastrophic potential information; (2) dread evoking information; (3) uncertainty; (4) perceived controllability; and (5) familiarity. This study used the coding instruments created by Fu ng et al. (2011) to see how risk characteristics are pres ented in coverage regarding NDM 1 in three different countries: India, the United Kingdom, and the United States. Hypotheses This study investigate d news reporting about the NDM 1 superbug in different countries by examining h ow journalists describe NDM 1 in terms of risk dimensions.
40 Based on the risk perception research literature and research goals, the author has proposed a set of hypotheses. A content analysis conducted by Fung et al. (2011) indicat ed that news coverage of avian flu in different countries varied significantly in the inclusion of dread evoking information, presentation of catastrophic potential, uncertainty related information and familiarity related information. As the NDM 1 was init ially isolated in India and has already spread in India and the United Kingdom the researcher expects that more uncertainty, less dread and less familiarity will be observed in U S news coverage regarding the NDM 1. Hence, the following hypotheses have b een proposed: Hypothesis 1a: News coverage of the NDM 1 superbug in Indian and UK newspapers will include more worst case scenario information about ND M 1 risk than coverage in U.S. n ewspapers. Hypothesis 1b: The placement of this worst case scenario infor mation will be more prominent in Indian and UK coverage than in U S coverage. Hypothesis 2a: News coverage of the NDM 1 superbug in Indian and UK newspapers will use more loaded words than coverage in U.S. newspapers. Hypothesis 2b: The placement of loade d words will be more prominent in Indian and UK coverage than in U.S. coverage. Hypothesis 3a: News coverage of the NDM 1 superbug in U.S. newspapers will contain a greater proportion of uncertainty related information than coverage in Indian and UK newspa pers. Hypothesis 3b: The placement of this uncertainty related information will be more prominent in U.S. coverage than in Indian and UK coverage. Hypothesis 4a: In comparison to news coverage in U.S. newspapers, n ews
41 coverage of the NDM 1 superbug in Indi a n and UK newspapers will provide more risk magnitude information about human infection / death. Hypothesis 4b: In comparison to news coverage in U.S. newspapers, n ews coverage of the NDM 1 superbug in India and UK newspapers will provide more information a bout the magnitude of risks for financial loss to society. As the World Health Organization (2010a) reported, although antimicrobial resistance has been an important issue confronting countries worldwide, it is escription of antimicrobials is unregulated unlikely that the situation of of antibiotic misuse in India will have any observable change in the short term (WHO, 2011e). T hus, the researcher infers that the Indian press will consider this issue less controllable. Moreover, based on the health literacy situation in these three countries, the researcher also expects that UK news coverage will contain more information about th e prevention of NDM 1. Further hypot heses are formulated as follows: Hypothesis 5a: In n ews coverage of the NDM 1 superbug UK newspapers will be most likely and Indian newspapers will be least likely to include personal protection information. Hypothesis 5b: In n ews coverage of the NDM 1 superbug UK newspapers will be most likely and Indian newspapers will be least likely to include societal protection information. suggested that n ews reports can characterize a risk by referring to similar issues in the past, which constitutes a key indicator of familiarity related information. For example, in the NDM 1 case, news coverage might refer to the MRSA superb ug
42 (Methicillin resistant Stap hylococcus aureus) which was identified in 1961 and has spread throughout the world. Fung et al. (2011) also found that U.S. news coverage tends to contain more statements regarding the potential of local risks to become global risks. Hence, Hypothesis 6a Hypothesis 6b and Hypothesis 7 are formulated as follows, Hypothesis 6: In n ews coverage of the NDM 1 superbug U.S. newspapers will contain a greater proportion of information regarding comparisons to known risk s than Indian and UK newspapers. Hypothesi s 6b: The placement of information regarding comparisons to known risk s will be more prominent in U.S. newspapers than in Indian or UK newspapers. Hypothesis 7: In n ews coverage of the NDM 1 superbug U.S. newspapers will contain a greater proportion of in formation regarding comparisons to known risk related information in other countries than Indian and UK newspapers. Here, information about risk comparison to other countries refers to "all regions or countries that appeared in each article for comparison" (Fung et al., 2011, p.898).
43 Table 2 1. Health related data in India, United Kingdom, and the United States India United Kingdom The United States Total expenditure on health as % of GDP* 4.2 8.7 15.2 Government expenditure on health as % of total expenditure on health* 32.4 82.6 47.8 Private expenditure on health as % of total expenditure on health* 67.6 17.4 52.2 General government expenditure on health as % of total government expenditure 4.4 15.1 18.7 Infant mortality rate (probability of dyi ng by age 1 per 1000 live births)** 50 5 7 Children under the age of 5 underweight (%)* 43.5 N/A 1.3 Life expectancy at birth (years)** 66.8 80.05 78.37 Life expectancy at birth (country comparison to the world)** 161 28 50 Physicians (per 10,000 popu lation) 6 27.4 26.7 Hospital beds (per 10,000 population)* 9 34 31 Note : World health statistics 2011 2011; ** The Central Intelligence Agency (CIA), 2011a, b, c.
44 Table 2 2. Eighteen risk characteristics Risk characteristics Explanations 1) Controllability Control over the damage of certain unfortunate events 2) Preventability Control over "the occurrence of a mishap" (Slovic, Fischhoff, & Lichtenstein, 2000: p.139) 3) Dread The extent that a risk can make people feel dread 4) Fatality W hether this risk causes any fatal harm; certainty of fatality 5) Equitable distribution of risks and benefits Whether the benefits can be equitably distributed among those at risk (especially for technology related issues, e.g., nuclear power) 6) Catastr ophic potential The degree to which a risk may cause "catastrophic loss of life" (Slovic, Fischhoff, & Lichtenstein, 2000: p.152) 7) Global catastrophic potential The degree to which a risk may cause catastrophic consequences across the whole world 8) F uture impact Whether a risk threatens future generations 9) Reducibility Whether the risk is easily reducible 10) Risk growth Whether the risk is increasing or decreasing 11) Voluntariness Whether people face the risk voluntarily 12) Personal effects Whether the risk affects everyone personally 13) Observability "Are the damage producing processes observable as they occur?" (Slovic, Fischhoff, & Lichtenstein, 2000: p.138) 14) Immediacy of effect Immediacy of consequences of certain unfortunate even ts 15) Newness Whether the risk is new/novel or old/familiar 16) Unknown to those exposed The degree to which a risk is understood by those exposed 17) Unknown to science The degree to which a risk is unknown to science 18) The population involved The number of people exposed to the risk Note: summarized from Slovic, F ischhoff, & Lichetenstein ( 2000 ).
45 CHAPTER 3 METHOD Design This study was designed to investigate the differences in media coverage of the risks posed by the NDM 1 superbug in newspap ers from India, the United Kingdom and the United States These countries were chosen in this study due to the different development stages of NDM 1 in each country It was initially discovered in India, then become prevalent in India and the United Kingdo m, and now is emerging in the United States. The method used for this study is a comparative quantitative content analysis. Sample Newspapers were the chosen medium for analysis because they have several advantages. The first advantage is the accessibilit y of newspaper articles. Further, newspapers have the ability to communicate lengthy, complex and detailed information. More importantly, "issues and events are perceived differently across town, around the country and throughout the world" (The Newsbank, n.d., para. 4); and among all local news sources, "newspapers offer the most extensive, up to date record of emerging and ongoing local issues, personalities and stories of interest" (The Newsbank, n.d., para. 3). Newspaper articles were retrieved from the LexisNexis Academic database language newspaper articles discussing the NDM 1 and published in India, the United Kingdom, and the United States between August 2009 and December 2011. The data range was chosen because the first newspaper article covering the issue appeared in The Guardian (London), August 13, 2009. Duplicated and unrelated articles were excluded. was an acronym for
46 so those articles were considered as irrelevant and were excluded. Procedure After articles were collected and screened for use in the study, the coding of all newspaper articles was completed by the researcher a nd another graduate student who is a native speaker of English. Both of the coders had been trained in quantitative content analysis. First, both of the coders independently coded 20 articles and calculated the intercoder reliability. T he percentage agree ment was good for most questions, except for Q1 (worst case scenarios) and Q10 (societal protection information). After discussion, the coders independently coded another 20 articles and achieved substantial levels of agreement for all questions. Next, one c oder coded about one third of the remaining 226 articles (70 articles) and the other coder coded two thirds (156 articles) Data from each article were entered into the spreadsheet. Measures This study uses a quantitative coding sheet (see Appendix A). E ach news article we coded based on coding items adopted from previous research (Fung et al., 2011; Slovic et al., 1982). Based on the established research instruments developed by Fung et al. (2011), eight major risk dimensions of risk information were cod ed for each article, including: 1. Worst case scenarios: this term indicates whether the news article provides worst case scenario information of NDM 1. For example, a statement such as the following would be coded as worst case scenario information : "We are essentially back to an era with no antibiotics" (WHO, 2010a, para. 2). First, this variable will be coded as present or absent. Second, if it was present, the placement of this variable
47 was coded in terms of whether or not the information appeared in the h eadline, in the lead paragraphs, or in the body of the story. 2. Loaded words: this term indicates whether the news article uses any "emotionally charged language" (Dudo, Dahlstrom, & Brossard, 2007, p.438), such as life threatening, fatal, deadly, kill, ala rming, untreatable, etc. First, this variable will be coded as present or absent. Second, if it was present, the placement of this variable was coded in terms of whether or not the information appeared in the headline, in the lead paragraphs, or in the bod y of the story. Finally, the exact word or phrase which appears in this article will be recorded. 3. Uncertain words: this term indicates whether the news article uses any words to describe NDM 1 related issues as uncertain or unknown, such as not sure, unsur e, unknown, questionable. undetermined, remains to be determined, remains to be seen. First, this variable will be coded as present or absent. Second, if it was present, the placement of this variable was coded in terms of whether or not the information ap peared in the headline, in the lead paragraphs, or in the body of the story. Finally, the exact word or phrase which appears in this article will be recorded. 4. Risk magnitude information about human infection /death: this variable will be coded yes or no for each of the following choices: 0) information about human infection /death is absent; 1) the article provides qualitative/non numerical information about human infection/death; 2) the news article provides quantitative information (chances of infection, nu mber of infected persons) about human infection/death at numerator level (e.g., "there have been 50 cases identified in the UK"); 3) the news article provides quantitative information about human infection/death at numerator/population level (e.g., "at lea st 3% of people infected"). 5. Risk magnitude information about financial loss to society: this variable will be
48 coded yes or no for each of the following choices : 0) information about financial loss to society is absent; 1) the news article provides qualitat ive information about financial loss to society caused by NDM 1 (e.g., "suffered a deadly blow"); 2) the news article provides quantitative information about financial loss (e.g., "dropped by 30%"). 6. Personal protection information: this term refers to per sonal protection information to decrease personal risk of NDM 1 (e.g., wash hands, use antibacterial surface wipes, etc.). First, t his variable will be coded as present or absent. If present, the statement about personal protection was recorded. 7. So cietal p rotection information: T his term refers to any actions taken by domestic and/or foreign governments or international organizations to prevent NDM 1. This variable will be coded as present or absent. 8. Information comparing NDM 1 to known risks: (1) Risk compari son: T his term indicates whether the news story mentions about any other superbugs which are known (e.g., MRSA, MSSA, C Difficile, etc.). This variable will be coded into three categories: 1) one type of other superbugs mentioned ; 2) two or more types of o ther superbugs mentioned ; 3) T he news story doesn't mention any other superbugs. (2) Information about known risks: This variable will be coded into four categories (multiple choices: the coder can choose one or more answers): 1) the article provides qualitati ve/non numerical information about any other superbugs which are known (e.g., MRSA, MSSA, C Difficile, etc.); 2) the news article provides quantitative information (chances of infection, number of infected persons) about other superbugs at the numerator le vel (e.g., "there have been 50 cases identified in the UK"); 3)the news article provides quantitative information about other superbugs at
49 numerator/population level (e.g., "at least 3% of people infected"); 0) information about other superbugs is absent. (3) Ri sk comparison to other countries: any country other than the country where the newspaper was published. For example, for and so on. Firs t, this variable was coded as present or absent. If present, the country name was recorded. As has been noted previously that, Fung et al. (2011) investigated five dimensions of risk characteristics in their study of Hong Kong and U.S. coverage of avian f lu. However, their coding instrument for the risk dimension of catastrophic potential information does not apply in the case of NDM 1. In order to examine catastrophic potential information, Fung et al. (2011) examined the presence of historical flu outbre aks, including the 1918 Spanish Influenza, the 1957 Asian fl u, and the 1968 Hong Kong flu. These historical flu outbreaks had one point in common: all of them killed millions of people within a short time. But in the case of NDM 1, there is no record of ma jor superbug outbreaks which were as disastrous as those historical flu outbreaks. Thus, the risk dimension of catastrophic potential information could not be examine d in this study.
