AN INTERPERSONAL PERCEPTION APPROACH TO UNDERSTANDING
COMMUNICATION BETWEEN PHARMACISTS AND PATIENTS
MICHELLE TAL ASSA
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
Michelle Tal Assa
This manuscript is dedicated to my parents, whose constant love and support have
given me the opportunity to realize this goal; to my grandmother and extended family,
who have always been there to encourage me; and to my friends, who have made these
years in Gainesville much more enjoyable.
My sincere gratitude goes to Dr. Carole Kimberlin, my major advisor, for her
tireless efforts throughout this project. I would also like to thank Drs. Donna Berardo,
Doug Ried and Rebecca Cline for their guidance and contributions to this research.
Many thanks to the Pharmacy Health Care Administration faculty for their
direction over the years. I would also like to thank my fellow graduate students who
were always there to commiserate and joke. My appreciation goes to DeLayne, Debbie,
Ned and the office staff for their flexibility and assistance which has made my work
easier in so many ways.
Finally, I would like to extend my love and appreciation to my family and friends
who have encouraged me throughout my graduate school experience.
TABLE OF CONTENTS
ACKNOWLEDGMENT ...................................... iv
ABSTRACT ............................................. vii
1 INTRODUCTION ..................................... 1
Problem Statement ................................. 1
Significance ..................................... 2
Background ..................................... 4
2 THEORETICAL FRAMEWORK AND SUPPORTING LITERATURE .... 8
Background ..................................... 8
Models of Interpersonal Perception ..................... 13
Interpersonal Perception Literature ..................... 16
Research Questions ............................... 20
3 REVIEW OF LITERATURE .............................. 23
Introduction and Background ......................... 23
Effect of Pharmacist-Patient Communication ............... 25
Extent of Patient Counseling ......................... 30
Pharmacist Views of Patient Counseling .................. 35
Pharmacist Variables Associated with Counseling ............ 39
Patients' Views of Pharmacist Counseling ................. 40
Changing Patients' Views of Pharmacy ................... 42
Summary ...................................... 44
Research Hypotheses .............................. 48
4 METHODOLOGY .................................... 50
Overview ..................................... 50
Instrument Development and Validation .................. 51
Study Variables .................................. 61
Analyses ...................................... 69
5 RESULTS .................
Characteristics of the Sample
Descriptive Analysis ......
Testing the Hypotheses ....
6 DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS
Interpersonal Perception Method ................
Perceptions and Pharmaceutical Care .............
Future Study ............................
A QUESTIONNAIRES ............................
Pharmacist Survey .........................
Patient Survey ...........................
B REVISED QUESTIONNAIRES .
Pharmacist Survey .......
Patient Survey .........
BIOGRAPHICAL SKETCH ..........
Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
AN INTERPERSONAL PERCEPTION APPROACH TO UNDERSTANDING THE
COMMUNICATION BETWEEN PHARMACISTS AND PATIENTS
MICHELLE TAL ASSA
Chairperson: Professor Carole Kimberlin
Major Department: Pharmacy Health Care Administration
The purpose of this study was to examine the relationships between pharmacists'
and patients' perceptions of pharmaceutical care type services and the extent to which
these perceptions are associated with the types of services offered to patients. Using an
interpersonal perception approach to understanding the perceptions that pharmacists and
patients have, this study examined i) pharmacists' agreement with patients, ii)
pharmacists' understanding of patients, iii) patients' understanding of pharmacists, iv)
pharmacists' realization about patients' understanding of them, v) patients' realization
about pharmacists' understanding of them, vi) pharmacists' feeling understood and vii)
patients' feeling understood. Further, this study analyzed the extent to which pharmacists'
agreement, pharmacists' understanding, pharmacists' realization and pharmacists' feeling
understood predicted pharmacists' reported provision of pharmaceutical care. Data were
collected via telephone interviews with 147 pharmacists practicing in the community
setting, and 151 adult patients who had a prescription filled during the previous six
Findings revealed that, according to the interpersonal perception method,
pharmacists and patients disagree on the pharmaceutical care services they believe would
be beneficial for patients. Overall, pharmacists perceived the services to be more
beneficial for patients than patients did. Further, pharmacists were found to
misunderstand patients' perceptions regarding pharmacy services. Likewise, patients
misunderstood pharmacists' perceptions. Results also indicate that pharmacists failed to
realize that patients misunderstood pharmacists' perceptions of the services. Patients, on
the other hand, realized that pharmacists misunderstood patients' perceptions of pharmacy
services. Despite this realization, patients did not feel misunderstood by pharmacists.
Finally, pharmacists felt erroneously understood by patients with regard to
pharmaceutical care services.
Pharmacists' understanding of patients' perceptions and pharmacists' agreement
with patients regarding the benefits of pharmaceutical care type services were found to be
significant predictors of pharmacists' provision of those services. However, only a small
percentage of the variability in pharmacists' behavior was explained by the model.
Based on these results, the study suggests that interventions designed to improve
communication between pharmacists and patients should focus on helping pharmacists to
realize their misunderstanding of patients and improving their understanding of each
others' perceptions of the benefits of pharmaceutical care type services.
Pharmacists, administrators, legislators and academics are calling for the
profession of pharmacy to move forward toward a more clinical, patient-oriented focus
in an attempt to provide better patient care and to improve patient outcomes from drug
therapy. One approach is the promotion of the concept of pharmaceutical care. Under
this philosophy, pharmacists work directly with patients and physicians in order to
design, implement and monitor patient medication regimens (Hepler and Strand 1990).
Pharmacists actively participate in decision making; monitoring and evaluating a
patient's use of medication. Despite noted benefits from this process, such as improved
compliance (Edwards and Pathy, 1984; Gotsch and Liguori, 1982; Sharpe and Mikeal,
1974; McKenney et al., 1973) and reduced cost of care (Knowlton and Knapp, 1994;
Forstrom et al., 1990), pharmacists in the community setting have been slow to adopt
this practice. Legal measures also have been aimed at expanding the type of services
provided by community pharmacists, yet few changes have occurred (Rumore, Feifer
and Rumore, 1995; Raisch, 1993; Alkhawajah and Eferakeya, 1992; Campbell et al.,
1989). In the community setting, pharmacists continue to dispense medications and
provide medication information sheets to patients in order to comply with regulations
established by the Omnibus Budget Reconciliation Act of 1990 (OBRA '90) (U.S.
Government Printing Office, 1990). But many pharmacists have yet to begin collecting
specific patient information or designing, implementing and monitoring patient-specific
In order for the profession of pharmacy to move toward these expanded roles,
barriers in the pharmacy environment (including pharmacist-related, patient-related,
informational/philosophical as well as other) must be overcome. These barriers will be
discussed in further detail in Chapter Two. Pharmacists often cite lack of patient
demand for these services as a primary rationale for not providing them (Nelson,
Zelnio and Beno, 1984; Raisch, 1993; Herrier and Boyce, 1994). Despite pharmacists'
perceptions, patients report wanting additional information as well as additional
services by pharmacists (Erstad et al., 1994; Enlund et al., 1991; Hirsch et al., 1990).
It is precisely this incongruence which is at the heart of this study. Of further interest
is a related issue regarding the provision of pharmaceutical care type services. Because
pharmacists cite lack of patient demand as a barrier to providing pharmaceutical care
services, it stands to reason that if pharmacists understand patient demand, they may
indeed be more likely to provide those services. To better understand this issue, one
must understand the relationships between pharmacists' and patients' perceptions.
As the profession of pharmacy begins to shift toward more patient centered
activities, communication with patients becomes more important. Although dispensing
the right drug in the right dose to the right patient remains paramount to the profession
of pharmacy, it no longer is sufficient to define the role of the professional pharmacist.
The implementation of pharmaceutical care is a step toward a more active, patient
centered role for pharmacists. Such a practice model depends upon a pharmacist's
ability to interact with patients, as well as other health care professionals.
In addition, pharmacists are faced with changes in health care financing.
Pharmacists are increasingly being asked to justify their role in health care. As noted
by Knowlton and Knapp (1994), pharmacist involvement in therapeutic decision making
can save a payor money. For the pharmacy itself, however, providing patient services
can be a way of satisfying patients' needs and desires while at the same time saving the
health care system money.
By understanding patients' desires for services such as the provision of
information about medications and monitoring of medication effects, pharmacists can
more accurately direct the types of services they provide for their patients. This study
will attempt to characterize the level of understanding between pharmacists and patients
in terms of what types of services would be beneficial for patients.
Researchers have shown that often pharmacists do not fully understand what
their patients would like from them (Herrier and Boyce, 1994; Hirsch, 1990; Carroll
and Gagnon, 1984a). Similarly, because patients have not had experience with
expanded pharmacy services such as pharmacist development, implementation and
monitoring of therapeutic plans, patients may not fully understand what pharmacists are
capable of providing or what services pharmacists would like to provide for them
(Chewning and Schommer, 1996; Gagnon, 1978). This difference in expectations and
understandings can lead to a pharmacist being frustrated by a patient's lack of patience
when picking up a prescription; or, a patient being frustrated by his or her pharmacist's
lack of attention when dispensing a prescription. This study will attempt to illustrate
the extent to which pharmacists and patients agree or disagree on the types of pharmacy
services they perceive to be valuable. Further, it stands to reason that if a pharmacist
understood patients' desires for expanded services, that he or she would indeed provide
those services. This study will also attempt to test that hypothesis. If pharmacists
understand patients' desires for pharmacy services, are they more likely to provide
those services or do pharmacists merely understand patients' desires but not incorporate
them into practice?
Pharmacists are in a unique position to provide patients with medication
information. Not only are they the last health care professionals to come in contact
with the patient prior to medication consumption, but they maintain a distinct
knowledge base of drug information which can provide patients with essential
information to prevent adverse drug effects and potential drug interactions. Policy
makers have recognized this potential and mandated through OBRA '90 that
pharmacists counsel all Medicaid patients receiving prescription medications beginning
January 1, 1993. Mandatory counseling laws are now widespread. The most recent
National Association of Boards of Pharmacy Survey of Pharmacy Law (1997) reported
that 45 states, including the District of Columbia, have required counseling for
Medicaid patients. One state had legislation pending to institute mandatory counseling,
and the remaining five states required an offer to counsel Medicaid patients. In
addition, 36 states required counseling for all other patients. Michigan and the District
of Columbia had legislation pending, while only seven states had no requirements for
counseling all other patients.
Pharmacists themselves, along with patients, recognize the need for patient
counseling. For example, Ortiz et al. (1984) asked pharmacists to complete a
questionnaire assessing their level of agreement with statements about the role of
community pharmacists. In response to the survey, 96 percent of pharmacists agreed
that community pharmacists should counsel patients about prescription and
nonprescription medications. They also agreed that pharmacists should keep up with
current drug information, should regularly attend continuing education programs and
have good working relationships with other health care professionals. Schommer and
Wiederholt (1994) found that pharmacists believe it is important to provide both oral
and written information to patients in order to avoid potential problems and
misunderstandings. Further, pharmacists have been shown to customize patient
counseling activities by identifying different elements of importance in patient
counseling based on various situations (Schommer and Wiederholt, 1994). However,
contrary to this view of pharmacist involvement, Hirsch et al. (1990) found that patients
want more drug information but reported that pharmacists do not always provide such
information. Alkhawajah and Eferakeya (1992) support this notion with their findings
that although pharmacists provided more information about medications than
physicians, patients still did not receive crucial information. None of the patients
involved in the study was told about potential adverse drug reactions and only 9 percent
of study patients were warned of important precautions.
Although pharmacists are in a position to counsel patients, and both pharmacists
and patients recognize the importance of patient counseling, many pharmacists still do
not counsel patients. To explain this phenomenon, pharmacists cite excessive
workload, lack of privacy, patient attitudes and store layout as barriers to providing
patient specific counseling services (Raisch, 1993; Herrier and Boyce, 1994). Carroll
and Gagnon (1984) noted that pharmacists' perceptions of patient demand is as
important as the actual patient demand in determining the provision of "patient-oriented
pharmacy services". The study explored pharmacists' perceptions of consumers'
demand for patient oriented pharmacy services. Researchers collected information
from both pharmacists and consumers and upon comparison of the two perspectives,
concluded that "pharmacists underestimate [patients'] demand for patient medication
records but not for voluntary provision of advisory services" (p.640). Herrier and
Boyce (1994) offer further support that, "one barrier regularly listed was low patient
expectations. Pharmacists felt that patients were not open to counseling because they
did not expect it" (p.23).
The extent to which pharmacists' perceptions serve as a barrier to counseling is
yet unknown. Examining pharmacists' perceptions of patients' expectations relative to
patients' actual expectations may provide a better understanding of any potential
barriers to counseling. This study will attempt to characterize pharmacists' perceptions
of patients' beliefs, patients' perceptions of pharmacists' beliefs as well as the extent to
which the congruency of these perceptions affects the communication that occurs
THEORETICAL FRAMEWORK AND SUPPORTING LITERATURE
The relationships between pharmacists' and patients' perceptions of the
communication that occurs, or can occur, between them is the primary focus of this
study. Communication between pharmacists and patients is an interpersonal process
that will be examined from an interpersonal communication approach to better
understand patient and pharmacist perceptions and the relationship between them. The
finding that pharmacists often perceive patients to be uninterested in learning more
about their medications despite patients' expressed desire for additional information,
may indicate that pharmacists hold inaccurate perceptions of patients' expectations.
Likewise, the fact that some patients may perceive their pharmacist as uninterested in
helping them may also be inaccurate. The congruency of pharmacists' and patients'
perceptions may determine if there is communication between the two and if so, the
nature of that communication. Thus, both pharmacist and patient perceptions must be
examined in order to understand and characterize the potential gap between them.
