Citation
The Possible Impacts of Alcohol Consumption on Suicide Rates in the United States

Material Information

Title:
The Possible Impacts of Alcohol Consumption on Suicide Rates in the United States
Creator:
Moscone, John R.
Publication Date:
Language:
English

Notes

Abstract:
Notable sources have indicated that suicide rates in the United States have increased significantly in recent years. Alongside this rise has been the increase of alcohol consumption in the United States. In this paper, suicide rates are studied as a function of alcohol consumption per capita and several socioeconomic variables including, but not limited to, average household income and inequality measures. The results show that alcohol consumption per capita, as well as several of the included variables, has a statistically significant effect on suicide rates in the United States. It is concluded that a better understanding of substance use and abuse, as well as social factors, is necessary in order to curb the rising rate of suicide. ( en )
General Note:
Awarded Bachelor of Arts, summa cum laude, on May 8, 2018. Major: Economics
General Note:
College or School: College of Liberal Arts and Sciences
General Note:
Advisor: Michelle A. Phillips. Advisor Department or School: Economics Department (CLAS)

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright John R. Moscone. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

UFDC Membership

Aggregations:
UF Undergraduate Honors Theses

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

! 1 The Possible Impacts of Alcohol Consumption on Suicide Rates in the United States John R. Moscone University of Florida April 2018

PAGE 2

! 2 I. INTRODUCTION The suicide epidemic has been and continues to be a growing issue in the United S tates. In 2015, the average suicide rate in the U.S. was 13.3 deaths per 100,000 individuals, yet this number varies between genders and across states 1 Suicide is the 10 th leading cause of preventable death in the U S summing up to an average of 121 dea ths per day and even more attempts 2 At the same time, alcohol consumption in the U S has been rising, with over 50% of people ages 18 and older reporting that they have consumed alcohol in the past month 3 While it is known that the excessive use and abu se of alcohol has many deleterious health effects could growing alcohol consumption be playing a role in Am erica's increasing suicide rate ? In this paper, suicide rates will be studied as a function of alcohol consumption per capita and several socioeconomi c variables. The goal is to find out whether or not alcohol consumption has a positive and statistically significant effect on suicide rates in the U S through regression analysis Beyond the scope of this study are the many personal and psychological rea sons why one might take their own life For the purpose of this paper the determinants of suicide will be observed from a purely socioeconomic perspective using variables that have been proven to have an affect on the rate of suicide While causality is not claimed in r egard to the relationship between alcohol consumption and suicide rates, t he results could lead to a better understanding of the many possible determinants of suicide. This, in turn could motivat e changes in policy that may reduce suicide rate s in the U S and worldwide. 1 Centers for Disease Control and Prevention National Center for Health Statistics (2015) 2 American Foundation for Suicide Prevention (2015) 3 National Institut e on Alcohol Abuse and Alcoholism (2017)

PAGE 3

! 3 II. SAMPLE The sample for this study is the adult population in each of the fifty states in the U S in 2014 and 2015 Washington D C and U.S. territories were excluded from the study since several values for these areas prove d to be outl iers. 2014 and 2015 were chosen as the year s of study because both suicide rat es and alcohol consumption were reported to have reac hed extraordinarily high levels during these years. III. DEPENDENT VARIABLE Suicide Rate (SUI RATE) The dependent var iable measures the adult suicide mortality rate in the U.S This variable measures the number of completed suicides per 100,000 individuals in each of the fifty states in 2014 and 2015 An independent variable will contro l for gender distribution, as ma le and female suicide rates differ significantly. The s e data were obtained from the Centers for Disease Control and Prevention's National Center for Health Statistics 1 IV. INDEPENDENT VARIABLES Alcohol Consumption Per Capita by Gallons of Ethanol (ALC) T his variable measures alcohol consumption per capita by gallons of ethanol consumed in the U.S. in 2014 and 2015. Overall, alcohol consumption in the U S has been rising since the 199 0s. While it can be argued that alcohol in small doses may relie ve stres s and anxiety, heavy drinking and alcoholism are known symptoms of