50 CHAPTER 4 RESULTS Review of Method Descriptions of the Sample The total sample included 266 news articles, of which 148 were from newspapers in India, 97 were from the United Kingdom and 21 were from the United States. Articles from India accounted for 55.6% of the total sample, while articles from the United Kingdom and the United States accounted for 36.5% and 7.9%, respectively. This distribution is consistent with the different development stages of the NDM 1 in each country. The unit of the analysis was each individual article. The average length of newspaper articles i n India, the United Kingdom and the United States were 512 words (range = 127 to 1795 words) 417 words (range = 64 to 1572 words), and 573 words (range = 133 to 1393 words) respectively Not surprisingly, the appearance of Indian and UK news articles deal ing with th e NDM 1 tended to increase after 2009 (see Ta ble 4 1). For example, in the United Kingdom there were four articles concerning NDM 1 in 2009, 23 articles in 2010, and 70 articles in 2011. However, in the United States, the number of articles on the NDM 1 issue tended to decrease from 2010 (18 articles) to 2011 (three articles). The U.S. news media's lack of interest is possibly the main reason for this phenomenon. Intercoder R eliability lity, the overall percentage of the intercoder agreement was .90. With some variables, such as personal protection information, risk magnitude information about human infection and i nformation about risk comparison to other countries, the intercoder agree ment achieved .95. The lowest percentage of agreement was for the worst case scenarios,
51 which achieved .81 (See Appendix C) Statistical analysis The researcher conducted chi square tests to investigate potential differences among newspaper articles from t he various countries in their use of risk information, as mentioned in the chapter of methodology. Review of Findings The coding process uncovered an interesting phenomenon well worth further investigation. Apart from "pure" health news, this study identif ied a number of articles from India that focused on why the deadly gene was named after the Indian capital of New Delhi or how the medical tourism business in India would be affected by the NDM 1 crisis. Some articles denied the NDM 1 problem by stating th at the issue was international slander. They claimed that Western countries had ulterior motives and were trying to discredit the medical tourism industry in India. Mostly published in 2010 (the early stage of the NDM 1 problem), those articles avoided des cribing the NDM 1 issue as a public health risk. Instead, they tended to expre ss the following opinions: (1) T he Lancet study, which was conducted by British scholars and raised public awareness of NDM 1 for the first time, was unscientific; (2) T he Lancet study was supported by some pharmaceutical compan ies that make antibiotics; (3) T he likelihood is great that the problem is not as serious as feared; and (4) Western countries want to suppress the booming medical tourism industry in India, as "150,000 Ame ricans travelled overseas last year for medical care to save on costs"; thus the NDM 1 crisis had became "a PR nightmare for India's expanding health tourism industry" (Choudhury, 2010). Although a number of news articles in India portrayed the NDM 1 supe rbug issue as a huge controversy, an unscientific conclusion, or a slander effort with ulterior
52 motives and unworthy of attention, many Indian news articles also warned of the potential dangers of NDM 1, noting that it is untreatable or unmanageable. Some articles in India also covered the severe situation of antibiotic over use and resistance in India, as well as necessary measures that should be taken. Moreover, since September 2010, Indian newspapers began to request ask the public to wash their hands fre quently or boil water before drinking In April 2011, Indian newspapers announced that the government had began to finalize new guidelines to prevent misuse of strong antibiotics, indicating that the Indian government had realized the problem of irrational use of antibiotics and started to take corresponding measures to solve the problem. Major findings from this study are presented in detail below. H1a: Presence of W orst C ase S cenarios H1a posited that the coverage of the NDM 1 in India and the United King dom would include more worst case scenarios about the risk of NDM 1 than coverage in the United States. Table 4 2 shows the presence of worst case scenarios in the news coverage of NDM 1 in each country. C overage in the United Kingdom and the United States w as more likely to cover worst case scenarios than those in India (India & UK: (1, N = 245) = 3.94, p (1, N = 169 ) = 4.06 p < .05 1 ). Also, as 28.5% of U.S. articles and 21.6% of UK articles mentioned worst case scenarios, t (1, N = 118 ) = .47, p > .05). Therefore, H1a was not supported. H1b: Placement of W orst C ase S cenarios H1b posited that the placement of worst case scenarios would be more 1 P value is the significance level For example, a p value of 0.10 means that there is a 10% chance that this deviation is due to chance alone. Based on standard p < 0.05, this is not within the range of acceptable deviation and the difference is statistically significant.
53 prominent in Indian and UK cov erage than in U.S. coverage. The placement of worst case scenario statements w as coded into three categories: headlines, lead paragraphs, and story bodies. As shown in Table 4 3, four of 148 (2.7%) of Indian news articles and two of 21 (9.5%) U.S. news art icles mentioned worst case scenarios in the headlines or lead paragraphs. Thus the placement of worst case scenarios was not more prominent in Indian coverag (1, N = 169 ) = .90, p >0.05). Moreover, as six of 97 (6.2%) UK news articles presented worst case scenarios in the headline or lead paragraph, there was no significant difference between the presence of worst case scenarios in U K and U.S. coverage (1, N = 118 ) = .01, p > .05). Therefore, H2b was also not supported. H2a: The P resence of L oaded W ords H2a posited that India and UK coverage of NDM 1 would use more loaded words to describe NDM 1 risk than U.S. coverage, because th ere have been more contracted cases in India and the United Kingdom. But as Table 4 4 shows a significant ly larger number of loaded words was presented in articles from the United Kingdom and the United States than in (1, N = 245 ) =31.06, p (1, N = 169 ) =5.70, p < .05), and there was no significant difference between the United Kingdom and the United States concerning (1, N = 118 ) = .66, p > .05). Therefore, H2a was not s upported. Table 4 6 shows the most frequently used loaded words in each country. It is were the most frequently used loaded word in all three countries worrying/worryingly
54 H2b: Placement of L oaded W ords H2b posited that the placement of loaded words would be more prominent in news coverage of NDM 1 from India and the United Kingdom. Compared with U.S. coverage, a much smaller proportion of loaded words was presented in lead (1, N = 169 ) = 28.85, p < .01, see Table 4 5), while there was no significant difference between UK and the U.S. coverage (headline: (1, N = 118 ) = 3.61, p > .05; lead paragraph: (1, N = 118 ) = 2.20, p > .05). Therefore, H2b was not supported. H3a: Use of W ords R eflecting U ncertainty H3a proposed that a greater proportion of U.S. news coverage than Indian and UK news coverage would contain statements reflecting uncertainty As Tab le 4 7 shows, the U.S. news paper s were most likely to use words suggesting uncertainty About 38.1% of articles from the United States included uncertainty related information regarding the NDM 1, while the percentage for articles in India was only 11.5%. Thus, the use of words suggesting uncertainty was significant ly different between India n and U.S. coverage ( (1, N = 169 ) = 8.33, p < .01). This difference was also observed between the United Kingdom and the United States ( (1, N = 118 ) = 21.05, p < 01), as only 3.1% of UK articles presented uncertainty related information. This hypothesis was supported. Table 4 9 shows the most frequently used uncertain ty reflecting words in news coverage for each country. For example, Indian and UK newspapers were more likely while U.S. newspapers were H3b: Placement of U ncertain W ords H3b posited that the placement of uncertainty related information would be more
55 prominent in U.S. coverage th an in India and UK coverage. But the study found that all words reflecting uncertainty were placed in the body of the story regardless of country No uncertain ty connoting words were found in headlines or lead paragraphs. Therefore, H3b was not supported (see Table 4 8). H4a: Information of R isks of H uman I nfection/ D eath H4a proposed that news coverage of the NDM 1 superbug in India and the United Kingdom would provide more information about the magnitude of risks for human infection or death. As Table 4 10 exhibits, articles in the U.S. presented more qualitative level of information on human infection than those in India ( (1, N = 169 ) = 44.35, p < .01) and UK ( (1, N = 245 ) = 42.28, p < .01). At the numerator level, articles in UK provided more nume ric information on human infection than those in India ( 245 ) = 61.91, p < .01) and the U.S. ( 118 ) = 10.80, p < .01), and there were no significant difference between India and the U.S. ( 169 ) = 2.24, p > .05). But at the numer ator and population level, India was the only country that offered information on human infection. Therefore, H4a was partly supported. H4b: Risk I nformation A bout F inancial L oss H4b predicted that news coverage in India and the United Kingdom would conta in more information about the magnitude of risk for economic losses caused by the NDM 1 superbug. As Table 4 11 shows, very few articles in the three countries mentioned economic losses caused by the NDM 1 superbug. At the qualitative level, the data did n ot show significant difference among these countries (e.g., between the United Kingdom and the United States, 118 ) = .07, p > .05). At the quantitative level, none of these three countries presented information about the risk of financial loss Therefore, H4b was not supported.