Individuals' perceptions help form the basis for the communication in which they
engage. For example, if patients perceive pharmacists to be too busy, then perhaps
they will not ask questions they have about medications. This perception can facilitate
the pharmacists' inaccurate perceptions of uninterested patients. Thus, each party's
perceptions not only shape the way they communicate (or not communicate) but in this
case may also further define the role of the pharmacist as merely a dispenser of
medications. This chapter will explore perception of self and others, the development
of models to explain interpersonal perception and finally will review interpersonal
People have the ability to explore the concept of self and as such develop
generalized views of themselves, or self-concepts. Self-concepts are derived not only
through rumination, but more significantly from interactions with others. Laing,
Phillipson and Lee (1966) call attention to the fact that many philosophical,
psychological and sociological theories have developed primarily from an "I"
orientation. Such theories lack an acknowledgment of the fact that "others" also play
an important role in the definition of self. However, Cooley (1902), with the concept
of the "looking glass self," began to explore the interplay between self and others and
how this in turn affects the "self". Cooley (1902) proposed that one's understanding of
self develops from one's perceptions of others' reactions to oneself. That is to say, one
learns about self mainly through experiences with others. The idea of the looking glass
self suggests that one's impressions of self are seen as a mirror of other's reactions. In
Laing's (Laing et al., 1966) terms, a person's experience is made up of not only a
direct view of the self and of a direct view of the other, but also includes an abstraction
of self. This includes others' views of the self as well as the self's view of their views,
which coalesce to form abstractions, labeled metaperspectives. For example, consider
that although I am unable to see myself as others see me, I am able to perceive them
seeing me in particular ways. These perceptions, combined with my view of self, form
my self-identity. In summary, the theoretical constructs of self-identity (my view of
myself) and meta-identity (my view of your view of me) are formulated not only by
looking at ourselves but also by looking at others looking at us and our perceptions of
the views of others' toward us, called metaperspectives.
The term phantasyy" has been used to denote a "primary way of experiencing
self and others which contributes to, and sustains, our relations with others throughout
life" (Laing, 1961, p.9). Isaacs (1952) suggests that the notion of phantasy helps a
person to identify and relate to others, and is a creation of a mode of experience which
is private, or internal, and only accessible by the self. One element of phantasy is that
knowledge of another person is based solely upon inference. Because a person's
experience is private, and unknown to others, one must attempt to interpret perceptions
of another person's behavior and oral communication in order to understand the other
person's experience. Further, this inference presupposes that the other person's actions
are in some way a function of the other's experiences. It is through perception that we
are able to experience and comprehend our environment as well as interact with people
in our environment (Heider, 1959). One can influence another's experience through
modification of one's behavior or alteration of the other's perceptions.
A person's behavior can be seen as a function of the person's experience with
another person's behavior in a common situation. More specifically, Person A's
behavior influences Person B's experience of the situation. This, in turn, influences
Person B's behavior which again influences Person A's experience (see Figure 1). In
essence, it is a circle of experience and influence. That is to say, the "behavior of each
towards the other is mediated by the experience by each of the other, just as the
experience of each is
mediated by the behavior of
each" (Laing et al., 1966;
A's Behavior B's Experience
p. 10). However, the
A Common situation B
behavior does not directly
lead to experience. It must A's Experience B's Behavior
first be perceived and
interpreted by the individual.
Figure 1. Interpersonal Perception and Behavior
As perception occurs, Laing, Phillipson and Lee (1966), p.9
interpretations are made. Further, because the criteria by which behaviors are
interpreted vary among individuals, one action may be perceived quite differently by
two people. For example, "to feel loved is to perceive and interpret, that is, to
experience the actions of the other as loving" (p. 11). Further, although the behavior
may be agreed upon, the meaning conveyed by the behavior may be disagreed upon.
For example, to one person, actions demonstrating love may be perceived by another as
controlling. Thus, experience entails the perception of an act as well as the
interpretation of it. Interpretations are based on past learning and can also vary
according to context. When two people do not agree on the meaning of a particular
behavior, questions arise as to their understanding of the difference of the
interpretations. This situation arises when peoples' perceptions of one another are
inaccurate. Further, one must question whether they realize that their understandings
differ. At the same time, two individuals can misunderstand each other and can realize
or fail to realize their misunderstanding.
For example, misunderstanding and misinterpretations of communication can
lead to spirals between two people which have dramatic implications for their
relationship. What one person thinks of the other, affects the second person's view of
self. This, in turn, affects the second's actions in response to the first. This example
illustrates the potential effect perceptions have on behavior. In order to discern the
nature of these spirals, one can view them as having multiple levels of analysis based
on multiple levels of perception. "Through my behavior I can act upon three areas of
the other: on his experience of me; on his experience of himself; and upon his
behaviors. In addition, I cannot act on the other himself directly, but I can act on my
own experience of him" (p.22). Three levels of perception that help us perceive
relationships with others, called metaperspectives, will be discussed. These include the
direct perspective, the metaperspective and the meta-metaperspective (Laing et al.,
As Laing et al. describe (1966), the direct perspective is a direct experience, or
perception. It is one person's view of another, or a person's view of self. For
example, one can perceive oneself as trustworthy while at the same time perceive
another to be untrustworthy. Second, the metaperspective is one person's view of
another's view of their relationship. Continuing with the example of trustworthiness, a
statement from the metaperspective may be "I imagine you would say I am
trustworthy." In other words, the metaperspective is each person's view of the other
person's direct perspective. Third, the meta-metaperspective is what I think you think
about what I think about you. For example, a statement from the meta-metaperspective
may be "I think that you imagine that I would say 'you are trustworthy'." In other
words, it is each person's view of the other person's metaperspective.
Models of Interpersonal Perception
Newcomb's Coorientation Model (1953) serves as the foundation for approaches
to interpersonal perception and thus offers
the groundwork for viewing how
interpersonal perceptions shape A B-- B
communication. As depicted in Figure 2,
the model assumes that two persons, A and X
B, are attracted to each other either
positively or negatively. Further, they Figure 2. Newcomb's Coorientation
Model; Wilmot (1975), p.84
simultaneously maintain an orientation or
attitude toward an object of communication, X. Newcomb (1953) suggests that
symmetry must be maintained in the triangular relationship. Under conditions of
discrepancy, such as if A dislikes X while B likes X, a state of psychological tension is
produced. This psychological tension must be reduced in some way. Symmetry can be
restored if A's attitude
toward B, or toward X,
changes. Other ways to A'A-B anderstming or
about X aboutX
restore symmetry include A t B
trying to change B's Accuracy
evaluation of X, A A N
Perceptions of Peelpions of
discontinuing communication B's co5nitions A' coapitions
Figure 3 Expanded Coorientation Model
with B, A seeking support McLeod and Chaffee (1973), p.484
from other persons, or A cognitively distorting perception of the realities of the
An expanded Coorientation Model was proposed by Taguiri et al. (1958) and
further elaborated by Chaffee and McLeod (1968). The expanded model facilitates the
measurement of the conceptual model detailed by Newcomb and employs the use of the
terms agreement, congruency and accuracy to describe perceptual differences between
two people. Congruency refers to the degree of similarity between A's own cognitions
and A's perception of B's cognitions. Communication can affect congruency by either
increasing or decreasing it depending on its initial level and accuracy. Agreement
refers to the degree of similarity between A and B's perceptions. This can also be
referred to as understanding in the context of a person-to-person orientation. Finally,
accuracy refers to the extent to which A's estimate of B's perceptions reflect the true
nature of his perceptions. Figure 3 provides a representation of the expanded
Although Taguiri et al. (1958) are concerned with person-to-person relations,
rather than person to object orientations, this model can be applied to both. McLeod
and Chaffee (1973) propose that the Coorientation Model can be used in understanding
dyadic communication because it provides the groundwork for a relational analysis. In
order to understand a relationship between two people, one must examine the nature of
the relationship itself rather than merely observing the two as individuals. In other
words, the dyad is the unit of analysis rather than the individual. Therefore, McLeod
and Chaffee (1973) propose that any index of a dyadic relationship should include a
measure of what each person's attitudes or orientation toward the other person or object
are, as well as what each person perceives the other's attitude or orientation to be. This
type of approach yields the relational data regarding congruence and accuracy.
Newcomb's and Taguiri's models of communication are the beginnings of
relational models. Relational perspectives on communication began to develop in the
1950s and 1960s when researchers in Palo Alto, led by Bateson, brought into focus
relationships as systems (Reardon, 1987). They proposed that within relationships,
"patterns of interaction emerge and dysfunctional patterns, or problems, can often be
identified and altered" (Reardon, 1987; p. 43). Relational communication suggests that
not only does communication between two people serve to transmit content, it also
serves to define and maintain their relationship. Through metaperspectives individuals
gauge the relational communication that occurs between them. As Wilmot (1975)
explains, in transactions, each person 1) assumes a role for herself, 2) imputes a role to
the other and 3) estimates what role the other thinks she is assuming. These ideas are
analogous to Laing's terms, of direct, meta- and meta-metaperspectives in that each
person has a view of his and the other's role. In any given transaction, both
participants attempt to define themselves, the other and the relationship by interpreting
the relational cues from their communication. In a relationship, each participant has his
or her own definition of the situation and also an awareness of and ability to adjust to
how he or she thinks the other person defines the relationship. This leads to a mutual
definition of a relationship.
Based on the ideas of the looking glass self and relational communication,
people define themselves based on the others' reaction to their projected identity.
Further, they tend to seek support for the identities they project (McCall, 1970).
Through the meta-metaperspective, a measure of the degree of support for the projected
identity is possible.
Interpersonal Perception Literature
The IPM evolved from a psychological orientation and has been used
predominantly in the study of interactions between married couples (Drewery and Rae,
1969; Kottas, 1969; Sillars and Scott, 1983) or between friends (Sullins, 1992).
Further, studies have focused on accuracy of interpersonal perception, perceived
understanding (Sillars et al., 1994), resolution of marital conflict (Knudson et al.,
1980), and relationship control (Wichstrom and Holte, 1993).
In order to measure and analyze these perspectives and the relationships among
them, an Interpersonal Perception Method (IPM) was developed by Laing et al. (1966).
For this method, two people, A and B, can be asked to respond to a series of statements
from each perspective. Comparisons are then made between the two direct perspectives
to assess the pair's agreement or disagreement on that statement. A comparison of A's
metaperspective with the B's direct perspective yields a measure of A's understanding
or misunderstanding of B. A comparison of A's meta-metaperspective with the B's
metaperspective yields a measure of A's realization or failure to realize whether B has
understood or misunderstood A. Finally, a comparison between one's own direct and
meta-metaperspectives yields a measure of feeling understood or feeling misunderstood.
This will be discussed in further detail in Chapter Four.
Drewery (1969) further expanded Laing et al.'s (1966) Interpersonal Perception
Method (IPM), and distinguished his Interpersonal Perception Technique in three ways.
First, the technique uses the Edwards Personal Preference Schedule (EPPS), a
personality questionnaire designed to measure 15 personality traits. Second, the
technique limits the perceptual analysis, using a married couple as an example, to the
level of "husband's view of wife's view of a topic" which is labeled as Laing's
metaperspective. Laing extends his analysis to the meta-metaperspective which would
be the husband's view of the wife's view of the husband's view. Drewery believed that
the meta-metaperspective introduced a degree of conceptual difficulty into the task
which would seriously limit its usefulness. Third, a difference lies in the method of
comparison between dyads. Laing compares dyads by comparing the amount and
nature of the agreements or disagreements. Drewery's techniques, on the other hand,
compare dyads in terms of agreements and disagreements on established personality
traits (the EPPS).
As Sillars and Scott (1983) detail, numerous studies have shown a positive
relationship between understanding and adjustment in a relationship (Dymond, 1954;
Christensen and Wallace, 1976; Newmark et al., 1977; Laing et al., 1966; Murstein
and Beck, 1972; Sillars et al., 1994; Knudson et al., 1980). In a few of these studies,
the association has been found only when the wife responds in predicting the husband's
perceptions (Murstein and Beck, 1972; Stuckert, 1963). Subsequent studies (Sullins
1992; Alperson and Friedman 1983; Schullo and Alperson 1984) have corroborated the
finding that females tend to be more accurate in their perceptions of their partner or
friend than males. Researchers have also found perceived similarity, expectations of
agreement and feeling understood to be associated with higher levels of marital
satisfaction (Corsini 1956; Laing et al. 1966; Dymond 1954). However, Sillars et al.
(1983) suggest that the relationship between perception and marital satisfaction depends
on the couple's assumptions about marriage. For example, as Fitzpatrick (1982)
identifies, couples may engage in different type of relationships: traditional, in which
husbands and wifes function in traditional male and female roles; independent, in which
husbands and wives complement each other and work together but not in traditional
roles; and separate, in which husbands and wives function almost independently of one
another. If a "traditional" couple values sharing, their understanding may be very
important to marital satisfaction. On the other hand, if a "separate" couple values
emotional distance or autonomy, perceived understanding may not be related to marital
satisfaction. Research among disturbed couples (those who were seeking therapy or a
divorce) in comparison to nondisturbed couples has shown that using the IPM, there are
fewer disjunctions in the interactions of nondisturbed couples and that the nondisturbed
couples report a higher degree of harmony (Laing et al., 1966). Another study using
the Interpersonal Perception Technique (Drewery and Rae, 1969) showed differences
between alcoholic and non-alcoholic marriages. In non-alcoholic marriages, wives'
understanding of their husbands increased with duration of marriage. Conversely, in
alcoholic marriages, wives' understanding of their husbands decreased with length of
Finally, Wichstrom and Holte (1993) used a shortened, modified version of the
IPM, the DYADE, to study the relationship between perception and relationship
control. This study improved upon earlier studies, which relied solely on the self-
report of behaviors, by observing couples on three occasions. Results indicated that
couples who communicated in dysfunctional ways reported less satisfaction in the
marriage. They also were less accurate in their perceptions of the other's feelings.