PAGE 4

! 4 depression and depression may lead to suicidal tendencies Seeing that alcohol it self is a depressant, heavy drinking intensifies the negative emotions associated with depression. Furtherm ore, alcohol impairs judg ment and causes users to act impulsively while under the influence. Studies have shown that those who abuse alcohol have higher rates of both attempted and completed suicide than non abusers 4 Therefore, it is hypothesize d that the regression will show a positive and statistically significant correlation between alcohol consumption and suicide rate s Data for this variable were obtained from the National Institute on Alcohol Abuse and Alcoholism 5 Percent Male Population (PERC MALE ) This variable measur es the percentage of the adult population in each state that was reported to be male in 2014 and 2015 This is an important variable to control for since males are 4 times more likely to commit suicide than females; however, females a ttempt suicide 3 times as often as males 2 Seeing that males commit suicide more frequently than females, it is predict ed that the regression will show that states with a larger male population will observe higher rate s of suicide ; in other words, it is expected that there will be a positive and statistically significant relationship between a large male population and suicide rates Data were obtained from The Henry J. Kaiser Family Foundation 6 Unemployment Rate (UNEMP_ RATE ) 4 Montgomery County Emergency Service (n.d.) 5 National Institute on Alcoh ol Abuse and Alcoholism (2016). 6 The Henry J. Kaiser Family Foundation

PAGE 5

! 5 The unemployment rate measures the percentage of the labor force that is out of work Th e Bureau of Labor Statistics provides the unemployment rate for each state in the U.S. in 2014 and 2015 7 In a study by Carlos Nordt, a psychiatric research associ a te at the University of Zurich, it was concluded that 45,000 (or 1 in 5) suicides each year world wide are directly caused by unemployment 8 I t is hypothesized that the regression will show a positive and statistically significant relationship between the unemployment rate and suicide rates. A verage Household Income (AVG_INC) The average household income variable measures the median income earned by households in each of the 50 states in the U.S. in 2014 and 2015. Data for average income measu r ed in inflation adjusted dollars 9 were obtained fro m the United States Census Bureau 10 In the aforementioned study conducted by Carlos Nordt 8 it is suggested that income may play a role in one's decision to take their own life. Because earning a smaller income has been shown to increase the likelihood of suicide, this may help explain the apparent correlation between unemployment and suicide rates. Similar to other measures of economic wellbeing, the relationsh ip between unemployment income, and suicide rates suggest that suicide is more prevalen t during tough economic times Theref ore, a negative and statistically significant relationship between average household income and suicide rates is expected. 7 Bureau of Labor Statistics (2015) 8 World Economic Forum (2015) 9 Inflation adjusted based on previous year estimates 10 United States Census Bureau (2015).

PAGE 6

! 6 Gini Coefficient ( GINI ) The Gini coefficient measures the wealth distr ibution of a nation's re sidents and is a widel y used measure of inequality Valu es range between zero and one; z ero indicates perfect equality (everyone has the same income), while one indicates perfect inequality This v ariable will be used to measure the extent of income inequ ality in each state in 2014 and 2015. According to the World Psychiatric Association, income inequality is positively correlated with both suicide rates and mental health issues 11 Therefore, i t is hypothesized that the regression results will indicate a positive and statistically significant relationship between the Gini coefficient and suicide rates Data pertaining to th e Gini coefficient were obtained from World Atlas 12 and America's Health Rankings 13 Divorce Rate (DIV RATE) The divorce rate variable measures the number of divorces per 1,000 population in the U.S. in 2014 and 2015. According to Kposowa (2000) those who are divorced are about three times more likely to c ommit suicide than those who get married and stay married 14 While divorce rates in the U S are currently falling, t his is an impor tant variable to control for since approximately 40% of marriages in the U.S. end in divorce 15 Going off of these findings, it is expected that the regression will show a positive and statistically significant relationship between di vorce rates and suicide rates Because the chose n data set excludes divorce rates for California, Georgia, Hawaii, Indiana, and 11 Science Daily (2011) 12 World Atlas (2017). 13 America's Health Rankings (2015). 14 Kposowa, Augustine J. (2000), "Marital Status and Suicide in the National Lon gitudinal Morality Study". Journal of Epidemiology & Community Health. 15 American Psychological Association (2017)

PAGE 7

! 7 Minne sota, the regression will be run twice; once without the divorce rate variable, and again with the divorce rate variable ex cluding these five states. Data for divorce rates in the U S in 2014 and 2015 were obtained from the Centers for Disease Control and Prevention's National Center for Health Statistics 16 Gun Ownership (GUN OWN ) This variable measures the percentage of the population in ea ch state th at was reported to have own ed at least one firearm in 2014 and 2015 In both 2014 and 2015, approximately 50 % of suicides in the U.S. were commit t ed using firearms. According to Harvard researchers Matthew Miller and Deborah Azrael, suicide rates are higher in states where gun ownership is more prevalent, and lower where gun ownership is less common 17 Whether or not suicide attempts are successful depend largely on the availability of lethal means. Because firearms are highly lethal, suicide attempts carried out with these devices are more likely to be successful. Therefore it is expected that suicide rates will increase with gun ownership rates indicating a positive and statistically significant relationship Data were obtained from a 2015 British Medical Journal study on gun ownership in the U.S. 18 V. SUMMARY STATISTICS Regression One Summary Statistics 16 Centers for Disease Control and Prevention National Center for Health Statistics (2017) 17 Harvard School of Public Health (2008) 18 Kal esan et. al. (2015). "Gun O wnership and Social Gun Culture" British Medical Journal.