56 H5a: Personal P rotection I nformation H5a posited that UK newspapers would be most likely to present personal protection information and Indian newspapers would be least likely to teach people how to protect themselves agai nst superbugs. In comparison with news articles in India, a larger proportion of news articles in the United Kingdom ( 245 ) = 6.99, p < .01 ) and the United States ( 169 ) = 8.09, p < .01 ) presented personal protection information (see Table 4 12). N ews articles from the United Kingdom and the United States did not significantly differ in presenting pers onal protection information ( 118 ) = .64, p > .05). Therefore, H5a was partially supported. Personal protection information in this study included instructions for decreasing 1 through good hygiene, appropriate antibiot ic use and safe drinking water. As is shown in Table 4 13, coverage in the United Kingdom and the United States was more likely to mention good hygiene (e.g., wash hands frequently and properly, use hand gels and antibacterial surface wipes) than those in p 7.24, p < .05). Also, as 23.8% of U.S. articles and 17.5% of UK articles mentioned measures to maintain personal and environmental hygiene there was no significant di p > .05). In addition, coverage in the United States was more likely to promote appropriate antibiotic use than those in p < .05). On the other hand, India was the only country t hat offered information on drinking water safety (e.g., keep the water pipeline clean, drink purified water or drink boiled water). H5b: Societal P rotection I nformation Societ al protection information referred to any actions taken by domestic and/or
57 fore ign governments or international organizations to prevent NDM 1 or reduce its impact For example these organizations include d the World Health Organization (WHO), the Atlanta based Center for Disease Control and Prevention (CDC), the National Health Serv ice (NHS) in the United Kingdom, the India based National Centre for Disease Control (NCDC), and other research centers. H5b posited that articles in the United Kingdom would be most likely and articles in India would be least lik ely to present such infor mation. As Table 4 14 exhibits, 66 % of UK news articles, 38.1% of U.S. news articles and 29.1% of Indian news articles provided societal protection information when covering NDM 1. Significant differences were shown between the United Kingdom and the Unite d States ( = 118 ) = 5.64, p < .05) and between the United Kingdom and India ( 245 ) = 32.48, p < .01). There was no difference between India and the United States ( N = 169 ) = .71, p > .05). Therefore, H5b was partially supported. H6 a: I nformation C omparing NDM 1 R isk to K nown R isks P resence of O ther S uperbugs and I nformation T ype H6a posited that U.S. newspapers would be most likely to contain comparisons to known risk related inf ormation. As shown in Table 4 15 no significant dif ference was found between the United States and India ( 169 ) = 1.45, p > .05) and between the United States and the United Kingdom ( 118 ) = 1.99, p > .05). Moreover, a larger proportion of news articles from the United Kingdom mentioned two or more types of other superbugs in comparison to articles from India ( 245 ) = 29.58, p < .05) an d the United States ( 118 ) = 5.65, p < .05); thus, U.S. newspapers did not contain a greater proportion of comparisons to other superbugs than UK ne wspapers. Further, as Table 4 16 shows, I ndia was the least likely to provide qualitative and numeric information regarding comparisons to other
58 superbugs, but the U.S. and UK coverage was not significant ly different either on the qualitative level ( 118 ) = .85, p > .05) or the numerat or level ( 118 ) = 1.20, p > .05). Therefore, H6a was not supported. H6b: Placement of I nformation C omparing NDM 1 to K nown R isks H6b posited that the placement of risk information related to other superbugs would be more prominent in U.S. covera ge than in Indian and U K coverage. But, as Table 4 17 shows, news articles from the United Kingdom and the United States did not mention other superbugs in the headlines or in lead paragraphs, while a few Indian articles did (1.4%). Therefore, H6b was not supported. H7: Information C omparing R isks in O ther C ountries H7 posited that news coverage of the NDM 1 superbug in U.S. newspapers would present more comparisons about risks related to NDM 1 in other countries. In general, news stories in the United King dom and the United States were more likely to contain information about risk comparison related to other countries tha n those in India (see Table 4 18 ), but there was no significant difference between UK and U.S. coverage ( 118 ) = .73, p > .05). U.S. news articles were more likely to mention Canada and Australia as countries affected by NDM 1. As mentioned previously, the NDM 1 superbug was initially discovered in India; there was no significant difference between UK and U.S. coverage regarding co mparisons to risks related to NDM 1 in India, but UK news articles were more likely to mention India's neighboring countries, such as Pakistan and Bangladesh. Therefore, H7 was partly supported.
59 Table 4 1. Numbers of n ewspaper a rticles on NDM 1 in e ac h c ountry India United Kingdom The United States Total 2009 0 4 0 4 2010 62 23 18 103 2011 86 70 3 159 Total 148 97 21 266 Table 4 2. Frequencies of w orst c ase s cenarios in n ews c overage Frequency of worst case scenarios (Presence) Percentage (Presence) India 18 (Total = 148) 12.2% United Kingdom 21 (Total = 97) 21.6% The United States 6 (Total = 21) 28.5% India & United Kingdom: 3.94 (p = .047< .05) In dia & The United States: 4.06 (p = .04) United Kin gdom & The United States: .47 (p = .49) Table 4 3. Placement of w orst c ase s cenarios in n ews c overage India N = 148 (%) United Kingdom N = 97 (%) The Uni ted States N = 21 (%) In the headlines 3 (2.0%) 2 (2.1%) 0 (0%) In the lead paragraphs 1 (0.7%) 4 (4.1%) 2 (9.5%) In the headlines or lead paragraphs (not replicated) 4 (2.7%) 6 (6.2%) 2 (9.5%) In story body 14 (9.5%) 16 (16.5%) 4 (19%) (In the headlines or lead paragraphs) India & United Kingdom: 1.03 ( p =.31) India & The United States: .90 ( p =.34) United Kingdom & The United States: .01 ( p =.94)
60 Table 4 4. Frequencies of l oaded w ords in n ews c overage Frequency of loaded words ( p resence) Percentage ( p resence) India 79 (Total = 148) 53.4% United Kingdom 85 (Total = 97) 87.6% The United States 17 (Total = 21) 81.0% India & United Kingdom: 31.06 ( p = .00) India & The United States: 5.70 ( p = .02) United Kingdom & The United S tates: .66 ( p =.42) Table 4 5. Placement of l oaded w ords in n ews c overage India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) In the headlines a 11 (7.4%) 43 (44.3%) 4 (19.0%) In the lead paragraphs b 19 (12.8%) 40 (41.2%) 13 (6 1.9%) In story body 61 (41.2%) 71 (73.1%) 12 (57.1%) a India & United Kingdom: 46.43 ( p = .00) India & The United States: 3.07 ( p = .08) United Kingdom & The United States: 4.6 ( p = .03) b India & United Kingdom: 25.85 ( p = .00) India & The United States: 28.85 ( p = .00) United Kingdom & The United State s: 2.98 ( p = .08)
61 Table 4 6. Most f requently u sed l oaded w ords in n ews c overage India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) 1 Warn / Warning 26 (17.6 %) Warn / Warning 54 (55.7%) Warn / Warning 9 (42.9%) 2 Deadly 15 ( 10.1%) Threat / Threatening 42 (43.3%) Dangerous 9 (42.9) 3 Threat 15 (10.1%) Fear 39 (40.2%) Fear 8 (38.1%) 4 Worrying / Worryingly / Worrisome 15 (10.1%) Alarm /Alarming 22 (22.7%) Worry / Worrying(ly) / Worrisome 7 (33.3%) 5 Serious 12 (8.1%) Al ert 20 (20.6%) Alarm 4 (19.0%) 6 Dangerous / Danger 10 (6.8%) Deadly / Deadliest 17 (17.5%) Serious 3 (14.3%) 7 Alarm 6 (4.1%) Danger /Dangerous 13 (13.4%) Alert 3 (14.3%) 8 Alert 4 (2.7%) Worry / Worrying / Worryingly 11 (11.3%) Threat / Threatening 3 (14.3%) 9 Untreatable 4 (2.7%) Serious 10 (10.3%) Scary 2 (9.5%) 10 Fear 4 (2.7%) Life threatening 5 (5.2%) Kill / Killer 2 (9.5%) Continue Severe 5 (5.2%) Continue U ntreatable 5 (5.2%) Note : All the data refers to the number of articles used a particular word, not the frequency of occurrence of a word)
62 Table 4 7. Frequencies of u ncertain w ords in n ews c overage Country name Frequency of uncertain words Percentage India 17 (Total = 148) 11.5% United Kingdom 3 (Total = 97) 3.1% The United States 8 (Total = 21) 38.1% India & United Kingdom: 4.44 ( p =.04) India & The United States: 8.33 ( p =.00) United Kingdom & The United States: 21.05 ( p =.00) Table 4 8. Placement of u ncertain w ords in n ews c overage India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) In the headlines 0 (0%) 0 (0%) 0 (0%) In the lead paragraphs 0 (0.6%) 0 (0%) 0 (0%) In story body 17 (11.5%) 3 (3.1%) 8 (38.1%) Table 4 9. Most frequently u sed u ncertain w ords in n ews c overage Country name Uncertain words India Don't k now, no evidence, unknown, unpredictable United Kingdom Don't know, unclear, uncertain The United States Unknown, no consensus on, not sure, to o early to judge