In summary, studies have shown agreement and understanding between partners
to be positively associated with adjustment and relationship control among married
couples. It would also seem that females are better able to perceive their partner
(Sullins, 1992; Alperson and Friedman, 1983; Schullo and Alperson, 1984). Laing,
Phillipson and Lee's (1966) Interpersonal Perception Method "is designed to measure
and provide understanding .. [of] the conjunctions and disjunctions, of two
individuals in respect of a range of key issues with which they may be concerned in the
context of their dyadic relationship" (p.38). Although traditionally used in the analysis
of personal relationships, this model of dyadic relationship seems to reach the essence
of the potential misperceptions between pharmacist and patient. A key issue in the
pharmacist-patient relationship is how pharmacists' and patients' perceptions influence
their communication. As will be detailed in the following chapter, pharmacists report
that they attempt to provide patients with the type of services they perceive patients
want, while at the same time patients perceive pharmacists' to be too busy to answer
questions and as such leave the pharmacy with unanswered questions. An adapted
version of the Interpersonal Perception Method may be helpful in providing a better
understanding of this discrepancy.
In parallel to intimate relationships, the extent to which pharmacists and patients
agree upon the potential benefits of pharmaceutical care, understand each other's
perceptions, realize their understanding and feel understood, may have a dramatic effect
upon their relationship. The incongruence that appears to exist between pharmacists
and patients is explored through the application of the interpersonal perception model.
Based upon the type information that can be gleaned from collecting data from the three
levels of perception; the direct, metaperspective and meta-metaperspective,
comparisons are made between pharmacists' and patients' perceptions to answer the
1. Do pharmacists and patients agree on the type of pharmacy services
which they perceive would be beneficial to patients if provided by
2a. Do pharmacists understand patients' perceptions of pharmaceutical care
2b. Do patients understand pharmacists' perceptions of pharmaceutical care
3a. Do pharmacists realize or fail to realize that patients understand (or
misunderstand) pharmacists' perceptions of pharmaceutical care type
3b. Do patients realize or fail to realize that pharmacists understand (or
misunderstand) patients' view of pharmaceutical care type services?
4a. Do pharmacists feel understood by patients with regard to
pharmaceutical care type services?
4b. Do patients feel understood by pharmacists with regard to
pharmaceutical care type services?
5. What is the relationship between pharmacists' agreement with patients,
pharmacists' understanding of patients, pharmacists' realization of
understanding and pharmacists' feeling understood and the extent to
which pharmacists provide pharmaceutical care type services?
Research questions 3a and 3b were shaped by the results of research questions
2a and 2b. According to the theory, once understanding or misunderstanding has
occurred, the questions of realization can be addressed. If, in fact, patients were found
to understand pharmacists, results from research question number 3a would indicate
whether pharmacists realize or fail to realize that understanding had occurred.
Alternatively, if patients were found to misunderstand pharmacists, the question would
address whether pharmacists realize or fail to realize that misunderstanding had
occurred. Likewise, results from question 2b shaped question 3b for patients. The
following chapter will address what is known from the literature about communication
between pharmacists and patients which may help answer these research questions.
REVIEW OF LITERATURE
Introduction and Background
In 1967, American Pharmaceutical Association President Bill Apple (Apple,
1967) called for pharmacists to communicate with other health professionals and
patients about the proper use of medications and to avoid becoming preoccupied with
the potential liabilities. After nearly thirty years and several legislative efforts,
pharmacists have yet to fully embrace their potential role as drug advisors. Pharmacists
have not yet assumed the role of drug educator, and patients continue to leave
physician's offices and pharmacies uninformed or misinformed about their medications.
Not only can pharmacists provide patients with information, they can assess patients'
response to treatment, provide ongoing monitoring of patients' response to treatment
and make recommendations to optimize patients' therapy. This expanded role for
pharmacists has recently become a focus within the profession of pharmacy. Hepler
and Strand (1990) outline this role for pharmacists as the provision of pharmaceutical
care that involves pharmacists working with patients and physicians to design,
implement and monitor patients' drug therapy. More than legislative requirements for
the provision of information, this type of involvement in patients' therapy relies on the
pharmacist's ability to communicate effectively with both patients and physicians.
Pharmacists must first initiate the communication with patients and then must collect
patient-specific information and utilize it to develop patient specific therapeutic plans.
In 1993, in order to assist pharmacists with the transition toward greater patient
involvement, the American Journal of Hospital Pharmacy printed guidelines on
pharmacist-conducted patient counseling. Fourteen points applicable to both
prescription and nonprescription drugs were listed as pertinent information. These
criteria are consistent with legislation (OBRA '90) mandating such patient counseling.
Through both legal and professional authorities, pharmacists are being urged to provide
patients with information. Historically, however, pharmacists have not provided much
information to patients. Despite the noted absence of counseling, patients recognize its
importance (Gagnon, 1994; Carroll and Gagnon, 1984b; Hirsch et al., 1990; Gotsch
and Liguori, 1992; Norwood et al., 1976).
The issue of patient counseling is of obvious significance to patients as well as
pharmacists and the profession of pharmacy. In order to better understand patient
counseling, researchers have observed the interactions between pharmacists and
patients. However, as discussed by Chaffee and McLeod (1973), it is not sufficient to
merely observe two individuals to understand their interaction, rather one must also
include a measure of each person's attitudes toward the other or object. The
Interpersonal Perception Method (IPM) is designed for this purpose--to measure and
provide an understanding of the "conjunctions and disjunctions of two individuals in
respect of a range of key issues with which they may be concerned in the context of
their dyadic relationship," (Laing et al., 1966; p. 38). In this case, it is not sufficient to
observe pharmacists' interaction with patients to judge the benefit of the services they
provide. Rather, one must ask both pharmacists and patients about their perceptions of
patient counseling from a pharmaceutical care perspective. Extensive research has
addressed patient counseling; some has included pharmacists' and patients' perceptions.
This chapter reviews the literature detailing the benefits of patient counseling in order
to demonstrate salience of this issue. Second, literature describing the extent to which
pharmacists engage in counseling behavior is reviewed to demonstrate what has been
learned through study of the issues. Finally, literature addressing pharmacists' and
patients' perceptions of counseling is reviewed to illuminate any gaps in our
understanding of pharmacists' and patients' perceptions as well as the congruency or
incongruence between their perceptions.
Effect of Pharmacist-Patient Communication
Researchers have used numerous terms in their description of pharmacist-patient
interactions. "Patient education", "patient counseling" and "pharmacist-patient
communication" have often been used interchangeably (DeYoung, 1996). As DeYoung
(1996) explains, researchers have been vague in defining "counseling." Some have
focused on pharmacists' provision of information specific to the medication (Gotsch and
Liguori, 1982; McBean and Blackburn, 1982; Slama and Gurwich, 1978; Madden,
1973), while more recently, others have studied more extensive pharmacist-patient
interactions in which pharmacists addressed patient concerns and questions in addition
to providing information (Borgsdorf et al., 1994; Opdycke et al., 1992; Kimberlin and
Berardo, 1987). Overall, benefits have been linked to pharmacists' involvement with
patients through the provision of written and oral information. These benefits include i)
an improvement in patients' knowledge about their therapy (Rantucci and Segal, 1986;
Jonston et al., 1986; Ascione and Shimp, 1984; McBean and Blackburn, 1982;
Madden, 1973; McKenney et al., 1973), ii) improved compliance (Lipton and Bird,
1994; Kimberlin and Berardo, 1987; Edwards and Pathy, 1984; Gotsch and Liguori,
1984; Slama and Gurwich, 1978; Clinite and Kabat, 1974; Sharpe and Mikeal, 1974),
iii) reduction in medication related problems (Hammarlund et al., 1985) and iv) cost
savings (Forstrom et al., 1990; Knowlton and Knapp, 1994). In keeping with the
interpersonal focus of this study, this section focuses on improved patient knowledge
and compliance because of the direct impact the pharmacist-patient relationship can
have on these variables.
Numerous studies have demonstrated that provision of information can improve
patients' knowledge about their own therapy. Madden (1973) found a significant
difference in patients' knowledge about drug treatment in a group of 120 patients
studied in an experimental education group, as compared to an equal size control group
which received traditional pharmacy services. In another 1973 study, McKenney et al.
found that patients in the experimental group showed a significant improvement in their
knowledge of hypertension and its treatment compared to control group patients. Both
of these studies utilized patient recall as the measure of improved knowledge and found
that experimental group patients were able to provide significantly more correct
answers than controls. Later studies confirmed that patients' knowledge can be
improved through pharmacist intervention (Johnston et al., 1986; Rantucci and Segal,
1986; Ascione and Shimp, 1984; McBean and Blackburn, 1982). McBean and
Blackburn (1982) found that patients who had received both written and oral
information had significantly higher knowledge scores than patients who received no
information from the pharmacist. Further, in their study of over-the-counter
medications, Rantucci and Segal (1986) found that oral counseling did increase the
patients' knowledge of potential side effects, precautions and contraindications of the
medication. DeTullio and Corson (1987) showed that "instruction with demonstration
and practice resulted in better patient understanding and performance of the correct
steps for inhaler use, with improved bronchodilation as measured by ...[pulmonary
function tests]" (p. 1802). Contrary to these findings, when Slama and Gurwich (1978)
studied the effect of pharmacist consultation, they found that patients who were
counseled were more knowledgeable about their disease state and appropriate
medication use. Yet, patients who were not counseled were more knowledgeable about
medication names and renewal information.
Overall, pharmacist intervention generally has been shown to have a positive
effect on patient knowledge. Patients can benefit from pharmacists' knowledge and
accessibility, which allows pharmacists to answer questions and provide information
about medications and disease states. However, as DeYoung (1996) explains, the
relationship between improved patient knowledge and other patient outcomes is
unknown. For example, Kimberlin and Berardo (1987) found no significant
relationship between total knowledge and compliance, raising questions about the
usefulness of patient knowledge as an outcome of pharmacist counseling. Rantucci and
Segal (1986) also suggest that pharmacists' provision of information may not be
sufficient to affect change in patients' behavior. They suggest that it is a relationship
between pharmacists and patients that can prove most beneficial in improving patient
outcomes. Ascione and Shimp (1984) suggest that the pharmacist-patient relationship
can be developed by pharmacists tailoring their communication to the patient's specific
Research in the area of compliance has shown conflicting results. Some
researchers have found no effect resulting from patient counseling on compliance. Yet,
others have found counseling promotes significant improvements in compliance. Early
research showed a short term improvement in compliance but little effect on long-term
compliance. McKenney et al. (1973) found that when taken on a long-term basis,
patients may become noncompliant even after initial compliance with the medication
regimen. During the five month period of counseling, patients randomly assigned to
the treatment group were significantly more compliant than non-counseled patients.
However, after the study period, the patients in the experimental group returned to their
baseline level of compliance. One study, Slama and Gurwich (1978), found that
consultation with a pharmacist had no effect on compliance. Kimberlin and Berardo
(1987) found a significant difference in patients' compliance between the pre-
intervention period and the intervention period where patients received medication
information from the pharmacist.
Pharmacist counseling can be particularly important in helping elderly patients
manage their complex medication regimens. Although studies by Wandless and
Whitmore (1981) and Roden et al. (1985) each found no significant improvement in
elderly patients' compliance despite pharmacist counseling, many studies of patient
compliance in the elderly have found that pharmacist involvement does improve
compliance (Lipton and Bird, 1994; Edwards and Pathy, 1984; Gotsch and Liguori,
1982; Sharpe and Mikeal, 1974). Specifically, Lipton and Bird (1994) studied the
effect of pharmacist intervention with elderly patients discharged from the hospital.
They found that at initial follow-up the experimental group patients were not more
compliant; however, at subsequent follow up they did have better overall compliance.
This supports the view that ongoing pharmacist contact may provided beneficial results
in compliance. Edwards and Pathy (1984) found that a counseled group of patients had
better compliance after six days. Further, Kimberlin and Berardo (1987) found a
significant relationship between knowledge about medications and compliance among
patients over the age of 60.
Results are mixed on whether written information alone is sufficient to improve
compliance. Sharpe and Mikeal (1974) studied the effect of providing written
information to patients receiving prescriptions for antibiotics. In a comparison of two
groups of patients (those who had received no additional information beyond the
prescription label and those who had received an information sheet), those who
received additional information were more compliant with their medication regimen.
Gotsch and Liguori (1982) also found higher rates of compliance among patients who
received patient package inserts for their antibiotic therapy compared to patients who
received no information. Clinite and Kabat (1974) suggest that "written drug
information without oral reinforcement was 'counterproductive' to compliance" (p. 85).
This statement was based on the fact that patients who received only written
information made the most 'errors' (30%) in dosing, while those who received verbal
review of the written information made the fewest 'errors' (14%). This suggests that
pharmacist involvement in providing written and oral information on an ongoing basis
is beneficial for patients. Kimberlin and Berardo (1987) found that patients responded
favorably to both written and oral information, as well as the combination of the two.
Pharmacists in the study preferred the combination of written and oral information and
believed that it was most effective in educating patients.
In summary, while some researchers found no significant improvement in
patients' compliance despite pharmacist counseling, the majority of research supports
pharmacists' ability to enhance compliance, via education and monitoring, particularly
in the short run. This suggests that ongoing pharmacist involvement may be the key to
improving patients' compliance with medication regimens.