PAGE 8

! 8 Variables Mean Standard Deviation Minimun Maximum SUI RATE 15.33 4.21 7.8 28 ALC 2.45 0.53 1.37 4.55 PERC MALE 49.16 0.84 48 51 UNEMP _RATE 5.39 1.21 2.7 7.9 AVG_INC 55371.88 9186.98 39702 76596 GINI 0.46 0.02 0.42 0.51 GUN _OWN 33.09 13.47 5.2 61. 7 Regression Two Summary Statistics Variables Mean Standard Deviation Minimun Maximum SUI RATE 15.62 4.32 7.8 28 ALC 2.47 0.55 1.37 4.55 PERC MALE 49.18 0.86 48 51 UNEMP _RATE 5.39 1.2 2.7 7.9 AVG_INC 54875.88 9123.87 39702 76596 GINI 0.46 0.02 0.42 0.51

PAGE 9

! 9 DIV RATE 3.3 0.7 1.2 5.3 GUN OWN 33.05 13.94 5.2 61.7 VI. REGRESSION RESULTS As previously mentione d, two regressions were run; regression one excludes the divorce rate (DIV RATE) variable, and regression two includes the divorce rate variable, leaving out the five states where data were unavailable (California, Georgia, Hawaii, Indiana, and Minnesota). In regression one, the adjusted R squared is 0.66 ; this indicates that approximately 66 % of the variation in suicide rates can be explained by the variables included in the regression. In regression two, the adjusted R squared value increases to approximate ly 0. 74 In both regressions, the degrees of freedom are 100, and the range of statistically significant t statistics are determined accordingly. Linear equations for both regressions are as follows: Regression One: SUI RATE = 0 + 1 (ALC) + 2 (PERC MALE) + 3 (UNEMP_RATE ) + 4 ( AVG_INC ) + 5 ( GINI ) + 6 (GUN_OWN) + error Regression Two: SUI RATE = 0 + 1 (ALC) + 2 (PERC MALE) + 3 (UNEMP_RATE ) + 4 ( AVG_INC ) + 5 ( GINI ) + 6 ( DIV_RATE ) + 7 (GUN_OWN) + error Alcoho l Consumption Per Capita by Gallons of Ethanol (ALC)

PAGE 10

! 10 In both regressions, al cohol consumption per capita is proven to be statistically significant at the 98% confidence level and 90% confidence level, respectively Therefore, the null hypothesis that there is no correlation between alcohol consumption per capita and su icide rates can be rejected. This is the result that was hypothesized The c o efficients are approximately 1.2 1 in regression one and 0.92 in regression two mean ing that for every 1 gallon increase in per capita alcohol consumption (measured by gallons of ethanol) in a state suic ides per 100,000 population will increase by the number represented by the coefficient While correlation does not prove causation t here are several apparent reasons for this outcome For example, alcohol consumption is known to impair judgement and bring about impulsive behavior T hese characteristics of this substance may cause people with depression and/or suicidal tendencies to attempt suicide while under the in fluence. Furthermore, alcohol is classified as a depress ant and is involved in over 25% of suicides in the U.S. each year 4 According to the results of this study it can be concluded that state s with higher alcohol consumption per capita will observe more detrimental effects that may be caused by this substance such as higher rates of suicide. M any factors go into the determinants of suicide an d based on th is study alcohol consumption per capita appears to be a significant factor. Percent Male Population (PERC MALE) Both regressions indicate that the percentage of male population in a state is statistically significant at the 99 % confidence level The null hypothesis tha t there is no correlation between the percentage of male population and suicide rates can be rejected This is the result that was hypothesized The c oefficient s are approximately 1.34 in