63 Table 4 10. Risk m agnitude of h uman i nfection/ d eath Types of risk information about human infection /death India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) Qualitative level a 26 (17.6%) 15 (15.5%) 18 (85.7%) Numerator level b (Numeric information) 40 (27.0%) 76 (78.4%) 9 (42.9%) Numerator / population level 3 (2.0%) 0 (0%) 0 (0%) a India & United Kingdom: .19 ( p = .67) India & The United States: 44.35 ( p =.00) United Kingdom & The United States: 42.28 ( p =.00) b India & United Kingdom: 61.91 ( p =.00) India & The United States: 2.24 ( p = .13) United Kingdom & The United States: 10.80 ( p = 00) Table 4 11. Risk m agnitude of f inancial l oss Types of risk information about financial loss India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) Qualitative level a 3 (2.0%) 1 (1.0%) 1 (4.8%) Quantitative level 0 (0%) 0 (0%) 0 (0%) a India & United Kingdom: .01 (p=.93) India & The United States: .00 (p=1.00) United Kingdom & The United States: .07 (p=.79) Table 4 12. Frequencies of p ersonal p rotection i nformation in n ews c overage Frequency of personal protection information Percentage India 12 (Total = 148) 8.1 % United Kingdom 19 (Total = 97) 19.6 % The United States 6 (Total = 21) 28.6% India & United Kingdom: 6.99 ( p = .008 ) India & The United States: 8.09 ( p = .00 ) United Kingdom & The United States: .64 ( p = .46)
64 Table 4 1 3. Personal p rotection i nformation in n ews c overage Personal protection information India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) Maintain good hygiene ( personal hygiene, environmental cleaning ) a 7 (4.5%) 17 (17. 5%) 5 (23.8%) Drink boiled water or purified water / keep the water pipeline clean 8 (5.4%) 0 (0%) 0 (0%) Take antibiotics appropriately, b 2 (1.4%) 3 (3.1%) 3 (14.3%) a India & United Kingdom: 10.85 ( p =.00) India & The United States: 7.24 ( p =.01) United Kingdom & The United States: .04 ( p =.84) b India & United Kingdom: 0.23 ( p =.63) India & The United States: 6.68 ( p =.01) United Kingdom & The United States: 2.46 ( p =.12) Table 4 14. Frequencies of s ocietal p rotection i nformation in n ews c overage Country name Frequency of societal protection information (Presence) Percentage (Presence) India 43 (Total = 148) 29.1% United Kingdom 64 (Total = 97) 66.0% The Un ited States 8 (Total = 21) 38.1% India & United Kingdom: 32.48 ( p = .00) India & The United States: .71 ( p = .40) United Kingdom & The United States: 5.64 ( p = .02) Table 4 15. Comparison to o ther s uperbugs p resence India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) One type of other superbugs 14 (9.5%) 9 (9.3%) 4 (19.0%) Two or more types of other superbugs a 12 (8.1%) 35 (36.1%) 2 (9.5%) Total b 26 (17.6%) 44 (45.4%) 6 (28.6%) a India & United Kingdom: 29.58 ( p = .00) India & The United States: .05 ( p = .83) United Kingdom & The United States: 5.65 ( p = .02) b India & United Kingdom: 22.18 ( p = .00) India & The United States: 1.45 ( p = .23) United Kingdom & The United States: 1.99 ( p = .16)
65 Table 4 16. Comparison to o ther s uperbugs i nformation t yp e Types of risk i nformation about risk comparison to other superbugs India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) Qualitative level a 7 (4.7%) 15 (15.5%) 5 (23.8%) Numerator level b (Numeric information) 6 (4.1%) 19 (19.6 %) 2 (9.5%) Numerator / population level 2 (1.4%) 0 (0%) 0 (0%) a India & United Kingdom: 8.26 ( p = .00) India & The United States: 10.15 ( p = .00) United Kingdom & The United States: .85 ( p = .36) b India & United Kingdom: 15.43 ( p = .00) India & The United States: 1.22 ( p = .27) United Kingdom & The United States : 1.20 ( p = .27) Table 4 17. Placement of r isk c omparison to o ther s uperbugs in n ews c overage Placement of risk comparison to other superbugs India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) In the headlines 1 (0.7%) 0 (0%) 0 (%) In the lead paragraphs 1 (0.7%) 0 (0%) 0 (%) In story body 27 (18.2%) 44 (45.4%) 7 (33.3%)
66 Table 4 18. Risk c omparison to o ther Countries India N = 148 (%) United Kingdom N = 97 (%) The United States N = 21 (%) Comparison to other countrie s (overall) a 66 (44.6%) 75 (77.3%) 18 (85.7%) Comparison to India / 66 (68.0%) 16 (76.2%) Comparison to Pakistan, Bangladesh or Indian sub continent 25 (16.9%) 56 (57.7%) 5 (23.8%) Comparison to other Asian countries (Israel, Turkey, China, Hong Kong, Taiwan, South Korea, Japan or Singapore) 16 (10.8%) 9 (9.3%) 3 (14.3%) Comparison to UK 41 (27.7%) / 13 (61.9%) Comparison to other European countries (Sweden, Germany, Italy, Greece, France, the Netherlands, Belgium, France, Israel, Turkey, Portugal and Spain) 21 (14.2%) 18 (18.6%) 8 (38.1%) Comparison to the U.S. 17 (11.5%) 12 (12.4%) / Comparison to Canada 9 (6.1%) 9 (9.3%) 13 (61.9%) Comparison to Australia 10 (6.8%) 9 (9.3%) 13 (61.9%) a India & United Kingdom: 25.68 (p= .00) India & The Unit ed States: 12.43 (p= .00) United Kingdom & The United States: .73 (p= .39)
67 CHAPTER 5 DISCUSSION This study was designed to assess how newspapers in three countries are covering the NDM 1 superbug and how the coverage varies in three countries. Using the psychometric paradigm as a theoretical framework, the study provides quantitative analysis of NDM 1 superbug related coverage in news paper article s in three different countries: India, the United Kingdom and the United States. The following dimensions of risk characteristics were examin ed in this study: (1) dread evoking information, (2) uncertainty, (3) controllability, and (4) familiarity. While the previous chapter described how newspapers in India, the United States, and the United Kingdom covered the NDM 1 issue, this chapter presents major implications from this study, limitations, and recommendations for future research. Major Findings and Implications Dread evoking I nformation Worst case scenarios, loaded words, and information about the magnitud e of risks for human infection or death or for financial loss es related to NDM 1 are all types of dread evoking information. First, this study showed that UK and U.S. newspapers were more likely to include worst case scenarios and loaded words than were In dian newspapers, and the placement of loaded words was more prominent in the UK and the U.S. coverage. This phenomenon may be the result of India's resis tance to the public attention to NDM 1, especially with the consideration of India's booming medical to urism industry. Second, UK newspapers were most likely to include quantitative information about the risks of human infection or death, while U.S. newspapers were most likely to provide qualitative information about the magnitude of the risk of human inf ection or death
68 number of NDM Although some scientists believed that the NDM 1 superbug is an "Indian superbug," the media message in India was n ot scarier than that in the United States, where far fewer cases have been found than in India. This phenomenon shows the emphasis on health related issues in the British and American societies. Perhaps the most noticeable difference between this study an d previous research (e.g., Fung et al., 2011) is that, in this study, not all countries involved consistently treated the NDM 1 as a public health problem. In other words, the problem may lie in India was the exceptio n. In 2011, some articles were published with titles such as "Delhi Superbug Is New Global Health Threat" or "Wake Up! The 'Superbug' Threat Is Real," but this was not the case in 2010. During the first few months following the Lancet paper that identified the NDM 1 enzyme ( August 2010 ), many Indian news articles took a denial or debate stance, rather than encouraging awareness or action. Therefore, it was not surprising that coverage of NDM 1 in India did not present as much dread evoking information as wa s expected. Uncertainty This study also examined the presence and placement of information suggesting uncertainty about the NDM 1 superbug in the news coverage. In general, only a few news articles included uncertainty related information regarding NDM 1 As Table 4 7 shows, of the 266 articles examined in the current study, only 28 used uncertain ty connoting words such as "unknown" or "unclear." However, a greater proportion of U.S. articles (38.1%) presented uncertain reflecting statements than did arti cles in India and the UK. This finding is consistent with the previous research that remoteness can affect the presence of uncertain words in news coverage of public
69 health risks (Fung et al., 2011). The high proportion of articles in the U.S. may be the r esult of geographical remoteness; stories of NDM 1 in India might seem geographically remote to most U.S. citizens not traveling to South Asia. People can easily recognize uncertain reflecting words in news stories (Johnson and Slovic, 1995), and such unc ertainty related information influences how a risk is understood by people/science (Slovic et al., 1980; Slovic, Fischhoff, and Lichetenstein, 2000). Therefore, the investigation into uncertainty related information has important implications for risk comm unication research; when the uncertainty of a risk is stressed in the press, people's perceptions of risk will possibly be affected. Controllability Significant differences were shown in the dimension of controllability among newspaper coverage in three co untries. This dimension of controllability was measured by the presence of personal protection information and societal protection information. Fung et al. (2011) found that information about personal protection was seldom mentioned, and there were no sign ificant differences in personal protection information between the Hong Kong and U.S. coverage of avian flu. However, the current study found that although personal protection information such as hand hygiene was also infrequently mentioned in all three co untries, Indian newspapers were least likely to present personal protection information such as advice on good hand hygiene. One possible reason is that many Indian articles were still debating the seriousness of NDM 1. As a result, health educator and com munication practitioners may be reluctant to provide personal protection information. Another possible reason is that the state of both health literacy and media literacy in the United Kingdom and the United States is relatively better than that in develop ing countries.