Extent of Patient Counseling
Until the late 1960s few studies of pharmacists' communication with patients
were conducted. Knapp et al. (1969) suggest that while many articles may be found on
the philosophy of the pharmacist's role as a drug advisor, relatively few studies have
attempted to measure the quantity or quality of the pharmacist's performance. Knapp et
al. conducted one such a study, wherein a researcher posing as a diabetic patient
purchased a medication that is contraindicated in patients with diabetes. Of 36
pharmacists, only six refused to sell the medication to the patient. In another scenario,
a patient presented a prescription for a monoamine oxidase inhibitor used to treat severe
depression, and one week later returned with a prescription for Tofranil, another
medication for depression. Only one pharmacist out of 12 refused to dispense the
second prescription. This study helped identify the problem of the lack of pharmacists'
counseling in situations where a potential drug related problem exists.
A body of literature has emerged as researchers have attempted to quantify the
percent of patients who receive information from pharmacists. Early researchers
consistently found that approximately 70 per cent of patients do not receive oral
counseling from the pharmacist when they receive a prescription. Using pharmacy
students as mock patients, Rowles et al. (1974) found that 73 percent received no oral
counseling from the pharmacist when they picked up their prescriptions. Similarly,
Morris (1980) found that 72 percent of women received no oral counseling when
picking up a prescription for estrogen, while Ross et al. (1981) found that 69 percent
of patients received no oral counseling. Mason and Svarstad (1984) and Carroll and
Gagnon (1984) each found pharmacists to be providing counseling to more patients than
earlier studies. In these studies, 62 percent, 30 percent and 50 percent of patients did
not receive counseling, respectively. Mason and Svarstad (1984) used a mock patient,
whereas Carroll and Gagnon (1984b) conducted a mail survey of households.
Berardo, Kimberlin and Barnett (1989) conducted direct observation of
community pharmacists' interactions with patients. Prior to an educational intervention
with pharmacists, they found that in 381 encounters between pharmacists and patients,
between zero and 48 percent of patients received counseling from the eight different
pharmacists observed. Raisch (1993) also conducted an observational study in
conjunction with a self-report of pharmacists' patient counseling activities to compare
counseling based on different payment methods and practice settings. The percentage
of patients that were counseled ranged from only 7.9 to 12.9, with those in capitation
programs receiving significantly less.
Other national estimates are much higher. Meade (1994) reports that
pharmacists initiate conversations with patients 48 percent of the time, up from 39
percent in 1988. An earlier survey, reported by Meade (1992), found that 72 percent
of pharmacists offer advice to patients before they ask for it and that six out of ten
patients are counseled.
Some research has focused on the patient as a source of information regarding
pharmacists' counseling activities rather than relying on mock patients or pharmacist
self-report. Boyd et al. (1974) conducted interviews with patients, and found that 87
percent reported receiving no oral counseling from the pharmacist. Morris (1982)
conducted a telephone survey of 1,223 individuals to assess the level of counseling
provided to them by their pharmacist. Only 11 percent said that they had been
informed about potential side effects from their medication. Most (72 percent),
reported that nothing had been said to them at the pharmacy. Rather, they had received
written information in the form of auxiliary labels or leaflets. Some 19 percent
reported receiving no information at all from the pharmacist. Enlund et al. (1991)
surveyed 623 hypertensive patients, only 31 percent of whom were satisfied with the
amount of information they received on potential adverse effects. Finally, Wiederholt,
Clarridge and Svarstad (1992) report on an unpublished study in which Svarstad et al.
conducted direct observations of patients receiving prescriptions. Of the 558 patients
observed, 50 percent did not receive counseling. These studies suggest that only a
small fraction of patients are actually receiving counseling from pharmacists. Further,
the patients were not satisfied with the amount of drug information they received.
The National Pharmacy Consumer Survey, conducted in 1996, asked patients
about the information they received from their pharmacists. As Stover (1996) reports,
89 percent of patients report that they receive written drug information either "always"
or "often" for new prescriptions, and 70 percent of patients reported receiving written
drug information for refill prescriptions. Further, approximately half of patients
reported that they "always" or "often" speak with the pharmacist directly about their
medications. Despite these results, the patients also reported wanting more
information. This is consistent with an earlier study by Hirsch et al. (1990). Their
findings also suggested that patients were pleased with pharmacy services, but thought
pharmacists should supply more drug information.
One early study revealed differences in counseling based on practice setting. In
1978, Puckett et al., surveyed counseling practices in 95 pharmacies. They found that
practitioners in pharmacies where only prescription and nonprescription medications as
well as health care accessories were sold had the highest incidence of volunteering
information to patients when compared to practitioners in traditional pharmacies and
chain or discount pharmacies. Additionally, they found that when information was
given, it was most frequently related to warnings about the drug's action, the frequency
of administration and the need to avoid certain foods or activities. Although this
information was given, pharmacists generally failed to be specific with regard to the
timing of doses and many were not complete in the information provided. The
frequency of counseling varied by practice setting; in the traditional practice site, 36
percent of pharmacies provided voluntary counseling; in professional settings, 54
percent volunteered information; and in the chain or discount setting, 29 percent
provided voluntary counseling.
In summary, researchers have attempted to quantify the amount of patient
counseling that occurs in community pharmacies. Over time, there seems to have been
improvement. Recent results (Stover, 1996; Meade, 1994) have indicated that more
patients are receiving written and oral counseling from pharmacists than in the past
(Ross et al., 1981; Boyd et al., 1974). As described in numerous studies, many
factors may account for the observed variation, including practice setting, practice
environment and pharmacist characteristics. However, despite the benefits associated
with counseling and the wide range of estimates, it appears that pharmacists are not
providing consistent counseling to their patients.
Pharmacist Views of Patient Counseling
When exploring the apparent lack of patient counseling, one must consider the
views of pharmacists themselves, or pharmacists' direct perspectives of involvement
with patients. Many studies have attempted to understand the lack of pharmacist
counseling from the pharmacists' perspective. Zelnio, Nelson and Beno (1984) began
to explore potential barriers to pharmacists' involvement with patients. Community
pharmacists were sent mail questionnaires asking them to rate their willingness and
perceived level of competency for five potential levels of counseling ranging from
minimal to maximal interaction. Results indicated that both competency and
willingness may serve as barriers to pharmacists' provision of counseling services in
the community setting.
In further analysis of the responses to the mail questionnaire, Nelson, Zelnio
and Beno (1984) identified the following as barriers pharmacists perceived to their
provision of counseling services: lack of revenue from services, pharmacists' attitudes,
lack of time, physicians' attitudes, legal barriers, patients' attitudes, lack of patient
contact, pharmacist incompetence, lack of demand, services being too costly, lack of
facilities, inadequate pricing methods and pharmacists' characteristics such as type of
entry level degree. These barriers are echoed in more recent studies of barriers to
counseling (Raisch, 1993). In addition, lack of privacy and store layout were identified
by the 73 pharmacists who completed the survey.
The most comprehensive evidence is offered by Herrier and Boyce (1994) with
the results from workshops with nearly 30,000 pharmacists. Participating pharmacists
were asked to identify barriers to patient counseling at their practice sites. Once again,
common themes emerged in the discussions. Pharmacy environment, as well as
barriers that were pharmacist-related, patient-related, informational/philosophical and
miscellaneous were addressed. Barriers in the pharmacy environment included
excessive workload and lack of time or staff. The physical layout of the pharmacy and
lack of privacy were also mentioned as potential barriers. Pharmacist-related barriers
included lack of formal education or lack of knowledge about the prescribed drug, or
poor counseling skills.
Patient related barriers were most frequently identified as the patient being in a
hurry or uninterested in receiving information about the medication. The perceptions of
patients being uninterested in the information would be considered as coming from
pharmacists' metaperspective. Philosophical or information barriers were related to
pharmacists' questioning of their abilities to affect patient outcomes through counseling
activities. Finally, miscellaneous barriers include concerns over liability and lack of
reimbursement as barriers to providing counseling services. Consistent with previous
findings, Schommer (1994) also identified "lack of time" as the primary barrier to
counseling. Rumore, Feifer and Rumore (1995) in a study designed to explore
pharmacist' implementation of OBRA '90 regulations found that pharmacists listed time
constraints, lack of reimbursement and patient indifference as problems in
implementing OBRA '90 requirements. Finally, in one on one interviews pharmacists
identified the external barriers of time, practice setting, patient expectations and
personality as most salient (Assa, 1995).
Kirking (1982) offered a different, more positive view of pharmacists'
perception by looking at pharmacists' direct perspective of benefits of patient
counseling. In the study of pharmacists' perceptions of their patient counseling
activities as well the extent to which they engage in counseling activities, the
pharmacists reported that counseling would likely be beneficial for both patients
pharmacists. The pharmacists agreed that auxiliary labels and patient package inserts
cannot replace oral counseling. Further, they believed they were qualified to counsel
without additional training. Interestingly, they did not believe that the pharmacy layout
had an inhibitory effect on their counseling. Pharmacists' metaperspective of pharmacy
services was also included in the study. Pharmacists seemed to believe that although
patients may want counseling, they are unwilling to pay for those services.
Oliver and Barnes (1983) conducted an anonymous survey of 50 pharmacists
who identified communicating information about medications, gathering information
and responding to patients' anxieties as most important in patient counseling. They also
believed that pharmacists should initiate patient counseling rather than wait for the
patient to initiate it. Nelson, Zelnio and Beno (1984) found that pharmacists saw a
great need for their provision of clinical services, yet only a small demand for them.
Miller and Ortmeir (1995) also found that pharmacists believed they could provide
beneficial services that would meet the needs of patients. The pharmacists surveyed
ranked the services of providing a computerized patient profile, drug allergy screening
and drug interaction screening along with oral patient counseling and over the counter
medication counseling as the most important services they could provide for patients.
Consistent with the theoretical premise of interpersonal perception, pharmacists
rely on their perceptions of patients to define their role as pharmacists. Schommer and
Wiederholt (1994b) identified patient motivation, patient abilities and time available as
the most frequently cited determinants of the amount and type of counseling. The
situational elements were also identified as playing a role in pharmacists' decision to
counsel. Specifically, patients' familiarity with the medication and the pharmacists'
perceptions of seriousness of potential consequences with a particular medication were
important determinants of pharmacists' counseling. Schommer (1994b) found similar
results in a study of Ohio pharmacists. He, too, reports that pharmacists use "patient
desire for counseling and the type of medication to determine the amount and content of
counseling they give their patients" (p.765).
In summary, pharmacists recognize the importance of counseling yet find it
difficult to provide patients with the information they need due to many perceived
barriers. Pharmacists question their own knowledge and competency, yet also cite
external barriers such as lack of revenue, lack of time or lack of patient interest as key
issues. These barriers must be overcome if pharmacists are to proceed with what they
recognize as beneficial services. Pharmacists report that they perceive they are
behaving in accordance with what they think patients want from them. However, the
accuracy of pharmacists' perceptions is unknown.
Pharmacist Variables Associated with Counseling
The barriers identified by pharmacists suggest situational as well as personal
reasons patients may not be receiving as much counseling as they might desire or find
helpful. Differences between pharmacists may also be important when looking at the
provision of counseling services. Factors such as practicing independent pharmacy
settings, holding advanced pharmacy degrees, having completed more hours of
continuing education, working more hours per week and being more willing to
participate in continuing education programs have been found to be associated with a
higher level of provision of counseling services (Zelnio, Nelson and Beno, 1984).
Watkins and Norwood (1976) studied the relationship of practice environment and
pharmacist age to the quality of counseling pharmacists provide. They hypothesized
that there would be a difference in attitude, knowledge and behavior of pharmacists
based on the length of time since graduation and that there would be differences in
attitude, knowledge and behavior among pharmacists in different practice sites. They
found that there were no differences in the pharmacists attitudes or knowledge across
environment, yet there was a significant difference in behavior. Those pharmacists in
discount pharmacies had the lowest levels of counseling behavior, whereas those in
independent pharmacies had the highest. There was also a significant difference
between groups based on year of graduation on knowledge and attitude with those out
longer having less positive attitudes. No significant differences were found in
Researchers generally agree that practice setting does affect the behavior of
pharmacists, with those in independent type pharmacies exhibiting the most counseling
behavior. This difference may be due to the amount of time available to pharmacists
for counseling or a difference in emphasis placed on these services by managers. Some
research evidence suggests that barriers such as pharmacy environment and lack of time
in certain practice settings noted by pharmacists are indeed barriers to counseling
(Zelnio et al., 1984; Watkins and Norwood, 1976).
Patients' Views of Pharmacist Counseling
In contrast to the studies of pharmacists, many researchers have studied
pharmacist-patient interaction from the patient's perspective. For example, Hirsch,
Gagnon and Camp (1990) attempted to understand patients' direct perceptions of
pharmacy through consumer focus groups. Overall, the patients in the study were
pleased with their pharmacists, but believed that the pharmacists should supply more
drug information. In fact, some patients believed that pharmacists were deliberately
withholding information from them. Further, while some patients complained about
having to wait for their prescriptions, others merely wondered why a waiting room was
not available. Chewning and Schommer (1996) assessed patients' perceptions and
knowledge of pharmacists' roles. They found that patients perceived pharmacists to be
too busy to interact with them. Some patients also felt that pharmacists were rude.
As Gagnon (1994) details, patients rated talking with the pharmacist about issues
such as dosage directions, side effects, interactions and allergies as either important or
very important. This underscores earlier work by Carroll and Gagnon (1984), who
gathered consumer self-reports of importance of pharmacy services in an effort to
characterize the accuracy of pharmacists' perceptions of consumer demands.