PAGE 11

! 11 regression one and 1. 04 in regression two meaning that for every 1 percentage point increase in male population in a state, suicides per 100,000 population will increase by that amount The result is in line with the fact that males tend to commit suicide more often than females. Unemployment Rate (UNEMP_ RATE ) The unemployment rate variable is not proven to be statistically significant in either regression. Therefore, the null hypothesis that there is no correlation between unemployment rate s and su icide rates can not be rejected This is not the result that was hypothesized A possible reason for this may be d ue to the fact that the regression did not control for the relationship between unemployment and income. For example, an unemployed individual who earned a higher income prior to losing their job would be in a better financial situation than an individual who is unemployed and ear ned a lower income. It is for this reason that previous studies regarding this topic such as one conducted by Dr. Yong Hwan Noh (2009) 19 have concluded in mixed results. A verage Household Income (AVG_INC) The average household income variable is shown to be statistically significant at the 99% confidence level in regression one and at the 90% confidence level in regression two. The null hypothesis tha t ther e is no correlation between average household income and suicide rates can be rejected The c oefficient s are approximately 0.00012 in 19 Yong Hwan Noh (2009). "Does Unemployment I ncrease Suicide Rates? The OECD Panel Evidence". Journal of Economic Psychology

PAGE 12

! 12 regression one and 0. 00006 in regression two m eaning that for every one dollar increase in average household income, suicides per 100,000 population will de crease by the number represente d by the absolute value of the coefficient This result was expected due to the fact that those who earn relatively low incomes have been found to be more vulnerable to suicidal behaviors than their wea l thier counterparts. Gini Coefficient ( GINI ) Both regressions indicate that the G ini coefficient variable is statistically significant at the 99 % confidence level The null hypothesis tha t ther e is no correlation between the Gini coefficient and suicide rates can be rejected However, the regression s show a negative relationship, which is not the result that was hypothesized. Seeing that aforementioned studies have found there to be a positi ve correlation between income inequality and suicide rates it is surprising to observe a negative relationship between the Gini coefficient and suicide rates in this regression. Divorce R ate (DIV RATE) The divorce rate variable is only in cluded in regression two. It is proven to be statistically significant at the 99% confidence level. Therefore, the null hypothesis that there is no correlation between divorce rates and suicide rates can be rejecte d. This is the result that was hypothesized The coefficient is approximately 1.86 meaning that for every increase of one divorce per 1,000 population in a stat e, su i cides per 100,000 population will increase by approximately 0.019 The result comes as no surprise since suicide rates among divorcees tend to be high in the U.S

PAGE 13

! 13 Gun Ownership (GUN OWN) In both regressions, the gun ownership variable is proven to be statistically significant a t the 99% confidence level. T he null hypothesis that there is no correlation between gu n ownership rates and suicide rates can be rejected. This is th e result that was hypothesized. Th e regression s indicate that a 1 percentage point increase in gun ownership rates will increase suicides per 100,000 population by approximately 0. 1 2. Half of all suicides in the U.S. are committed using firearms, so it makes sense that an increase in gun ownership rates and therefore easier access to firearms, may contribute to an i ncrease in suicide rates. Where le thal means are more readily available, we can expect suicide rates to rise. V I I. CONCLUSION W hile correlation does not prove causation, t his study indicates that there is a positive and statistically significant correlation between alcohol consumption per capita and suicide rates in the Unite d States Studies co nducted by notable source s have observed this relationship, as well. As previously mentioned, alcohol is a depressant that amplifies the symptoms of mental disorders such as depression and causes those under the influence to act impulsively. This helps explain why over 25% of those who commit suicide each year are found to h ave alcohol in their system The increasing rate of suicide in the U.S. is an issue that comes as a result of many factors, not limi ted to substance use and abuse. For example, income, divorce rates and gun ownership rates were found to be highly correlated w ith suicide in this study In

PAGE 14

! 14 other studies, i t has even been found that suicide is highly correlated with increased social media use 20 This shows that even mundane factors, factors that have become staples of everyday life, may have an effect on su icide rates. While t he suicide epidemic is an issue that has gained national attention, the n umbers continue to rise Substance use and abuse is an issue that has been growing alongside suicide rates in the U.S. for some time. Economic and social factors play into both of these increasing rates. It is important that these issues continue to be studied and understood by the public, as well as by lawmakers. While recognition and understanding of this issue must be facilitated it is clear that some sort of policy changes m ust be enacted in order to curb the growing rate of suicide in the U.S. 20 Kelly, Laura (2017). "Teen suicide rate suddenly rises with heavy use of smartphones, social media". The Washington Times.