70 The finding suggested some practical implication for health reporters and health educators. News articles in all three countries mentioned that basic sanitation practice s are the most important personal protection measure s suggesting that health reporters and health educators should be encouraged to promote hygiene practices like hand washing in coverage of NDM 1 and other easily transmissible risks However, very few newspaper articles in the three countries told their readers to prevent o veruse or inappropriate use of antibiotics, suggesting that more measures should be taken to promote appropriate antibiotic use. Also worth noting is the fact that only India n newspapers present ed advice on drinking water, as British scientists stated that the NDM 1 gene had been found in community waste seepage (water pools in streets or rivulets) and tap water in New Delhi, India (Walsh, Weeks, Livermore, & Toleman). Second, UK news articles were most likely to mention societal protection information such as actions taken by governments and organizations to prevent NDM 1; a significantly smaller proportion of coverage in India and the U.S. mentioned such information. One explanation is that the United Kingdom has a national health service that guarantees a ll citizens have access to health care service. As mentioned in the section discussing health care systems in the literature review, government spending on and government control of health care differs greatly between the United Kingdom and India and betwe en the United Kingdom and the United States as health services in the United Kingdom are primarily financed by the government. Therefore, the NHS has been frequently mentioned in domestic health reporting. A large proportion of articles in the United King dom mentioned NHS hospitals, NHS patients, and the researchers, scientists and clinicians working in the NHS. Another explanation might be that the UK treated this issue as a serious threat, while India did
71 not consider it a pressing problem during the fir st few months and only a few cases were identified in the United States. Familiarity The dimension of familiarity was measured by analyzing the inclusion of information comparing NDM 1 risk to known risks and i nformation comparing the NDM 1 risk in the h ost country versus other countries. In terms of the presence of risk comparisons to other known risks (other superbugs), news coverage of NDM 1 in the United Kingdom and the United States included more comparisons than stories in India. Although the resear cher expected that the United States would include more information about other superbugs, there was no significant difference between UK and U.S. coverage. Additionally, UK newspapers were most likely to mention two or more types of other superbugs. One p ossible explanation is that a number of cases of NDM 1 had been identified in the United Kingdom and UK reporters consistently regarded it as a serious and urgent problem. With regard to i nformation comparing NDM 1 risks in the host country versus in other countries the data showed that news articles in India presented the least amount of information about risk compari son related to other countries. One possible reason is that the NDM 1 was initially identified in India, but Indian newspapers did not want It is not surprising that the U.S. news articles wer e more likely to mention Canada because of the countries' proximity. Interestingly, U.S. news articles were more likely to mention Australia, while UK news articles were more likely to mention India's neighboring countries such as Pakistan and Bangladesh. One possible explanation for this phenomenon is that health writers often rely on announcements from domestic public health departments to get info rmation on a public health issue in
72 other countries, so articles from the same country may repeat the same country names as those included in public health department documents Summary In sum, the current study expanded the application of the psychometric framework of risk perception to examine news coverage of global health risks. The current study supports previous research showing that social factors, such as geographical factors and social/cultural context of the involved country, may "influence the wa y newspapers report on a distant risk" (Fung et al., 2011, p.902). The study indicated the complexity of global health communication considering the features of the NDM 1 itself and complicated contextual variances across different regions. For example, I ndian newspapers denied the existence of the NDM 1 problem for a long time allegedly to protect the reputation of India's medical tourism business. On the other hand, the media messages may influence the way audience s think about the risk (Slovic, 2000); i f people in different regions perceive health threats very differently, it may become harder for countries t o gain international support to reduc e the risk (Fung et al., 2011). The findings suggested that the psychometric framework of risk perception may b e useful in understanding and predicting people's risk perception s and psychological acceptance of other health risks. Meanwhile, it is advis able that decision makers take more social factors into consideration to ensure the public receives correct and tra nsparent health information. Limitations Several limitations of this study should be noted. As described in the methods section, this study drew articles from newspapers in only three countries from an online database, so the results may not be generalizab le to other media platforms,
73 such as magazines, television, and network media. Also, the sampling process from an electronic news database made it impossible to take other facto rs into consideration, such as h eadline font (including font size and color), u se of images, and so on. A further limitation of this study is the lack of information about material types. For many articles examined in the study, the LexisNexis database does not specify if the article is an editorial, an informative article, a column, or anything else. Thus there is no ability to do further analysis. Furthermore, the cultural differences among these three countries may pose another limitation for this study, although it is very difficult to avoid in comparative research. Future Researc h Recommendations First, only the newspaper coverage of health risks was examined in this study. If this study were to be replicated, i t may be useful to include reports from other types of media platforms, such as magazines, health websites, health blogs, and so on. This study is also limited by the period of the evaluation. The researcher believes that new studies could also examine similar risk scenarios longitudinally and measure changes over a longer period. This could be particularly helpful for futur e research to reveal the changes of coverage in different developmental stages of public health issues. Further, this study did not examine the actual impact of health risk information on society across countries. Lab experiments can be designed to assess the relationship between risk perception and behavioral outcomes
74 A PPENDIX A CODING SHEET Item ID # __ ____ __ Coder n ame: DF / BB Code date: ____/____/2012 Basic information Newspaper Date: _____ /_____ /20___ Worst case scenarios 1. Does the n ews article provide worst case scenario information of NDM 1? (H1a) 1) Yes Go to Q2 2) No Go to Q3 2. Where is the worst case scenario information placed in the article? (H1b) 1) In the headline 2) In the lead paragraph 3) In story body Loaded words 3. Does the news article use any loaded words? (H2a) 1) Yes Go to Q4 2) No Go to Q5 4. Where are the loaded words placed in the article? (H2b) 1) In the headline 2) In the lead paragraph 3) In story body Specify the "loaded words": _____________________________________
75 Uncertain words 5. Does the news article use any uncertain words? (H3a) 1) Yes Go to Q6 2) No Go to Q7 6. Where are the uncertain words placed in the article? (H3b) 1) In the headline 2) In the lead paragraph 3) In story body Specify the "uncertain words": ____________________________________ Risk magnitude information about human infection /death 7. What types of information does the news article provide regarding the human infection/death cau sed by NDM 1? (H4a) 1) Qualitative risk information (non numerical NDM 1 related risk statements) 2) Quantitative risk information numerator level 3) Quantitative risk information numerator/population level 0) None Risk magnitude information abou t financial loss to society 8. What types of information does the news article provide regarding the financial loss to society caused by NDM 1? (H4b) 1) Qualitative risk information (non numerical NDM 1 related risk statements) 2) Quantitative risk info rmation (numerically based NDM 1 related risk statements) 0) None
76 Personal protection information 9. Regarding the NDM 1 risk, does the news article provide any personal protection information to decrease personal risk of NDM 1? (H5a) 1) Yes 2) No Societal protection information 10.Regarding the NDM 1 risk, does the news article provide any societal protection information to prevent people from NDM 1? (H5b) 1) Yes 2) No I nformation about risk comparison to known risks 11. Does the news article compare the risks of NDM 1 to other similar risk scenarios? (H6a) 1) Yes, one Go to Q12 2) Yes, more than one Go to Q12 3) No Go to Q14 12. What types of information does the news article provide regarding other similar risk scenarios? (H6a) 1) Qualitative risk information (non numerical NDM 1 related risk statements) 2) Quantitative risk information (numerator level) 3) Quantitative risk information (numerator/population level) 0) None
77 13. Where are the similar risk scenarios placed in the article? (H6b) 1) In the headline 2) In the lead paragraph 3) In story body 14. Does the news article provide i nformation about risk comparison to other countries? (H7) 1) Yes, specify the co untry name____________________________ 2) No
78 A PPENDIX B CODING GUIDELINES Basic information Date MM / DD /20 Worst case scenarios 1. Worst case scenarios indicate whether the news article provides worst case scenario information of NDM 1 by circling yes or no. E xample s of the worst case scenarios are: "We are essentially back to an era with no antibiotics." (WHO, 2010a, para. 2) "Less than a handful of antibiotics are currently in the pipeline to combat antibiotic resistant bacteria, and the worldwide sprea d of severe resistance genes is considered a nightmare scenario." (WHO, 2011 c, para. 9) 2. Placement of worst case scenario information if the news article provides any information about worst case scenario indicate where the information is placed: 1) h eadline; 2) lead paragraph; 3) story body Loaded words 3. Loaded words indicate whether the news article uses any "emotionally charged language" (Dudo, Dahlstrom, & Brossard, 2007, p.438), such as life threatening, fatal, deadly, kill, alarming, untreatab le, etc. 4. Placement of loaded words if the news article uses loaded words indicate where the word(s) is placed: 1) headline; 2) lead paragraph; 3) story body. Also, write down the word(s). Uncertain words 5. Uncertain words indicate whether the news artic le uses any words to describe NDM 1 related issues as uncertain or unknown, such as not sure,
7 9 unsure, unknown, questionable. undetermined, remains to be determined, remains to be seen. 6. Placement of Uncertain words if the news article uses uncertain word s, indicate where the word(s) is placed: 1) headline; 2) lead paragraph; 3) story body. Also, write down the word(s). Risk magnitude information about human infection /death 7. Information of the human infection/death I f the news article does not provide any information about human infection/death (e.g., chances of infection, number of infected persons), circle 0). I f the news article provide s qualitative/non numerical information about human infection/death circle 1) I f the news article provides quantitati ve information about human infection/death at numerator level (e.g., "there have been 50 cases identified in the UK"), circle 2). I f the news article provides quantitative information about human infection/death at numerator/population level (e.g., "at lea st 3% of people infected"), circle 3). Risk magnitude information about financial loss to society 8. Information about financial loss to society I f the news article provide s qualitative information about financial loss to society caused by NDM 1 (e.g., "suf fered a deadly blow"), circle 1) I f the news article provides quantitative information about financial loss (e.g., "dropped by 30%"), circle 2). I f the news article does not provide any information about financial loss to society, circle 0). Personal prot ection information 9. Personal protection information indicate whether the news article include any personal protection information to decrease personal risk of NDM 1 (e.g.,
80 wash hands, use antibacterial surface wipes, etc.) Societal protection information 10. Societal protection information indicate whether the news article include any actions taken by domestic and/or foreign governments or international organizations to prevent NDM 1. I nformation about risk comparison to known risks 11. Risk comparison indicat e whether the news story mention about any other superbugs which are known (e.g., MRSA, MSSA, C Difficile, etc.). I f the news article mentions one type of other superbugs, circle 1) I f the news article mentions two or more types of other superbugs, circle 2). If none, circle 3). 12. Information of known risks I f the news article does not provide any information about other similar risks (e.g., chances of infection, number of infected persons), circle 0) I f the news article provide s qualitative/non numerical information about other similar risks, circle 1). I f the news article provides quantitative information about human infection/death at numerator level (e.g., "there have been 50 cases identified in the UK"), circle 2). I f the news article provides quantit ative information about human infection/death at numerator/population level (e.g., "at least 3% of people infected"), circle 3). 13. Placement of similar risk scenarios if the news article mentions similar risk scenarios, indicate where the information is p laced: 1) headline; 2) lead paragraph; 3) story body. 14. R isk comparison to other countries I f the news article provides similar risk scenarios in other regions and/or country(countries), circle 1) and record the name of other regions and/or country(countr ies) appear in the article. If not,
81 circle 2).