Consumers rated the voluntary provision of advisory services as the most
important service that could be provided. Also noted as important by consumers were,
successively, the friendliness of pharmacists, availability of advisory services on
request, nominal increase in prescription price (for additional time spent with the
pharmacists) and maintenance of patient medication records were rated as important by
consumers. Finally, Perri et al. (1995) conducted a survey to assess the impact of
OBRA '90 on pharmacists and patients. They found that patients felt pharmacists
provided useful information. They also reported that patients were willing to give the
pharmacist personal information to help the pharmacist improve their care.
Patients can receive information about their medications from a variety of
sources: while they still look primarily to their physician, they do value the pharmacist
as an advisor. Stratton and Stewart (1990) found that the public ranked "physicians
more highly than the pharmacist as a source of drug information only 19 of 95
respondents (20 percent) ranked the pharmacist as the first person from whom they
would seek information about a prescription drug" (p.21). Nonetheless, pharmacists
were ranked second among consumers as a source for drug information and as sources
of assistance in monitoring compliance as well as prevention of drug interactions.
Interestingly, almost two thirds (65 percent) of the consumers ranked the pharmacist as
the primary source of information about nonprescription drugs. In an earlier study of
rural consumers, Norwood et al. (1976) found that pharmacists were thought to be
extremely important as advisors concerning drugs, poisons and sickroom supplies.
In the National Pharmacy Consumer Survey (Stover, 1996) patients were asked
whether they would be interested in pharmaceutical care type services. Specifically,
they were asked if they would be interested in a service where a pharmacist would 1)
review the patient's medical history and develop a plan for taking the medication
properly, 2) meet with the patient to discuss the medication plans and answer questions,
3) recommend changes in the treatment plan to physicians and 4) contact the patient
occasionally to discuss how well the medication is working and any side effects the
patient may be experiencing. The majority of patients (69 percent) had a favorable
attitude toward those services.
Changing Patients' Views of Pharmacy
Early research by Norwood (1975) suggests that by improving patient
communication about drug therapy, consumers' attitudes toward pharmacy can be
improved. Consumer attitudes were measured using a scale that included items such as
patient reports about whether or not the pharmacist had a good knowledge of health
problems and of medications. In a study designed to determine the effect of written
information on compliance, Gotsch and Liguori (1982) found that patients in the
experimental group, who received written information, reported interest in receiving
patient package inserts routinely with new prescriptions. Berardo and Kimberlin (1987)
showed that when pharmacists are trained to provide medication information, they can
change patients' attitudes toward pharmacists. Most patients (91 percent) interacting
with pharmacists who provided them with written and oral information reported that the
pharmacists showed personal concern. Erstad et al. (1994) conducted a study in which
a randomized experimental group received increased contact with the pharmacist.
Patients in the experimental group reported being more satisfied with the increased
levels of pharmacy service as well as appreciating the greater contact. Patients also
reported desiring more contact with the pharmacist.
Patients may be uncertain of appropriate behavior for their role in the pharmacy.
Chewning and Schommer (1996) studied the effect of providing patients with a
pamphlet which described potential pharmacy services to patients on patients'
knowledge of pharmacists' roles. Patients who received the pamphlet demonstrated
significantly more knowledge about pharmacists' abilities than those who did not.
Patients cite embarrassment and ignorance that it was appropriate to seek information
from pharmacists as reasons they did not do so more often (Chewning and Schommer,
1996). This suggests that patients, too, use cues from pharmacists to define their
In summary, research suggests patients value contact with the pharmacist and
feel the interactions can be helpful. In other words, from the patients' direct
perspective, pharmacy services are quite beneficial. Even when patients have not had
interaction with a pharmacist in the past, it appears that direct perspectives can be
altered by experience. If patients are given more opportunity to interact with
pharmacists, they appreciate it and come to prefer it. As pharmacists attempt to gauge
the type of services the patient would like to receive, they undertake a risk of
misunderstanding patients' desires. In order to more accurately direct services to
patients' wishes, pharmacists must first understand what those wishes are. Further,
pharmacists must understand that patients may not expect a high level of interaction
from the pharmacist based on past experiences, but that when additional interaction
with the pharmacist and additional information about medications are provided, patients
Both pharmacists and patients rely on perceptions of one another to define their
relationships. Patients report feeling embarrassed or being unaware that it was
appropriate to seek information from pharmacists and that pharmacists are too busy or
were rude. They also suggested that pharmacists "give direct clues that they expect and
want to take an active patient consultation role" (Chewning and Schommer, 1996;
p. 1303). Yet, when patient expectations are met or exceeded, higher levels of patient
satisfaction are found (Schommer, 1995). Pharmacists report utilizing their perceptions
of patient attitudes to determine the extent of counseling they provide (Schommer,
1994b; Schommer and Wiederholt, 1994a).
From the literature detailing pharmacists' and patients' perceptions about their
interaction, inferences can be made about the relationships between their perceptions.
According to the comparisons made in using the Interpersonal Perception Method
(IPM), of interest is whether pharmacists and patients agree on the issue of patient
counseling from a pharmaceutical care perspective; whether each group understands or
misunderstands the other, whether each group realizes or fails to realize that they are
understood or misunderstood; and finally, whether each group feels understood.
Research suggests that pharmacists recognize the benefits of offering their expertise to
patients (Schommer and Wiederholt, 1994b; Ortiz et al., 1984; Oliver and Barnes
1983; Kirking, 1982) and patients recognize the benefit of that information as well
(Stover, 1996; Erstad et al., 1994; Carroll and Gagnon, 1984; Gotsch and Liguori,
1982; Norwood et al,. 1976). For example, McGhan et al. (1980) found a high
correlation (0.89) between pharmacists' and consumers' rankings of pharmacy services.
Pharmacists' and patients' direct perspectives seem to be similar, which implies that
they may agree on the types of pharmaceutical care services they perceive as beneficial
Recent surveys of pharmacists (Herrier and Boyce, 1994; Schommer and
Wiederholt, 1994; Rumore et al., 1995) have shown that pharmacists identify patient-
related factors as barriers to counseling. Pharmacists in these studies perceived patients
to be in a hurry or uninterested in receiving additional information about their
medications. To the contrary, patients report wanting more information (Erstad et al.,
1994; Hirsch et al., 1990) and valuing the information provided by pharmacists
(Gagnon, 1994; Carroll and Gagnon, 1984; Norwood et al., 1976). This suggests that
pharmacists metaperspectives are not congruent with patients' direct perspectives,
which would indicate that pharmacists misunderstand patients' perceptions of
counseling. Patients, too, may misunderstand pharmacists' perceptions of counseling
behavior (Gagnon, 1978; Chewning and Schommer, 1996). One study found that some
patients believe that pharmacists were deliberately withholding information from them
(Hirsch et al., 1990). Based on the experiences patients have traditionally had in
pharmacies, they may not understand what pharmacists are capable of providing for
them (Chewning and Schommer, 1996; Erstad et al., 1994). Because patients are
unaccustomed to the additional attention of a pharmacist, they may believe that
pharmacists perceive it to be unnecessary. Patients' metaperspectives do not seem
congruent with pharmacists' direct perspectives, which indicates that patients may
misunderstand pharmacists' perceptions of the benefits of pharmaceutical care type
services. Thus, both pharmacists and patients may misunderstand the other's
perceptions of the benefits of counseling and fail to recognize their misunderstanding of
Little research exists as a basis for hypotheses about pharmacists' and patients'
realization of their misunderstanding because no research has examined pharmacists'
and patients' meta-metaperspectives directly. Pharmacists cite many barriers (e.g.
pharmacist-related or financial barriers) to counseling in the community setting
including patient demand. Logically, however, if pharmacists realized patients' desire
for counseling, patient demand would not be one of the perceived barriers they cite. It
seems that pharmacists may fail to realize that patients misunderstand them. It also
seems that patients fail to realize that pharmacists misunderstand their perceptions.
Researchers (Herrier and Boyce, 1994; Schommer and Wiederholt, 1994b; Schommer
1994a) have found that pharmacists consider patient desire for counseling as a factor in
deciding when to provide counseling. Patients report desiring more information, yet
pharmacists report perceiving them as uninterested. Perhaps if patients were to realize
that pharmacists perceived them to be in a hurry or uninterested, they would make their
views about counseling known.
Because of the suggestions that pharmacists and patients misunderstand each
other and each group fails to realize that the other group misunderstands them, as
described above, both groups feel misunderstood by the other with regard to patient
counseling activities. Further, because of the external barriers reported by pharmacists,
perhaps pharmacists feel that patients would not understand their perceptions of the
benefits of pharmaceutical care services. Finally, patients might expect that if the
pharmacist was aware of their desires for additional information, the pharmacist would
indeed provide counseling.
In order to predict pharmacists' behavior, it is logical to consider pharmacists'
perceptions of that behavior. Because pharmacists claim to consider patient demand
when providing counseling services (Herrier and Boyce, 1994; Schommer and
Wiederholt, 1994b; Schommer 1994a), it is reasonable to expect pharmacists'
understanding of patient perceptions to predict their behavior. Also, their agreement
with patients as to the benefits of that behavior is likely to be a strong predictor of that
behavior. Other data from the IPM, such as pharmacists' realization that patients
understand (or misunderstand) them and pharmacists' feelings of being understood (or
misunderstood), are likely to be weaker predictors of their behavior but will be
included in the analysis because they represent pharmacists' perceptions.
Based on the literature and theory of interpersonal perception and literature
regarding pharmacists' and patients' perceptions of pharmaceutical care, the following
hypotheses were proposed.
H1 Pharmacists and patients agree on the types of pharmaceutical care
services they perceive as beneficial for patients if provided by
H2, Pharmacists misunderstand patients perceptions of the benefits of
pharmaceutical care type services.
H2b Patients misunderstand pharmacists' perceptions of the benefits of
pharmaceutical care type services.
H3, Pharmacists fail to realize that patients misunderstand their perceptions
of the benefits of pharmaceutical care type services.
H3b Patients fail to realize that pharmacists misunderstand their perceptions
of pharmaceutical care type services.
H4, Pharmacists feel misunderstood by patients with regard to
pharmaceutical care type services.
H4b Patients feel misunderstood by pharmacists with regard to
pharmaceutical care type services.
H5 The level of pharmaceutical care type services reported by pharmacists is
predictable from the level of pharmacists' agreement, level of
pharmacists' understanding, level of pharmacists' realizations and level
of pharmacists' feeling understood.
This study investigated the relationships between pharmacists' and patients'
perceptions of the benefits of pharmaceutical care type services and the extent to which
those services are provided by pharmacists. By implementing an adapted version of the
Interpersonal Perception Method, pharmacists' and patients' perceptions were
compared to determine the extent to which i) pharmacists and patients agree on the
types of services they perceive would be beneficial for patients if provided by
pharmacists, ii) pharmacists understand patients' perceptions of these services, iii)
patients understand pharmacists' perceptions of these services, iv) pharmacists realize
that patients understand their perceptions of these services, v) patients realize that
pharmacists understand their perceptions of these services, vi) pharmacists feel
understood by patients with regard to the provision of pharmacy services, vii) patients
feel understood by pharmacists with regard to the provision of pharmacy services and
viii) the preceding factors predict pharmacists' reported provision of pharmaceutical
care type services. Data were collected via telephone interviews from both pharmacists
and patients throughout the state of Florida to characterize and compare their
perceptions. The remainder of this chapter describes instrument development and
validation, data collection procedures, study variables and data analysis.
Instrument Development and Validation
Interpersonal Method--Pharmaceutical Care Questionnaire
The interpersonal perception method was applied to aid in understanding
pharmacists' and patients' perceptions of beneficial pharmaceutical care type services.
The instrument included modified items from the Pharmaceutical Care Questionnaire
(PCQ) developed by Badejogbin (1994) as well as questions about the types of activities
pharmacists would perform if they were practicing according to the pharmaceutical care
practice model. These activities represented the domains of patient assessment, patient
consultation, patient record screening, implementation of therapeutic objectives,
documentation of patients' medical information and verification of patient
understanding. PCQ questions were chosen based on the visibility of the tasks to
patients and were modified to lay language. Some pharmaceutical care activities such
as documenting information may not be visible to patients. Because the visibility of
these services may be low and patients unaware of their benefit, these items were not
used. In addition to the relevant PCQ questions chosen, questions specific to the
communication that occurs between pharmacist and patient were developed through a
review of relevant literature to enhance the domain of patient consultation.
Once the items were generated, questions were worded to reflect the
interpersonal perception method. This involved phrasing the questions as statements
from each of the perspectives. For example, items on the patient questionnaire were
phrased from the patient's perspective, what the patient believes the pharmacist thinks
and finally what the patient thinks the pharmacist thinks of the patient's beliefs.
Consider the item "discuss the patient's drug therapy with him or her." On the patient
questionnaire this item appeared in the following forms: 1) the patient's direct
perspective "I think it would be beneficial if my pharmacist were to discuss my drug
therapy with me," 2) the patient's metaperspective "My pharmacist would say 'I think it
would be beneficial for my patients if I were to discuss the patient's drug therapy with
him or her,'" and 3) the patient's meta-metaperspective "My pharmacist thinks I would
say 'I think it would be beneficial if my pharmacist were to discuss my drug therapy
The pharmacists' questionnaire included identical items although they were
originally phrased from the pharmacist's perspective and led to the pharmacist's
perception of the patient's perspective and finally to what the pharmacist thinks the
patients think about the pharmacists' beliefs. Continuing with the same example, items
appeared in the following forms: 1) the pharmacist's direct perspective "I think it
would be beneficial for my patients if I were to discuss the patient's drug therapy with
him or her," 2) the pharmacist's metaperspective "My patients would say 'I think it
would be beneficial if my pharmacist were to discuss my drug therapy with me,'" and
3) the pharmacist's meta-metaperspective "My patients think I would say 'I think it
would be beneficial for my patients if I were to discuss the patient's drug therapy with
him or her.'"