PAGE 15

! 15

PAGE 16

! 16

PAGE 17

! 17 Regression One Correlation Matrix "#$%&'() '*+ ,)&+%-'*) #.)-,%&'() '/0%$.+ 0$.$ 0#.%12. !"#$%&'( ) &*+ ,-). ) /(%+$0&*( ,-12 ,-33 ) "4(0/$%&'( 5 ,-3 5 ,-31 5 ,-63 ) &78$#4+ 5 ,-39 ,-3. ,-): 5 ,-31 ) 8#4# 5 ,-12 5 ,-6) 5 ,-1) ,-11 5 ,-33 ) 8"4$;<4 ,-:: 5 ,-) ,-63 5 ,-,3 5 ,-2) 5 ,-3: ) Regression Two Correlation Matrix "#$%&'() '*+ ,)&+%-'*) #.)-,%&'() '/0%$.+ 0$.$ 3$/%&'() 0#.%12. !"#$%&'( ) &*+ ,-)= ) /(%+$0&*( ,-19 ,-3 ) "4(0/$%&'( 5 ,-3 5 ,-36 5 ,-6: ) &78$#4+ 5 ,-31 ,-3= ,-)1 5 ,-36 ) 8#4# 5 ,-1: 5 ,-63 5 ,-19 ,-1) 5 ,-3) ) >#7$%&'( ? ,-1: ? ,-,,1 ? ,-)9 ? ,-). ? 5 ,-6 ? 5 ,-)2 ? ) 8"4$;<4 ,-:= 5 ,-)3 ,-6: ,-,2 5 ,-2: 5 ,-33 ,-22 ? )

PAGE 18

! 18 Sources: Centers for Disease Control and Prevention National Center for Health Statistics (2015). Re trieved 23 September 2017. https ://www.cdc.gov/nchs/pressroom/sosmap/suicide mort ality/suicide.htm American Foundation for Suicide Prevention (2015). Retrieved 25 September 2017. https://afsp.org/about suicide/suicide statistics/ National Institute on Al cohol Abuse and Alcoholism (2015). Retrieved 25 September 2017. https://www.niaaa.nih.gov/alcohol health/overview alcohol consumption/alcohol facts and statistics Montgomery County Emergency Service (n.d.). Retrieved 25 September 2017. http://www.mces.org/pages/suicide_fact_alcohol.php National Institute on Alcohol Abuse and Alcoholism (2016). Retrieved 26 September 2017. https://pubs.niaaa.nih.gov/publications/surveillance104/CONS14.pdf The Henry J. Kaiser Family Foundatio n (2016). Retrieved 25 October 2017. https://www.kff.org/other/state indicator/distribution by gender/ Bureau of Labor Statistics (2015). Retrieved 27 September 2017. https://data.bls.gov/map/MapToolServlet World Economic Forum (2015). Retrieved 27 September 2017. h ttps://www.weforum.org/agenda/2015/02/the link between unemployment and suicide/ United States Census Bureau (2015). Retrieved 16 November 2017. https ://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk Science Daily (2011). Retrieved 27 September 2017. https ://www.sciencedaily.com/releases/2011/02/110204130416.htm

PAGE 19

! 19 World Atlas (2017). Retrieved 12 February 2018. https://www.w orldatlas.com/articles/us states by gini coefficient.html Kposowa, Augustine J. (2000), "Marital Status and Suicide in the National Longitudinal Morality Study". Journal of Epidemiology & Community Health Retrieved 30 September 2017. http://jech.bmj.com/content/54/4/254 American Psychological Association (2017). Retrieved 26 October 2017. http://www.apa.org/topics/divorce/ Centers for Dis ease Control and Prevention National Center for Health Statistics (2015). Retrieved 29 September 2017. https://www.cdc.gov/nchs/nvss/marriage divorce.htm "Guns and Suicide: A Fatal Link" (2008). Harvard T.H. Chan School of Public Health Retrieved 30 September 2017. https://www.hsph.harvard.edu/news/magazine/guns and suicide/ Kalesan et. al. (2015). "Gun Ownership and Social Gun Culture British Medical Journal Retrieved 30 September 2017. http://injuryprevention.bmj.com/ Yong Hwan Noh (2009). "Does Unemployment Increase Suicid e Rates? The OECD Panel Evidence". Journal of Economic Psychology Retrieved 30 October 2017 http://www.sciencedirect.com/science/article/pii/S0167487009000361 Kelly, Laura (2017). "Teen suicide rate suddenly rises with heavy use of smartphones, social media". The Washington Times. Retrieved 1 D ecember 2017 https://www.washington times.com/news/2017/nov/14/teen suicides rise with smartphone social media us/