82 A PPENDIX C Item Q1. Presence of worst case scenarios 0.85 Q2. Placement of worst case scenarios 0.81 Q3. Presence of loaded words 0.93 Q4. Placement of loaded words 0.95 Q5. Use of words reflecting uncertainty 0.9 Q6. Placement of uncertain words 0.98 Q7. Information of risks of human infection/death 0.95 Q8. Risk information of financial loss 0.85 Q9. Personal protection information 0.95 Q10. Societal prot ection information 0.85 Q11. Information comparison NDM 1 to known risks information 0.83 Q12. Information comparison NDM 1 to known risks information --information type 0.93 Q13. Placement of information comparing NDM 1 to known risks 0.81 Q14. Info rmation comparing risks in other countries 0.95 Overall 0.9
83 A PPENDIX D LOADED WORDS Alarm, alarming, alert, danger dangerous, deadly, deadliest fatal, fear, frightening, horror, impossible to treat, kill, killer, lethal, life threatening (life threat ening), panic, scare, scary, serious, severe, threat, threatening, trouble, troublesome, unmanageable, unprecedented, unstoppable, untreatable, warn, warning, worry, worrying(ly), worrisome, worst
84 A PPENDIX E UNCERTAIN WORDS Do not know, further studies se em necessary, impossible to say, "It has to be seen ...", little data, need to be confirmed, no consensus on, no conclusions, no evidence, no records, not sure, there can't be any assumptions made, too early to judge, uncertain, uncertainty, unclear, unkno wn, unpredictable
85 REFERENCE LIST About us (n.d.) Medical Journalists' Association (MJA). London, UK. Retrieved April 1, 2012, from http://www.mjauk.org/Contact Us A.D.A.M. Medical Encyclopedia (2010, January 14). Septic shock Atlanta, GA : American Accred itation HealthCare Commission (A.D.A.M., Inc.) Retrieved June 21, 2012 from http://www.nlm.nih.gov/medlineplus/ency/article/000668.htm Advanced Post Graduate Diploma in Scientific Writing and Medical Journalism (n.d.). James Lind Institute. Andhra Prade sh, India. Retrieved April 2, 2012, from http://www.jli.edu.in/diploma_sw_mj.php Ahmad, A., Krumkamp, R. & Reintjes, R. (2009). Controlling SARS: a review on affected countries. Tropical Medicine & Internationa l Health 14: 36 45. doi: 10.1111/j.1365 3156.2008.02146.x American Medical Association. (1999). Health literacy: report of the Council on scientific a ffairs Ad Hoc Committee on Health Literacy for the Council on Scientific Affairs, AMA. JAMA, 281(6), 55 2 557. Arul A. & Suresh P. (2009). Challenges facing media e ducation in India. Media Education in Asia 2009, 121 130, DOI: 10.1007/978 1 4020 9529 0_9 Bennett, N.J. (2012, May 3). Bacteremia Retrieved June 21, 2012, from http://emedicine.medscape.com/article/961169 overview Bloom, D. E., & Canning, D. (2008). Population health and economic growth Washington: The International Bank for Reconstruction and Development / The World Bank. Retrieved June 21, 2012 from http://www.growthcommission.org/storage/cgdev/documents/gcwp024web.pdf Bomlitz, L. J., & Brezis, M. (2008). Misrepresentation of healt h risks by mass media. Journal of Public Health 30 (2): 202 204. doi: 10.1093/pubmed/fdn009 Boyle, S. (2011). United Kingdom (England): Health system review. Health Systems in Transition 2011; 13(1):1 486. Retrieved May 22, 2012 from http://www.euro.who .int/__data/assets/pdf_file/0004/135148/e94836.pdf Bronfman, N. C., Cifuentes, L., deKay, M. L., & Willis H. H. (2007). Accounting for variation in the explanatory power of the psychometric paradigm: The effects of a ggregation and f ocus. Journal of Risk Research 10(4), 527 554. doi:10.1080/13669870701315872 Brown, J.A. (1998). Media literacy perspectives. Journal of Communication March 1998. Volume 48, Issue 1, 44 57.
86 Business Knowledge Resource Online. (n.d.) Service sector: Media. Retrieved Januar y 11, 2012, from http://business.gov.in/Industry_services/media.php Cadge, W., Ecklund, E. & Short, M. (2009). Religion and spirituality: A barrier and a bridge in the everyday professional work of pediatric physicians. Social problems 56(4),702 721. DOI: 10.1525/sp.2009.56.4.702 Census of India. (2011). Provisional p opulation t otals: India: Census 2011 Retrieved February 7, 2012, from http://www.censusindia.gov.in/2011 prov results/indiaatglance.html Central Intelligence Agency. (2011a). The w orld f actbook: I ndia November 15, 2011. McLean, VA : Central Intelligence Agency. Retrieved December 20 2011 from https://www.cia.gov/library/publications/the world factbook/geos/in.html Central Intelligence Agency. (2011b). The world f actbook: United Kingdom November 15, 2011. McLean, VA: Central Intelligence Agency. Retrieved December 20, 2011 from https://www.cia.gov/library/publications/the world factbook/geos/uk.html Central Intelligence Agency. (2011c). The world f actbook: United States. November 17, 2 011. McLean, VA: Central Intelligence Agency. Retrieved December 20, 2011 from https://www.cia.gov/library/publications/the world factbook/geos/us.html Centers for Disease Control and Prevention. (2012, January 19). E. Coli: General i nformation Atlanta, GA : Centers for Disease Control and Prevention Retrieved March 16, 2012 from World Wide Web: http://www.cdc.gov/ecoli/general/index.html/ Choudhury, U. (2010, September15). 3 Americans hit by s uperbug after India treatment; d octors in the US slam Delhi f or overusing antibiotics, saying it could lead to rapid spread of virus among several people. DNA (Daily News & Analysis) Electronic version available at http://www.dnasyndication.com/dna/article/DNMUM185274 Christie, L. (2011, September 13). Number of p eople without health insurance climbs. Retrieved May 22, 2012, from CNNMoney.com website: http://money.cnn.com/2011/09/13/news/economy/census_bureau_health_insu rance/index.htm Combat Antimicrobial Resistance (n.d.). Geneva, Switzerland: World Health Organ ization. Retrieved January 11, 2012, from http://www.who.int/world health day/2011/WHD201_FS_EN.pdf Courtney, L. (2004). Health risk reporting. Society. 11, Volume 42, Issue 1, 49 56.
87 Cut ler, D.M. (2008). The American healthcare s ystem. Me dical S oluti ons May 2008. Retrieved December 20, 2011 from http://www.economics.harvard.edu/faculty/cutler/files/The%20American%20H ealthcare%20System.pdf Den tzer, S. (2009). Communicating medical news pitfalls of health care j ournalism. The New England Journal of Medicine 2009; 360:1 3. Retrieved April 2, 2012, from http://www.nejm.org/doi/full/10.1056/NEJMp0805753 Department of Health and Human Services. (n.d.). Healthcare associated i nfection (HAI) Washington, D.C: U.S. Department of Health & Human Services. R etrieved June 21, 2012, from http://www.hhs.gov/ash/initiatives/hai/index.html Department of Health and Human Services (2008). America's h ealth l iteracy: Why w e n eed a ccessible h ealth i nforma tion. Retrieved January 5, 2012, from http://www.health.gov/communication/literacy/issuebrief/ DeSilva, M., Muskavitch, M., & Roche, J. (2004). Print media coverage of antibiotic resisitance. Science Communication 26(1): 31 43. Djankov, S., McLiesh, C., Nenova, T., & Shleifer, A. (2001). Who Owns the Media? Retrieved June 22, 2012 from http://siteresources.worldbank.org/INTWDRS/Resources/477365 1257315 064 764/2423_djankov.pdf Dudo, A., Dahlstrom, M., & Brossard, D. (2007). Reporting on a potential pandemic: A content analysis of avian influenza newspaper coverage. Science Communication 28, 429 454. Eckersley R. (2006). Is modern Western culture a h ealth hazard ? International Journal of Epidemiology 2006; 35: 252 258. E.coli death toll increases to 50 (2011, July 1). Retrieved March 16, 2012, from World Wide Web: http://yourlife.usatoday.com/fitness food/safety/story/2011/07/Ecoli death toll i ncre ases to 50/49023846/1 Ellison, C.G., & Levin, J.S. (1998). T he religion health connection: E vidence, theory, and future directions. Health Education & Behavior. Vol. 25 (6): 700 720. Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S. & Combs, B. (1 978). How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits Policy Sciences 9(2), 127 152. Fung, T. K., Namk oong, K. & Brossard, D. (2011). Media, social proximity, and risk: A comparative analysis of newspaper coverage of avian f lu in Hong Kong and in the United States, Journal of Health Communication 16:8, 889 907
88 Gargano, L.M., Schreier, B., & Hughes, J. M. (2010). Factors influencing emerging infectious diseases in the southeastern United States [conference summary]. Emerging Infectious Diseases [serial on the Internet]. Retrieved June 22, 2012 from http://wwwnc.cdc.gov/eid/article/16/4/09 1910.htm Glik, D. C. (2003). Health c ommunication in p opular m edia f ormats American Public Health Association Annual Meeting presentation. San Francisco, CA. Retrieved April 6, 2012, from Medscape Public Health website: http://www.medscape.org/viewarticle/466709 Golik eri, P. (2011, April 7). Indiscriminate use, wrong dose making bugs resistant to these drugs: Health experts, doctors. Daily News & Analysis Retrieved from http://www.searo.who.int/LinkFiles/WHD 11_DNA.pdf Hay, J., Coups, E.J., Ford, J., & DiBonaventura, M. (2009). Exposure to mass media health information, skin cancer beliefs, and sun protection behaviors in a U.S. probability sample. Journal of the American Academy of Dermatology. 2009; 61 (5): 783 92. Health Protection Agency. (2011, mid March). Epidem iological data (carbapenem resistance and NDM 1). London: Health Protection Agency. Retrieved April 2, 2012, from http://www.hpa.org.uk/Topics/InfectiousDiseases/InfectionsAZ/CarbapenemRe sistance/EpidemiologicalData/. Health Reporter Training Summer 2012 (2012, March 11). Athens, Greece : Health Reporter Training (HeaRT) Project. Retrieved April 1, 2012, from http://www.healthemergency.org.uk/pdf/HeaRT_Leaflet.pdf Hesse, B. W., Nelson, D. E., Kreps, G. L., Croyle, R. T., Arora N. K., Rimer, B. K., & Viswa nath, K. (2005). Trust and sources of health information: the impact of the Internet and its implications for health care providers: findings from the first Health Information National Trends Survey. Archives of Internal Medicine 165 (22):2618 24. doi: 10. 1001/archinte.165.22.2618. 165/22/2618 HinduMinistries. Org. (n.d.). Comparison of cZ hristian and Hindu b eliefs. Seffner, FL : HinduMinistries. Org Retrieved May 14, 2012, from http://www.hinduministries.org/downloads/Comparison_Of_Beliefs.pdf History of AHCJ (n.d.). Columbia, MO : Association of Health Care Journalists Retrieved April 2, 2012, from http://www.healthjournalism.org/about history.php Hofstede, G. (n.d.). India Retrieved April 26, 2012, from http://geert hofstede.com/india.html Hofstede G. (n.d.). United Kingdom Retrieved April 26, 2012, from http://geert hofstede.com/united kingdom.html