Fourteen items were included in the original version of the Interpersonal
Perception Method-Pharmaceutical Care (IPM-PC) questionnaire for patients and
pharmacists (see Appendix A). The items were scaled on a 7-point scale with the
anchors of 'strongly agree' and 'strongly disagree'. A 7-point rather than a 5-point
scale was chosen to elicit greater variability in response to the items.
Pretest of Instrument
A small number of patients and pharmacists (five each) were asked to complete
the instrument and were interviewed after completing the instrument to discuss any
problems they had in understanding the questions and to comment on the ease of
completing the instrument. Of particular concern were the statements from the meta-
metaperspective. As Drewery (1969) suggests, respondents may be confused by the
phrasing of meta-metaperspective items and have difficulty completing them. If that
was found to be the case, items would have been converted to direct questions of
feeling understood, or eliminated if necessary. This was not the case, however,
because the pharmacists and patients who completed the survey reported being able to
understand the questionnaire. When asked to explain their understanding of the three
perspectives, all of the pharmacists and four of the five patients gave accurate
descriptions of the perspectives. This suggested that it was reasonable to include the
meta-metaperspectives statements in the pilot study. Few editorial revisions were made
to the instrument prior to a larger pilot study of approximately 30 pharmacists and 30
Sampling Procedure. Pharmacists were selected randomly from the current
listing of active pharmacists in the state of Florida. The database received from the
Board of Professional Regulations included pharmacists' home address. Only
pharmacists residing in the state of Florida were included. Because the database did not
include home telephone numbers, once names were randomly selected, pharmacists'
home phone numbers were looked up in the most recent telephone directories for their
cities. A world wide web site (Database America) was also used to attempt to identify
phone numbers that were unlisted in the city directories.
In order to identify patients, a random sample of telephone numbers throughout
the state of Florida was purchased, from Genesys, Inc., by the survey research firm
assisting with the interviews. The sample purchased was a random digit dial sample
from which certain exchanges known to be primarily business numbers had been
Interview procedure. Data collection was via telephone interview by the Florida
Survey Research Center at the University of Florida. This methodology was chosen for
data collection because of the ease of obtaining the data and the increased response rates
over mail questionnaires (Dillman, 1978). Both pharmacist and patient interviews were
completed between March 10 and March 14, 1997. Individuals were contacted by
telephone by trained interviewers between the hours of five pm and nine pm.
Interviewers recorded the results of each telephone call (e.g. if there was no answer,
busy signal, answering machine etc.) and attempted to contact each individual three
All interviews began with identification of the interviewer, and an explanation
that the interviewer was calling from the University of Florida. This was followed with
a brief description of the survey. Respondents were informed that their participation in
the study was voluntary and verbal consent was obtained from each participant prior to
establishing eligibility. At that point, respondents were screened to determine if they
were eligible to participate in the survey. Pharmacists were asked in what type of
setting they practiced. Only those pharmacists who practiced in the community setting
were eligible to complete the survey. Patients were asked if there was someone in the
household who had a prescription dispensed during the previous six months and if that
person was over the age of 18. If that person was available, he or she was asked for
consent to be interviewed. When respondents did not meet those criteria, they were
thanked for their willingness to participate and the telephone call was terminated.
Pharmacists and patients were called until 30 of each had completed the interview.
Patient questionnaires included the interpersonal perception method
questionnaire (IPM-PC) assessing how beneficial they perceived the pharmacy services
would be to them if they were provided by their pharmacist. Information about the
patient such as the number of prescription medications currently being taken, number of
pharmacies visited on a regular basis and number of times in the past six months they
have had a prescription dispensed was also collected. Questions were asked about
patients' experience with pharmacists including; "Do you know the name of the
pharmacist who usually fills your prescriptions?," "Does your pharmacist usually
volunteer information about your medicine when you get a new prescription filled?,"
"Does your pharmacist usually ask if you are having any problems when you get a
refill?," and "What type of pharmacy do you go to?." Demographic information
including age, gender, level of education and income was also collected. The mean
length of patients' interviews was 12.3 minutes (s.d. =3.6).
Pharmacist questionnaires included the interpersonal perception method
questionnaire (IPM-PC) related to the types of pharmacy services they believed should be
provided and the self-report measure of the pharmaceutical care type services they
actually provided to patients (PCQ). Information about the pharmacist's education (e.g.
type of degree held and number of years in practice) and information about the pharmacy
(e.g. type of pharmacy and number of prescriptions dispensed per day) as well as
demographic information about the pharmacist (age and gender) were collected. The
mean length of pharmacists' interviews was 17.8 (s.d. =2.8) minutes.
Data was entered to a SAS file by the Florida Survey Research Center and was
checked for errors via examination of the frequencies of the variables. Cases were also
spot checked for accuracy of data entry after the file was obtained from the Research
Results of Instrument Pretest
Sample. Twenty eight pharmacists completed the questionnaire. Of those, 35
percent (N= 10) were female. Half of the pharmacists (N= 14) reported holding
Bachelor of Science degrees, while 46 percent (N= 13) reported holding Doctor of
Pharmacy degrees. One pharmacist held both a Bachelor of Science and a Doctor of
Pharmacy degree. Ten (36 percent) of the pharmacists had been in practice for over 20
years, eight (29 percent) reported being in practice between 11 and 20 years, and 10
(36 percent) had been in practice less than ten years. When asked to describe their
practice, six (21 percent) pharmacists reported practicing in independent pharmacies,
12 (42 percent) reported practicing in drug chain store such as Eckerd's and 10 (36
percent) reported practicing in discount chain pharmacies such as WalMart. More than
half of the pharmacists (54 percent, N= 15) described their position as an employee
pharmacist, 39 percent (N= 11) described their position as a pharmacy manager, and 7
percent (N=2) described their position as pharmacy owner.
Thirty one respondents completed the patient version of the questionnaire. Of
those, 68 percent (N=21) were female. Twenty-six percent (N=8) of the patients were
over the age of 60 years, 36 percent (N= 11) were between the ages of 40 and 60 years,
and 48 percent (N= 12) were between the ages of 18 and 39. Thirty-six percent
(N= 11) of patients reported patronizing a discount chain pharmacy such as WalMart,
36 percent (N= 11) reported patronizing a drug chain such as Eckerd's, 10 percent
(N=3) reported patronizing an independent pharmacy, 13 percent (N=4) reported
patronizing a grocery store pharmacy, and 13 percent (N=4) reported using another
type of pharmacy. Nine patients (29 percent) reported having an approximate
household income of less that $20,000, fifteen (49 percent) reported having an income
between $20,000-$59,999, and four patients (13 percent) reported a household income
greater than $60,000.
Instrument. Responses to the pretest from pharmacists and patients indicated
some problems with the instrument. Respondents indicated that the questions were
sometimes confusing and that the questionnaire was too long. They also had difficulty
with the 7-point range of responses. Based on the comments from the pretest, it was
decided to reduce the number of items in an effort to shorten the questionnaire. This
was intended to reduce the amount of time people were kept on the telephone and
increase response rate, but also to ensure that respondents would not be fatigued upon
making the final transition to thinking from the meta-metaperspective. Also, it was
decided that the range from one to seven may be too confusing. It is possible that it
was more difficult for people to use the seven point response range over the telephone
when compared to a written format of the questionnaire. The range was increased to
one through ten in order to maximize variability while at the same time offering
familiar response categories. Additionally, a ten point scale eliminates a true midpoint
which elicits either positive or negative reactions from respondents.
Reliability. Statistical procedures were used to assess the reliability of each
instrument and to determine which items would be eliminated. Internal consistency
reliability is designed to measure the extent to which the items in an instrument are
homogeneous. Cronbach's alpha was used to establish the internal consistency of the
subscales IPM-PC (direct, metaperspective and meta-metaperspective scales for
pharmacists and patients). These reliabilities can be found in Table 4-1.
Table 4-1. Reliabilities of Pretest Scales
Scale Original Alpha Revised Alpha
Fourteen Items Seven Items
Direct 0.87 0.82
Meta 0.90 0.82
Meta-Meta 0.96 0.91
Direct 0.90 0.83
Meta 0.97 0.92
Meta-Meta 0.98 0.95
In addition, the inter-item correlation matrices and corrected item to total correlations
were analyzed to determine whether any item did not meet criteria for inclusion in a
summated scale. Table 4-2 (page 75) lists the corrected item to total correlations and
alpha coefficients for pharmacists' scales while Table 4-3 (page 76) provides the same
information for the patients' scales. Some suggest that acceptable correlations should
be 0.50 or greater (Bearden et al., 1989); however, Nunnally (1978) suggests that
corrected correlations of greater than 0.30 are sufficient. For this study, corrected item
to total correlations greater that 0.30 were considered acceptable.
Modifications. In making decisions regarding the modifications of the
instrument, results from both pharmacists' and patients' questionnaires were reviewed.
The consensus of pretest subjects was that the instrument needed to be shortened. One
way of accomplishing this was to eliminate items from domains represented by more
than one item. Also, the relative visibility of each behavior to patients was
reconsidered in determining which items would be included in the questionnaire. Table
4-4 (page 77) details the Interpersonal Perception Method-Pharmaceutical Care items
and pharmaceutical care domains they represent. In revising the instrument, items A
("Explain to patients how to use their medicationss)) and B ("Tell patients about other
drugs which may cause problems with their medicationss)) were selected to represent
the domain of consultation (provision of information). Items F ("Ask patients questions
to find out if they might be having any problems with their medications") and J ("Ask
patients questions to figure out if their goals) for their medications) are being
reached") were selected to represent the domain of assessment. Items K and L were
merged to form one item ("Design and carry out follow-up plans to measure my
progress toward my goals) for my medicationss).) which represented implementation
of therapeutic plans. Item H ("Carry out plans to solve (or prevent) problems with
patients' medications") was also included in the domain of implementation. Finally,
item N ("Make sure that patients understand the information given to them") was
included for the domain of verification of patient understanding.
Content validity of the revised IPM-PC was assessed via a panel of experts who
matched items to the performance domains to ensure that the scale represented the
intended behavior domain.
Pharmacist and patient interviews were conducted according to the procedure
followed for the pretest, described earlier in this chapter. Patient data were collected
over a period of four days from March 24 through March 28, 1997. Pharmacist data
were collected over a period of four days from April 8 through April 11, 1997.
The necessary sample size was determined based the standard deviation and
means for the scale variables from the pretest data using ST-plan (Brown et al., 1993)
software. In order to detect differences between the group means using T-tests and based
on achieving a statistical power of 0.80 and a Type I error rate of 0.05, sample size
estimates were calculated for each of the comparisons to be made with the interpersonal
perception method. The largest sample size was needed to detect differences between
pharmacists' metaperspective and patients' direct perspective. For this comparison, the
sample size estimate was 150 respondents. Estimates made based on the other IPM
comparisons resulted in fewer respondents needed. Thus, 150 was chosen as the desired
sample size for pharmacist and patient groups.
According to the interpersonal perception method, by comparing the three
perspectives (direct, metaperspective and meta-metaperspective), agreement,
understanding, realization of understanding and feeling understood can be calculated for
pharmacists and patients. Using the IPM-PC questionnaire, data were collected from
patients and pharmacists for all three of the perspectives. The questionnaire addressed
the perceived benefits of pharmacists engaging in behavior that is consistent with a
pharmaceutical care approach to patient counseling. Specific examples include whether
or not it would be beneficial to patients if the pharmacist were to discuss potential side
effects of medications with patients; ask if patients have any questions to ask of the
pharmacist; and, address questions or concerns patients have. See Appendix B for the
final version of the questionnaire. The operational definitions of the variables appear
Direct Perspective (Pharmacist and Patient)
The "direct perspective" is the pharmacist's or patient's perceptions of the
benefits of pharmaceutical care type services derived from the responses to the IPM-PC
from the first level of perception--one's own. "Patient Direct Perspective" is a summed
score of the patients' responses to the items from the direct perspective (items la-lg on
patient questionnaire). These items addressed perceived benefits to patients of
pharmacist activities of patient assessment, patient consultation, record screening,
implementation of therapeutic plans and patient understanding verification. The parallel
variable, "Pharmacists' Direct Perspective", is calculated based on the pharmacists'
direct perspective statements (items la-Ig on pharmacist questionnaire). Each item was
measured on a 10 point scale anchored by "strongly disagree" (1) and "strongly agree"
(10). Because the questionnaire included 7 items, the potential range of scores for the
overall direct perspective variables is from 7 to 70.
Metaperspective (Pharmacist and Patient)
The "metaperspective" represents the pharmacist's or patient's responses to the
IPM-PC items from the second level of perception--the other person's. This is a
measure of pharmacists' and patients' perceptions of each other's perceptions of the
benefits of the pharmaceutical care type services. Specifically, pharmacists' responded
based on how they thought their patients would answer. Likewise, patients responded
based on how they thought their pharmacist would answer. A summated score of
patients' metaperspective scores responses (items 2a-2g on patient questionnaire)
yielded "Patients' Metaperspective", and a summed score of pharmacists'
metaperspective responses (items 2a-2g on pharmacist questionnaire) yielded
"Pharmacists' Metaperspective". An overall metaperspective score for the IPM-PC
questionnaire was calculated like the direct perspective score described above. It was
measured on a 10 point scale anchored by "strongly disagree" (1) and "strongly agree"
(10). The potential range of scores for the overall metaperspective variable is the same
as the direct perspective score range (7 to 70).
Meta-Metaperspective (Pharmacist and Patient)
The "meta-metaperspective" represents responses to the IPM-PC questionnaire
from the third level of perception--one's perception of another's perception of one's
own perception of the benefits to the patient of pharmaceutical care type services.