89 Hofstede, G. (n.d.). United States Retrieved April 26, 2012, from http://geert hofstede.com/united states.html International Journalists' network. (2006). Health reporting course open to mid career journalists Washington, DC: Inter national Center for Journalists Retrieved April 2, 2012, from http://ijnet.org/opportunities/health reporting course open mid career journalis ts Johnson, B. B. & Slo vic, P. (1995). Presenting u ncertainty in h ealth r isk a ssessment: Initial s tudies of i ts e ffects on r isk p erception and t rust. Risk Analysis 15: 485 494. doi: 10.1111/j.1539 6924.1995.tb00341.x Johnson, T., D. (2006). Graduate program brings public healt h, journalism together. The Nation's Health. October 2006, Vol.36, 18, p. 12(1). Retrieved April 1, 2012, from the Gale Health & Wellness Resource Center database. Record Number: A154331959. Kalaitzandonakes, N., Marks, L.A., & Vickner, S.S. (2004). Me dia coverage of biotech foods, and influence on consumer choice. American Journal of Agricultural Economics 86(5): 1238 46. Koenig, H.G. (2007). Spirituality in p atient c are: Why, h ow, w hen, and w hat Philadelp hia: Templeton Foundation Press. 2007. pp.108 122. Kreeft, P. (1987, May). Comparing Christianity & Hinduism. National Catholic Register Retrieved May 14, 2012, from http://www.catholiceducation.org/articles/apologet ics/ap0008.html. Kreuter, M. W., & McClure, S. M. (2004). The role of culture in health communication. Annual Review of Public Health 25, 439 455. Livingstone, S. (2003). Media literacy and the challenge of new information and communication technologies Communication review 1 (7). 3 14. Online version available at: http://eprints.lse.ac.uk/1017 Livingstone, S., & Thumim, N. (2003). Assessing the media literacy of UK adults: A review of the academic literature Report commissioned by the Broadcastin g Standards Commission/ Independent Television Commission/National Institute of Adult and Continuing Education. March, 2003. Retrieved February 08, 2012, from http://eprints.lse.ac.uk/21673/1/Assessing_the_media_literacy_of_UK_adults. pdf Livingstone, S. ( 2004). What is media literacy? Intermedia, 32 (3). pp. 18 20. Liu, Y., Liu, M., Xiao, H., Cai, J., & Xu, W. (2010). A c ontent a nalysis of n ews c overage of s kin c ancer in China n ewspapers. Health Communication 25 (8), 647 649. doi:10.1080/10410236.2010.5 21904
90 Long, E. (2011, May 12). How the latest Pew report points to link between health literacy and media literacy Retrieved April 6, 2012, from http://www.thelampnyc.org/2011/05/12/how the latest pew report points to lin k between health literacy and med ia literacy/ MacDonald, M.F. (2008). Media literacy in action: An exploration of teaching and using media literacy constructs in daily classroom practice. (Doctoral dissertation). Retrieved from Proquest Dissertations & Theses database. University of Cali fornia, Los Angeles, 2008, 281 pages; AAT 3332575. Manocha S. (2010). The top most widely read newspapers in India Retrieved June 26, 2012, from http://www.ch illibreeze.com/articles/top 10 newspapers in India 1011.asp Medical j ournalism: Separating f acts f rom f iction. (12 Aug 2011). James Lind Institute. Andhra Pradesh, India. Retrieved April 2, 2012, from http://www.jli.edu.in/blog/2011/08/medical journalism separating facts from fict ion/ Medical journalism at its best. (n.d.). Medical Journalists' Association (MJA). London, UK. Retrieved April 1, 2012, from http://www.mjauk.org/ Mehr, S.R. (2008). Report on the h ealth s ystem of India Retrieved May 28, 2012 from TPG International Health Academy website: http://www.tpg iha.com/pdf/AIHSIndiaReportFinal.pdf Meissner, H. I., Potosky, A. L., & Convissor, R. (1992). How sources of health informa tion relate to knowledge about use of cancer screening exams. Journal of Community Health 17, 153 165. MIHR report to CIPIH (2005 April). Innovation in developing countries to meet health needs: Experiences of China, Brazil, South Africa and India. WHO ref. CIPIH study 10d. p.112. Retrieved March 20, 2012, from http://www.who.int/intellectualproperty/studies/MIHR INNOVATION%20EXPE RIENCES%20OF%20South%20Africa,%20CHINA,%20BRAZIL%20AND%20I NDIA%20MIHR CIPIH%20REPORTS%2014 04 05.pdf Miller, W.R., & Thore sen, C.E. (2003). Spirituality, religion, and health an emerging research field. American Psychologist. 2003; 58: 24 35. Moellering R.C. (2010). NDM 1 A c ause for w orldwide c oncern. The New England Journal of Medicine 2010; 363: 2377 2379 Retrieved f rom http://www.nejm.org/doi/full/10.1056/NEJMp1011715#t=article National Center for Education Statistics. (2003). Literacy skills Retrieved February 07, 2012, from http://nces.ed.gov/naal/fr_skills.asp
91 Ofcom (the Office of Communications). (2011). UK Adul ts' Media Literacy April 2011. Retrieved March 14, 2012, from http://stakeholders.ofcom.org.uk/binaries/research/media literacy/media lit11/ Adults.pdf Office for National Statistics. (2001). United Kingdom c ensus 2001 Religion Newport, South Welsh: Uni ted Kingdom. Retrieved May 14, 2012, from http://www.free enterprises.co.uk/Religion Statistics/Office of National Statisti cs/Census 2001/ Office of The Registrar General & Census Commissioner. (2001). Census d ata 2001 India at a glance Religious c ompo sition Office of t he Registrar General & Census Commissioner, India, New Delhi. Retrieved May 14, 2012, from http://censusindia.gov.in/Census_Data_2001/India_at_glance/r eligion.aspx Paasche Orlow, M.K., Parker, R.M., Gazmararian, J.A., Nielsen Bohlman, L.T., & Rudd, R.R. (2005). The prevalence of limited health literacy. Journal of General Internal Medicine 2005;20:175 84. doi: 10.1111/j.1525 1497.2005.40245.x Pappas, G., Siozopoulou, V., Saplaoura, K., Vasiliou, A., Christou, L., Akritidis, V., & Tsianos, E. V. (2007). Health literacy in the field of infectious diseases: The paradigm of brucellosis. Journal of Infection Volume 54, Issue 1, Pages 40 45. Park, K. (2007) Preventive and social medicine. 19th edition (2007), M/s Banarsidas Bhanot, Jabalpur, India. Parker, J. C., Woelfel, R. L., Hart, E. S. & Brown, G (2009). The c hallenge of h ealth c are and d isability. Chapter 1 in Parker, J. C. and Thorson, E., (Eds.), Hea lth communication in the new media landscape New York, NY: Springer Publishing Company. Parker, R. M., Ratzan, S. C., & Lurie, N. (2003). Health literacy: A policy challenge for advancing high quality health care. Health Affairs, 22(4), 147 153. Persell, S.D., Osborn, C.Y., Richard, R., Skripkauskas, S., & Wolf, M.S. (2007). Limited health literacy is a barrier to medication reconciliation in ambulatory care. Journal of General Internal Medicine 2007; 22(11): 1523 1526. Phillips, M. (2010 March 23). Wha t's in the h ealth c are b ill? Retrieved May 26, 2012, from The White House Blog: http://www.whitehouse.gov/blog/2010/03/23/whats health care bill Potter, W. (2010). The s tate of m edia l iteracy. Journal o f Broadcasting & Electronic Media 54(4), 675 696. doi :10.1080/08838151.2011.521462 Professional Diploma in Medical Journalism. (n.d.). James Lind Institute. Andhra Pradesh, India. Retrieved April 2, 2012, from http://www.jli.edu.in/certificate_medical_journalism.php
92 Rao, S. (2005). Health Insurance in India. In Pranay, G. Lal, Byword Editorial Consultants, (eds). NCMH: Background papers on Financing and delivery of Health Care Services in India. New Delhi: Cirrus Graphics Private Limited; 2005 August. pp. 275 95. Retrieved May 22, 2011 from http://www.who.int/macrohealth/action/Background%20Papers%20report.pdf Renn, O., & Rohrmann, B. (2000). Cross cultural risk perception: A survey of empirical studies Boston: Kluwer Acade mic Publishers.. Reporters Without Borders. (2012 January 25). A Press Freedom Index 2011 2012. Paris, France: Reporters Without Borders. Retrieved May 22, 2012, from http://en.rsf.org/IMG/CLASSEMENT_2012/C_GENERAL_ANG.pdf Roe, A.M., & Liberman, A. (20 07). A comparative analysis of the United Kingdom and the United States health care systems. The Health Care Manager (Frederick). 2007; 26: 190 212. Rohrmann, B. & Renn, O. (2000). Risk p erception r esearch An i ntroduction. In Renn O. & Rorhmann, B. (E ds.), Cross cultural risk perception: A survey of empirical studies (pp. 11 53). Dordrecht: Kluwer Academic Publishers. Rubin, A.M. (1998). Media literacy. Journal of Communication. March 1998, Volume 48, Issue 1, 3 4. Scheufele, D. A. (1999). Framing as a theory of media effects. Journal of Communication 49: 103 122. Schwarz, G. (2005). Overview: What is media literacy, who cares, and why? In G. Schwarz & P. Brown (Eds.), Media literacy: Transforming curriculum and teaching (pp. 5 17). Maiden, MA: Blackwe ll Publishing. Schweiger, W. (2000) Media c redibility Experience or i mage? A s urvey on the c redibility of the w orld w ide w eb in Germany in c omparison to o ther m edia. European Journal of Communication 15(1): 37 59. Schwitzer, G. (2009 ) The s tate of h ealth j ournalism in the U.S. A Report to the Kaiser Family Foundation. Menlo Park, California: He nry J. Kaiser Family Foundation Retrieved April 2, 2012, from http://www.who.int/pmnch/topics/advocacy/healthjournalism/en/index.html Science j ournalism : Career p rospects (n.d.). City University London. London, United Kingdom. Retrieved April 1, 2012, from http://www.city.ac.uk/courses/postgraduate/science journalism Science j ournalism: Overview. (n.d.). City University London. London, United Kingdom. Re trieved April 1, 2012, from http://www.city.ac.uk/courses/postgraduate/science journalism Shiell A. & Hawe P. (1996) Health promotion, community development and the tyranny of individualism. Health Economics 5, 241 247.