Pharmacists' responses to the meta-metaperspective versions of the items (3a-3g on the
pharmacist questionnaire) were summed to calculate the variable "Pharmacists' Meta-
metaperspective". Likewise, patients' responses to the meta-metaperspective versions
of the items (items 3a-3g on patient questionnaire) were summed to create the variable
"Patients' Meta-metaperspective". Responses were measured on a 10 point scale
anchored by "strongly disagree" (1) and "strongly agree" (10). The potential range of
scores for the overall instrument on the meta-metaperspective variables is the same as
described under the direct perspective.
Level of Pharmaceutical Care: The Pharmaceutical Care Questionnaire (PCQ)
The dependent variable in this study is the extent to which pharmacists engage
in pharmaceutical care type activities when interacting with their patients. Using the
Pharmaceutical Care Questionnaire (PCQ) (Badejogbin, 1994), pharmacists' self-report
of behavior was used as the measure of the types of services which they actually offer
to their patients. The PCQ asks the pharmacist to recall the last five patients who
presented a new or refill prescription for a medication used to treat a chronic condition
and to indicate with how many of these patients the pharmacist pursued patient specific
information sharing and discussion. Examples of pharmacist behavior include asking
what the patient would like to achieve from therapy, discussing the patient's drug
therapy and documenting all medication currently being taken by the patient. Further,
the PCQ asks what action was taken, if any, if a drug-related problem was detected.
The activities measured by the PCQ can be divided into six domains including 1)
documentation activities, 2) patient assessment activities, 3) therapeutic objective
implementation activities, 4) patient consultation, 5) patient understanding verification
and 6) record screening activity (Badejogbin, 1994). Pharmacist response scores were
summed based on the number of patients they indicate for each question (0-5). There
are eighteen questions over the six domains in the PCQ, thus, the total potential score is
90 for the overall instrument. Each of the six domains is described below.
Documentation. The domain of documentation included six items. They
include whether the pharmacist documented information about a) the patients' medical
condition, b) all medications currently being taken, c) the therapeutic objective, d) any
drug-related problems, and e) the therapeutic objective for each of the drug-related
problems identified and any interventions made. The range of scores is from zero to
thirty for this subscale.
Assessment. The domain of patient assessment was also comprised of six items.
These included, a) asking the patient to describe his/her medical condition, b) asking
what he/she wants to achieve from drug therapy, c) assessing patterns of actual
medication use, d) assessing whether the patient is experiencing any drug-related
problems, and e) assessing the patients' perceptions of the effectiveness of medications
and assessing whether the therapeutic objectives are being reached. Again, the range of
scores for this subscale was from zero to thirty.
Implementation. Implementation of therapeutic objectives was a domain
measured using three items. The range of scores possible for this subscale is from zero
to fifteen. The items include whether the pharmacist a) implemented a strategy to
resolve any drug-related problems, b) established follow-up plans to evaluate progress
toward therapeutic objectives, and c) carried out the established follow up plans.
Screening. Patient record screening was a domain measured using one item to
assess the frequency of pharmacists checking patient records for potential drug related
problems. Since this subscale was measured using one item, the range of scores was
from zero to five.
Consultation. The domain of consultation most directly addresses the
communication between pharmacists and patients. It was measured by the item
assessing the frequency of pharmacists discussing the patient's drug therapy with the
patient. The range of scores for this subscale was from zero to five.
Verification. Verification of patient understanding was a domain measured
using an item asking pharmacists the frequency with which they verified patient
understanding of the information which was presented. In the factor analysis conducted
by Badejogbin (1994), this item was found to be a separate domain. The range of
scores for this subscale was from zero to five.
In a survey of 793 community pharmacists, the overall reliability of the PCQ
was found to be 0.88 (Badejogbin, 1994). The reliabilities of the six domains were
found to be as follows: 0.72 for documentation, 0.90 for patient assessment and 0.74
for therapeutic objective implementation. The reliabilities for the other three domains
(patient record screening, patient consultation and patient understanding verification)
were not calculated because they were measured by a single item.
Pharmaceutical Care Score II
Pharmaceutical care score II was a variable which measured the extent to which
pharmacists engaged in the pharmaceutical care activities described above that were
included in the IPM-PC questionnaire. Thus, it measures the extent to which
pharmacists engage in the behaviors, visible to patients, that they were asked to respond
to about how beneficial they are for patients. The PCQ items included in this variable
are as follows i) "verify that the patient understands the information I present to him or
her," ii) "discuss the patient's drug therapy with him or her," iii) "ask the patient
questions to find out if he/she is experiencing drug-related problems," iv) "ask the
patient questions to ascertain whether the therapeutic objectives) is (are) being
reached," v) "implement a strategy to resolve (or prevent) the drug related problems,"
vi) "establish follow-up plans to evaluate the patient's progress toward his/her drug
therapy objectives" and vii) "carried out the follow up plans established for the patient's
progress toward his/her drug therapy." PC-2 was calculated as a summated score of
the PCQ items listed above.
Pharmacists' agreement is the extent to which a pharmacist is in agreement with
patients' on the responses to the IPM-PC scale. It was calculated for each pharmacist
as a difference score between the pharmacist's total score for the overall instrument and
the mean of all patients' responses on the scale. As a difference score, the scores for
this variable can range from the negative value of the largest difference (-63) to the
positive value of the largest difference possible (+63). The largest difference possible
is calculated based on the minimum and maximum potential scores for the scale, which
are seven and 70 respectively for the seven-item instrument. The difference between
them, 63, represents the largest potential difference score. Thus, the potential range of
scores for this variable is -63 to +63. According to the Interpersonal Perception
Method, a comparison from the pharmacists' perspective is made by entering the
pharmacists' perspective first. Thus, a positive value would imply that pharmacists
rated the items higher than patients. In other words, a positive difference suggests that
pharmacists believed the pharmacy services would be more beneficial than patients did.
On the other hand, a negative value would suggest that pharmacists believed the
pharmacy services would be less beneficial than patients did.
Pharmacists' understanding is the extent to which pharmacists understand
patients with regard to pharmaceutical care activities. It was calculated according to the
interpersonal perception method by comparing the pharmacists' metaperspective to the
patients' direct perspective. This comparison was made for each pharmacist based on
their overall metaperspective score compared to patients' mean overall direct
perspective score. This difference score was used in the multiple regression. As
discussed above, the potential range of scores for the seven-item instrument on this
variable is -63 to 63. For this variable, a positive value would suggest that pharmacists
rated patients' perceptions of the services higher than the patients actually did. A
negative value would suggest that pharmacists rated patients' perceptions of the
pharmacy services lower than the patients actually did, or in other words, that patients
perceived the services as more beneficial than pharmacists thought they would.
Pharmacist realization is the extent to which pharmacists realize that patients
understand them. It was derived by comparing pharmacists' meta-metaperspective to
patients' metaperspective. Pharmacists' individual responses were compared to mean
patient responses on the overall scale to calculate the difference score. Accordingly,
the potential range of scores is from -63 to +63. A positive value would suggest that
the pharmacist rated patients' perceptions of pharmacists' perceptions of the services
higher than the patients actually did, while a negative value would suggest that
pharmacists rated patients' beliefs about pharmacists' perceptions of the pharmacy
services lower than the patients actually did, or in other words, that patients perceived
that pharmacists would rate the services as more beneficial than pharmacists thought
Pharmacists' Feeling Understood
Pharmacists' feeling understood is derived by comparing pharmacists' meta-
metaperspective with pharmacists' own direct perspective and was calculated as a
difference score on the overall scale between these two perspectives for each
pharmacist. The potential range of scores for this variable is from -63 to +63. A
positive value would suggest that the pharmacist believed patients' perceptions of
pharmacists' opinion higher than the pharmacist actually did. On the other hand, a
negative value would suggest that the pharmacist believed patients perceived
pharmacists' answers lower than the pharmacists themselves actually did.
Descriptive information about the sample and the variables was examined.
Associations between the IPM-PC and PCQ questionnaires were examined. In
addition, associations between patients' perceptions of the benefits of pharmaceutical
care type services and their pharmacy patronage were explored. The following
analyses were carried out to test the hypotheses.
Hi: Pharmacists and patients agree on the types of pharmaceutical care type services
they perceive would be beneficial for patients.
A sum of the item scores for both the pharmacists' direct perspective ("I think it
would be beneficial for my patients if I were to discuss the patient's drug therapy with
him or her") and patients' direct perspective ("I think it would be beneficial if my
pharmacist were to discuss my drug therapy with me") on the IPM-PC questionnaire
was calculated. A T-test between means of the summated scale scores was conducted to
determine if pharmacists and patients disagree on the type of pharmacy services they
believe would be beneficial. Significant difference between the means would indicate
that pharmacists and patients disagree on the overall scale. According to the
Interpersonal Perception Method, a nonsignificant difference would mean that
pharmacists and patients do agree on the types of services they believe would be
beneficial. Alpha was set at 0.10 in order to reduce the risk of committing a Type II
H2,: Pharmacists misunderstand patients' perceptions of the benefits of pharmaceutical
care type services.
In order to test this hypothesis, the scores from the pharmacists'
metaperspectives ("My patients would say 'I think it would be beneficial if my
pharmacist were to discuss my drug therapy with me'") were compared to the patients'
direct perspectives ("I think it would be beneficial if my pharmacist were to discuss my
drug therapy with me"). An independent samples T-test was conducted between
pharmacists' and patients' mean scores. Alpha was set at 0.05 for this and the
remaining analyses. A significant difference would indicate that pharmacists
misunderstand patients' perceptions about the type of pharmacy services which patients
believe would be beneficial. According to the Interpersonal Perception Method, a
nonsignificant difference would indicate that pharmacists understand patients'
perceptions of the benefits of pharmaceutical care type services.
H2b. Patients misunderstand pharmacists' perceptions of the benefits of pharmaceutical
care type services.
In this case, patients' metaperspectives ("My pharmacist would say 'I think it
would be beneficial for my patients if I were to discuss the patient's drug therapy with
him or her'") were compared to pharmacists' direct perspectives ("I think it would be
beneficial for my patients if I were to discuss the patient's drug therapy with him or
her"). An independent samples T-test was used to compare patients' and pharmacists'
mean scores. A significant difference between the means would indicate that patients
misunderstand pharmacists' attitudes toward the types of services pharmacists believe
would be beneficial. According to the IPM, a nonsignificant difference would indicate
that patients understand pharmacists' perceptions of the benefits of pharmaceutical care
H3,: Pharmacists fail to realize that patients misunderstand their perceptions of the
benefits of pharmaceutical care type services.
To test this hypothesis comparisons between pharmacists' meta-metaperspectives
("My patients think I would say 'I think it would be beneficial for my patients if I were
to discuss the patient's drug therapy with him or her'") and patients' metaperspectives
("My pharmacist would say 'I think it would be beneficial for my patients if I were to
discuss the patient's drug therapy with him or her'") must be made. An independent
samples T-test was conducted to compare the means of the pharmacists' meta-
metaperspective to the patients' metaperspective score. A significant difference would
indicate that pharmacists fail to realize that patients understand (or misunderstand)
pharmacists' perceptions of the types of pharmacy services they believe to be
beneficial. A nonsignificant difference would indicate that, according to the IPM,
pharmacists realize that patients understand (or misunderstand) their perceptions of the
benefits of pharmaceutical care type services.
H3b: Patients fail to realize that pharmacists misunderstand their perceptions of the
benefits of pharmaceutical care type services.
In order to determine if patients realize or fail to realize that the pharmacist
understands (or fails to understand) their perceptions, a comparison between the
patients' meta-metaperspectives ("My pharmacist thinks I would say 'I think it would
be beneficial if my pharmacist were to discuss my drug therapy with me'") and the
pharmacists' metaperspectives ("My patients would say 'I think it would be beneficial if
my pharmacist were to discuss my drug therapy with me'") was made. Again, this
comparison was made using an independent samples T-test between patients' mean
meta-metaperspective score and pharmacists' mean metaperspective score. A
significant difference would indicate that patients fail to realize that pharmacists
understand (misunderstand) patients' perceptions. A nonsignificant difference would
indicate, according to the IPM, that patients realize that pharmacists understand (or
misunderstand) their perceptions of the benefits of pharmaceutical care type services.
H4,: Pharmacists feel misunderstood by patients with regard to pharmaceutical care
By comparing the pharmacists' meta-metaperspectives ("My patients think I
would say 'I think it would be beneficial for my patients if I were to discuss the
patient's drug therapy with him or her'") with the pharmacists' own direct perspectives
("I think it would be beneficial for my patients if I were to discuss the patient's drug
therapy with him or her"), the pharmacists' feeling of being understood was analyzed.
Pharmacists' mean direct perspective score was compared with pharmacists' mean
meta-metaperspective score using a paired samples T-test. A significant difference
would indicate that pharmacists feel misunderstood by patients in regard to
pharmaceutical care type services. Based on the IPM, a nonsignificant difference
would mean that pharmacists feel understood by patients with regard to pharmaceutical
care type services.
H4b: Patients feel misunderstood by pharmacists with regard to pharmaceutical care
Likewise, in order to analyze patients' feelings of being understood, patients'
meta-metaperspectives ("My pharmacist thinks I would say 'I think it would be
beneficial if my pharmacist were to discuss my drug therapy with me'") was compared
to patients' own direct perspective ("I think it would be beneficial if my pharmacist
were to discuss my drug therapy with me"). Patients' mean meta-metaperspective score
were compared with patients' mean direct perspective score by a paired samples T-test
and a significant difference would indicate that patients feel misunderstood by
pharmacists with regard to patient care services. Based on the IPM, a nonsignificant
difference would mean patients feel understood by pharmacists with regard to
pharmaceutical care type services.