93 Shih, T. J., Brossard, D., & Wijaya, R. (2011). News coverage of public health issues: The role of news sources and the processes of news construction. International Public Health Journal 3(1), 87 97. Silverblatt, A., & Enright Eliceiri, E.M. (1997). Dictionary of media literacy Westport, CT: Greenwood Press. Sloan, R. P., Bagiella, E. and Powell, T. (1999). Religion, s pirituality, and m edicine. The Lancet .353 (1999): 664 667. Slovic, P. (1987). Perception of risk. Science Washington, DC. v. 236, n. 4799, p. 280 285. Slovic, P. (2000). The p erception of r isk. Earthscan, London. Slovic, P., Fischhoff, B., & Lichtenstein, S. (1980). Facts and f ears: Understanding p erceived r isk. In Albers, W.A. (Ed.), Societal Risk Assessment: How Safe Is Safe Enough? New York, NY: Plenum Press. Sl ovic, P., Fischhoff, B., & Lichtenstein, S. (1982). Why s tudy r isk p erception? Risk Analysis 2(2): 83 93. Slovic, P., Fischhoff, B. & Lichetenstein, S. (1985). Characterizing p erceived r isk I n R. W. Kates, C. Hohenemser C. & Kasperson J. X. (Eds.), Peri lous p rogress: Managing the h azards of t echnology ( pp. 91 123). Boulder, CO: Westview. Slovic, P., Fischhoff, B., & Lichtenstein, S. (2000). Facts and fears: Understanding perceived risk. In Slovic, P. (Eds.), The Perception of Risk (pp.137 153). Earthsca n, London. Sjberg, L., Kolarova, D., Rucai, A. A., & Bernstrm, M. L. (2000). Risk p erception in Bulgaria and Romania. In Renn, O., & Rohrmann, B. (Eds), Cross cultural risk perception: A survey of empirical studies. Cross cultural risk perception: A surv ey of empirical studies (pp. 145 184). Boston: Kluwer. Sosa, A. (2005). Antibiotic policies in developing countries. In Gould, I.M., Van der Meer J.W., (Eds). Antibiotic policies, theory and practice (pp. 593 616). New York: Kluwer Academic/Plenum Publishe rs. Speros, C. (2005). Health literacy: Concept analysis. Journal of Advanced Nursing 50, 633 640. Stableford, S., & Mettger, W. (2007). Plain language: A strategic response to the health literacy challenge. Journal of Public Health, 28, 71 93. doi:10.10 57/palgrave.jphp.3200102 Stryker, J. (2003). Media and m arijuana: A l ongitudinal a nalysis of n ews m edia e ffects on a dolescents' m arijuana u se and r elated o utcomes, 1977 1999. Journal o f Health Communication 8 (4), 305 328.
94 Sudore, R.L., Yaffe, K., Satterfi eld, S., Mehta, K.M., Simonsick, E.M., Newman, A., Rosano, C., Rooks, R., Rubin, S.M., Ayonayon, H.N., & Schillinger, N. (2006). Limited literacy and mortality in the elderly: the health, aging, and body composition study. Journal of General Internal Medic ine 2006; 21(8): 806 12. The Center for Disease Dynamics, Economics & Policy (CDDEP). (2007). Extending the c ure: Policy r esponses to the g rowing t hreat of a ntibiotic r esistance Retrieved March 10, 2012, from http://www.cddep.org/sites/cddep.org/files/pu blication_files/ETC_FULL.pdf?iss uusl=ignore The concept of the defined daily dose (2003). Geneva, Switzerland: World Health Organization. Retrieved March 20, 2012, from http://apps.who.int/medicinedocs/en/d/Js4876e/7.html#Js4876e.7.1 The Newsbank. (n.d.). The v alue of n ewspaper i nformation Naples, FL: Newsbank, inc. Retrieved March 11, 2012, from http://www.newsbank.com/index.cfm?content=172 The Pew Forum on Religion & Public Life. (2008, February). U.S. r eligious l andscape s urvey Washington, DC. Retrieve d May 14, 2012, from http://religions.pewforum.org/pdf/report religious landscape study full.pdf The Queensland Government (Australia). (2011). Health c are p roviders' h andbook on Hindu p atients Retrieved June 11, 2012 from the Queensland Government websi te: http://www.health.qld.gov.au/multicultural/support_tools/hbook hindu.pdf University of California Berkeley. (n.d.). Concurrent degrees. Master of p ublic h ealth (M.P.H.), Master of Journalism (M.J.). Retrieved April 2, 2012, from http://sph.berkeley.edu /students/degrees/programs/mj mph.php University of Minnesota. (n.d.). M.A. in h ealth j ournalism & c ommunication. Retrieved April 2, 2012, from http://sjmc.umn.edu/grad/hjComm.html UK Department for Business, Innovation and Skills. (2011). 2011 s kills for l ife s urvey: Headline findings Retrieved January 11, 2012, from http://www.bis.gov.uk/assets/biscore/further education skills/docs/0 9/11 1367 2011 skills for life survey findings.pdf UK Department of Health. (2009). Literacy s kills l evels in England and the i mpact on h ealth Retrieved January 11, 2012, from http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/di gitalasset/dh_095375.pdf
95 U.S. Department of Health and Human Services. (2000). Healthy People 2010. Washington, DC: U.S. Governme nt Printing Office. Originally developed for Ratzan SC, Parker RM. 2000. Introduction. In Selden, C. R., Zorn, M., Ratzan, S.C., & Parker, R. M., (Eds), National l ibrary of m edicine c urrent b ibliographies in m edicine: Health l iteracy. NLM Pub. No. CBM 2000 1. Bethesda, MD: National Institutes of Health, U.S. Department of Health and Human Services. U.S. Department of Health and Human Services. (n.d.). Immunization and i nfectious d iseases Retrieved March 16, 2012, from http://www.healthypeople.gov/2020/topi csobjectives2020/overview.aspx?topici d=23 Walsh, T. R., Weeks, J., Livermore, D. M., & Toleman, M. A. (2011). Dissemination of NDM 1 positive bacteria in the New Delhi environment and its implications for human health: A n environmental point prevalence stu dy. The Lancet Infectious Diseases. May 2011,Volume 11, Issue 5, p. 355 362. doi:10.1016/S1473 3099(11)70059 7. Wenham, K., Harris, N., & Sebar, B. (2009). Reflections of inequalities: The construction of HIV/AIDS in Africa in the Australian print media. Health Sociology Review 18(3), 284 296. Whitman, S. M. (2007). Pain and suffering as viewed by the Hindu religion. The Journal of Pain Vol. 8, No. 8 (August), 2007. pp 607 613. WHO Collaborating Centre for Drug Statistics Methodology. (2009, December 17 ). Definition and general considerations Oslo, Norway: WHO Collaborating Centre for Drug Statistics Methodology. Retrieved March 20, 2012, from http://www.whocc.no/ddd/definition_and_general_considera/ World Health Organization (WHO). (n.d.) Outbreaks of E. coli O104:H4 infection. Copenhagen, Denmark: WHO Europe. Retrieved March 16, 2012, from http://www.euro.who.int/en/what we do/health topics/emergencies/internation al health regulations/outbreaks of e. coli o104h4 infection World Health Organization. (1 998). Health p romotion g lossary Geneva, Switzerland: World Health Organization. Retrieved March 16, 2012, from http://www.who.int/healthpromotion/about/HPR%20Glossary%201998.pdf World Health Organization. (2006a). Country c ooperation s trategy: India Retr ieved May 22, 2011 from http://www.who.int/countryfocus/cooperation_strategy/ccsbrief_ind_en.pdf World Health Organization. (2006b). WHO c ountry c ooperation s trategy 2006 2011: India. Retrieved December 22, 2011 from http://www.who.int/countryfocus/coopera tion_strategy/ccs_ind_en.pdf
96 World Health Organization. (2010a). Antimicrobial resistance: R Bulletin of the World Health Organization 2010 Volume 88, Number 11, November 2010, 797 876. Geneva, Switzerland: World Hea lth Organization. Retrieved from http://www.who.int/bulletin/volumes/88/11/10 031110.pdf World Health Organization. (2010b). Better health for Europe: A dapting the WHO Regional Office for Europe to the changing environment Regional Director's speech at th e sixtieth session of the WHO Regional Committee for Europe. Retrieved January 11, 2012, from http://www.euro.who.int/en/who we are/regional director/speeches and prese ntations by year/2010/better health for europe adapting the who regional offi ce for euro pe to the changing environment World Health Organization. (2010c). Containing antimicrobial resistance: A renewed effort. Bulletin of the World Health Organization, Geneva, Switzerland: World Health Organization. 2010; 88: 878 878. doi: 10.2471/BLT.10.08423 6 World Health Organization. (2011a). Antimicrobial resistance. Media centre [Fact sheet] Geneva, Switzerland: World Health Organization. Retrieved January 11, 2012, from http://www.who.int/mediacentre/factsheets/fs194/en/ World Health Organization. (2011 b). Antimicrobial resistance: N o action today, no cure tomorrow. Geneva, Switzerland: World Health Organization. Retrieved January 11, 2012, from http://www.who.int/world health day/2011/en/index.html World Health Organization. (2011c). Complacency kills. Antibiotic resistance still on the rise in Europe Copenhagen, Denmark: WHO Europe. Retrieved March 14, 2012, from http://www.euro.who.int/en/what we publish/information for the media/section s/latest press releases/complacency kills. antibiotic resistance still on the rise in europe World Health Organization. (2011d). tackles antibiotic resistance. Copenhagen, Denmark: WHO Europe. Retrieved January 11, 2012, from http://www.euro.who.int/en/what we publish/informatio n for the media/section s/latest press releases/europe fights back whos new action plan tackles anti biotic resistance World Health Organization. (2011e). Mobilizing political will to contain antimicrobial resistance. Bulletin of the World Health Organizatio n Geneva, Switzerland: World Health Organization. 2011; 89: 168 169. doi:10.2471/BLT.11.030311 World Health Organization. (2011f). Race against time to develop new antibiotics. Bulletin of the World Health Organization 2011 Geneva, Switzerland: World Heal th Organization. 89: 88 89. doi:10.2471/BLT.11.030211
97 World health statistics 2011 (2011). Geneva Switzerland: World Health Organization. ISBN 978 92 4 156419 9. Retrieved May 22, 2011 from http://www.who.int/gho/publications/world_health_statistics/EN_W HS2011_Ful l.pdf World Information Access. (2008 July 10). Ownership d iversity in Muslim m edia s ystems Retrieved May 16, 2012 from http://www.wiaproject.org/index.php/68/ownership diveristy in muslim media s ystems WHO Regional Office for South East Asia. ( 2011, April 7). Media reports. Retrieved April 16, 2012 from http://www.searo.who.int/en/Section260/Section2659_15911.htm Yanovitzky, I., & Blitz, C. L. (2000). Effect of m edia c overage and p hysician a dvice on u tilization of b reast c ancer s creening by w ome n 40 y ears and o lder. Journal Of Health Communication 5 (2), 117 134. doi:10.1080/108107300406857
98 BIOGRAPHICAL SKETCH Bijie Bie enrolled in the Department of Advertising of the College of Journalism and Communications at the University of Florida in Aug ust 2010. In China, she completed her undergraduate study majoring in human resources management and minoring in advertising. After she received her M aster of Arts at the University of Florida, she will join the doctoral program in the College of Communica tion and Information Sciences at the University of Alabama starting in August 2012.