H5: The level of pharmaceutical care type services reported by pharmacists will be
predicted by the level of pharmacists' agreement, level of pharmacists' understanding,
level of pharmacists' realization and level of pharmacists' feeling understood.
Pharmacists' self-report of pharmaceutical care type services was measured
using the PCQ. Scores were summed to yield an overall pharmaceutical care score that
served as the dependent variable. Multiple regression analysis was conducted utilizing
the difference scores of pharmacist agreement, pharmacist understanding, pharmacist
realization and pharmacist feeling understood as the independent variables used to
predict pharmacists self report of pharmaceutical care activities performed.
Subsequent analysis repeated the multiple regression utilizing PC-II as the
dependent variable. This dependent variable represents the pharmaceutical care
behaviors which are represented on the IPM-PC questionnaire. As in the previous
analysis, the regression was conducted utilizing the independent variables of pharmacist
agreement, pharmacist understanding, pharmacist realization and pharmacist feeling
Table 4-2. Item to Total Correlations and Coefficient Alpha of Pharmacist Pretest
Scale Item-Total Correlation Alpha if Item Deleted
A 0.57 0.87
B 0.69 0.86
C 0.54 0.87
D 0.11 0.87
E 0.60 0.86
F 0.35 0.87
G 0.46 0.87
H 0.74 0.85
I 0.45 0.87
J 0.79 0.85
K 0.53 0.87
L 0.70 0.86
M 0.62 0.86
N 0.47 0.87
A 0.65 0.89
B 0.68 0.89
C 0.71 0.89
D 0.07 0.91
E 0.71 0.89
F 0.72 0.89
G 0.33 0.90
H 0.53 0.89
I 0.85 0.88
J 0.87 0.88
K 0.85 0.88
L 0.65 0.89
M 0.51 0.90
N 0.41 0.90
A 0.81 0.96
B 0.87 0.96
C 0.85 0.96
D 0.83 0.96
E 0.87 0.96
F 0.80 0.96
G 0.69 0.96
H 0.81 0.96
I 0.84 0.96
J 0.79 0.96
K 0.76 0.96
L 0.74 0.96
M 0.77 0.96
N 0.77 0.96
Table 4-3. Item to Total Correlations and Coefficient Alpha of Patient Pretest Scales
Scale Item-Total Correlation Alpha if Item Deleted
A 0.52 0.90
B 0.45 0.90
C 0.47 0.90
D 0.22 0.91
E 0.62 0.89
F 0.48 0.90
G 0.71 0.89
H 0.79 0.89
I 0.66 0.89
J 0.74 0.89
K 0.70 0.89
L 0.73 0.89
M 0.57 0.89
N 0.60 0.90
A 0.61 0.97
B 0.69 0.96
C 0.82 0.96
D 0.71 0.96
E 0.83 0.96
F 0.89 0.96
G 0.80 0.96
H 0.86 0.96
I 0.80 0.96
J 0.86 0.96
K 0.85 0.96
L 0.86 0.96
M 0.82 0.96
N 0.80 0.96
A 0.79 0.98
B 0.89 0.98
C 0.89 0.98
D 0.86 0.98
E 0.88 0.98
F 0.86 0.98
G 0.90 0.98
H 0.88 0.98
I 0.89 0.98
J 0.89 0.98
K 0.86 0.98
L 0.87 0.98
M 0.88 0.98
N 0.84 0.98
Table 4-4. Interpersonal Perception Method-Pharmaceutical Care Items and
Pharmaceutical Care Domains
Interpersonal Perception Method-Pharmaceutical Care Pharmaceutical
Item Care Domain
A. Explain to patients how to use their Consultation
B. Tell patients about other drugs which may cause
problems with their medicationss.
C. Tell patients about possible side effects of their
D. Ask patients if they have any questions to ask.
M. Answer any questions or concerns patients may
E. Ask patients questions to find out how well they Assessment
think the medications) they are taking is/are
F. Ask patients questions to find out if they might be
having any problems with their medications.
I. Ask patients what they want to achieve by taking
J. Ask patients questions to figure out if their
goals) for their medications) are being reached.
G. Check patients' records for potential problems Screening
with their medicationss.
H. Carry out plans to solve (or prevent) problems Implementation
with patients' medications.
K. Design follow-up plans to measure patients'
progress toward their goals) for their
L. Carry out the follow-up plans to measure
patients' progress toward their goals) for their
N. Make sure that patients understand the Patient
information given to them. Understanding
Characteristics of the Sample
The final sample was comprised of 147 pharmacists practicing in the state of
Florida in the community setting. Table 5-1 (page 102) describes the pharmacist sample
and compares the sample with national estimates. When asked to describe their positions,
52 percent identified themselves as employee pharmacists, 39 percent as pharmacy
managers, 5 percent as pharmacy owners and 3 percent described their position as
"other." None of the pharmacists who indicated a different type of position indicated
what position they held. Most of the pharmacists surveyed held Bachelor of Science
degrees (85 percent), while several held both B.S. and Doctor of Pharmacy degrees (11.6
percent). Only 5.4 percent reported holding a Doctor of Pharmacy degree as the sole
professional degree. Pharmacists were also asked to indicate how long they had been in
practice. Twenty percent of the pharmacists had been in practice 1-5 years, 18 percent
between 6-10 years, 12 percent 11-15 years, 14 percent 16-20 years and 37 percent
reported practicing pharmacy for over 20 years. Males accounted for 65 percent of the
Pharmacists also were asked to characterize the number of prescriptions that were
dispensed in their pharmacies on an average day. Seventeen percent of pharmacists
reported that under 100 prescriptions were dispensed, 27 percent reported that an average
of between 100 and 299 prescriptions were dispensed daily, 27 percent reported that
between 200 and 299 prescriptions were dispensed, 15 percent between 300 and 399, 8
percent between 400 and 499, and 7 percent 500 and over. The pharmacists' practice
settings were also recorded. Thirty eight percent of pharmacists practiced in chain
pharmacies such as Eckerd's, 31 percent in independent pharmacies and 31 percent in
discount chain stores (such as WalMart). One pharmacist reported working at a health
maintenance organization (HMO) outpatient pharmacy. Information regarding the
distribution of pharmacists in the state of Florida was unavailable, so comparisons
between the sample and national figures were made. By comparison, national figures
show that 33 percent of pharmacists practice in independent community pharmacies and
33 percent practice in chain pharmacies (Martin, 1993). According to the study reported
by Martin (1993) 86 percent of pharmacists hold a Bachelor of Science degree in
pharmacy and the remaining 14 percent held either a Doctor of Pharmacy degree or both
degrees. The sample seems to reflect the national distribution of pharmacists with respect
to practice site, degree held, age and gender as found in the Pharmacy Manpower Project
described by Martin (1993).
The most recent list of registered pharmacists in Florida included 18,798 names,
of which 12,239 listed addresses in Florida. From those residing in Florida, 1,700
pharmacists were randomly selected for identification of their telephone numbers. A total
of 1,105 (65 percent) of the telephone numbers were identified from the most recent city
directories available and Database America, a national online database including home
address and telephone numbers. Only 947 of the 1,105 numbers were called when the
sample size of 150 was reached. Of those, 150 were nonworking numbers, 35 were non-
household numbers (business, cellular phone, fax number) and 64 were wrong numbers.
Due to language barriers between the interviewer and the person who answered the
telephone, 35 of the pharmacists were not reached. Twenty calls were not completed
because the individual was deceased. Thus, 643 pharmacists were asked to complete the
questionnaire. There were 205 (31.9 percent of calls) refusals, 291 (45.2 percent of calls)
pharmacists were ineligible because they worked in hospital or mail order pharmacies or
were no longer in practice and 150 pharmacists completed the questionnaire. Three of the
pharmacists who completed the interview did not meet eligibility criteria. Because they
practiced in the hospital setting, they were eliminated from the sample. Thus, 147
eligible pharmacists completed the questionnaire. The response rate of eligible
pharmacists, assuming that all refusals came from eligible pharmacists, was 41 percent.
This is the worst case estimate for the response rate since it is unknown what percentage
of the pharmacists who refused to complete the questionnaire were, in fact, eligible for
the study. The mean length of pharmacist interviews was 9.5 minutes (s.d.=2.7).
Patient Sample Characteristics
The final sample of patients included 151 consumers throughout Florida who had
filled at least one prescription during the previous six months. Table 5-2 (page 103)
summarizes the characteristics of the patient sample in comparison to the 1990 census
report (U.S. Department of Commerce Bureau of the Census, 1990).
Females accounted for two-thirds of the respondents (67 percent). According to
the 1990 census, 52 percent of the population of Florida is female. The higher percentage
of female respondents may be accounted for by the fact that women tend to complete
questionnaires more often than men. Women also utilize health care, including
prescriptions, more often than men (Cockerham, 1992). Most of the patients were under
50 years of age (66.3 percent). Approximately one-fourth of all patients (23 percent)
were between the ages of 18 and 29 years, one-fifth were between 30 and 39 years (22
percent), one-fifth between 40 and 49 years (21 percent). Of those over the age of 50, 10
percent were between the ages of 50 and 59, 13 percent were between 60 and 69, and 11
percent were 70 years of age and older. This is consistent with the age distribution
throughout the state of Florida. In Florida, approximately 23 percent of people are
between the ages of 18 and 29, 20 percent between 30 and 39 years, 15 percent between
40 and 49 years, 12 percent between 50 and 59 years, and 30 percent over the age of 60
years (U.S. Census Report, 1990).
When asked to describe their education level, 34 percent reported having
completed some college, 36 percent reported being college graduates (including post-
baccalaureate degrees), 19 percent reported being high school graduates and 9 percent
reported less than a high school education. By comparison, in the state of Florida, 25
percent have less than a high school degree, 30 percent have completed high school, 27
percent have completed some college, 10 percent have completed college and 9 percent
have a professional degree or have completed at least some graduate school. Seventeen
percent of the patients reported annual household income below $20,000.
Twenty five percent reported income between $20,000 and $39,999, 20 percent
reported income between $40,000 and $59,999, 11 percent between $60,000 and $79,999
and 8.6 percent reported income in excess of $80,000. Many patients declined to answer
this question (19 percent). The categories used in this study to measure income are not
consistent with the categories used in the U.S. census, however, for comparison, the
census figures are included here. In the state of Florida, 15 percent of the population
reports earning less than $10,000 annually; 30 percent report incomes of between $10,000
and $24,999; 34 percent report household incomes between $25,000 and $49,999; and 21
percent report annual household incomes over $50,000. Comparison figures for the
population are shown in Table 5-2 (page 103). The sample seems to reflect a higher level
of education than the overall population in Florida, although based on age this sample
seems to reflect Florida's population.
In describing their pharmacy patronage, almost half of patients reported
patronizing a large discount chain store, such as WalMart, for their prescriptions (47
percent). Another third reported patronizing smaller pharmacy chains such as Eckerd's
(33 percent). Nine percent reported using an independent store, 6 percent used a grocery
store and 5 percent reported using other types of pharmacies (such as an HMO
pharmacy). This distribution of pharmacy patronage is comparable to national figures
from the 1993 National Prescription Buyers Survey described by Meade (1994). Most
patients reported patronizing only one pharmacy on a regular basis (72 percent), 17
percent reported using two pharmacies, and 3 percent reported using three pharmacies on
a regular basis. This, too, is consistent with other findings. According to Meade (1994),
75 percent of patients go to only one pharmacy to have their prescriptions dispensed.
Most patients did not know the name of the pharmacist who usually fills their
prescriptions (82 percent).
Finally, patients were asked to list what factors they considered when choosing a
pharmacy. The most frequent response included insurance requirements (17 percent),
location or convenience (38 percent) and price (17 percent). Customer services including
friendliness (10 percent), hours (4 percent), knowledgeable pharmacists (3 percent) and
efficiency (3 percent) were also named. One patient reported that the doctor had
recommended a particular pharmacy. These figures vary slightly from those described by
Meade (1994). By comparison, only 15 percent of those patients mentioned using a
designated pharmacy, 28 percent reported location, 22 percent mentioned price and 20
percent named characteristics of pharmacists or staff. See Table 5-3 (page 104) for a
description of the sample's pharmacy patronage.
Patients were also asked questions concerning their medication use including how
many prescription medications were taken on a regular basis and how many new and
refill prescriptions they had dispensed in the previous six months. In response to the
number of prescription medications taken on a regular basis, answers ranged from zero to
nine. Approximately one-third of the respondents used one prescription medication on a
regular basis (34 percent). Twenty-seven percent reported not regularly using any
prescription medications. Fifteen percent reported using two prescriptions on a regular
basis while 24 percent use three or more prescriptions on a regular basis.
When asked about number of new prescriptions in the previous six months, the
range of responses was from zero to 12 with a mean of 1.7 (s.d. = 2.0). Patients reported
a range of zero to 48 refill prescriptions dispensed over the previous six months (mean
4.1 6.6). Twenty-five percent of respondents had no refill prescriptions dispensed
over the past six months, 19 percent had one, 17 percent had 2. Twelve percent of
patients had between 3 and 5 prescriptions dispensed while 15 percent had 6 refills.
Thirteen percent of patients had over six prescriptions refilled in the previous six
When asked to describe their interaction with their pharmacist, 43 percent of
patients reported that their pharmacist usually volunteers information when they have a
prescription dispensed. This is consistent with the national figure (43 percent) reported
by Meade (1994). Twenty-two percent of patients reported that their pharmacist
usually asks if they are having any problems when they receive refill prescriptions.
This is higher than the percentage reported (10 percent) for the control group of a study
on the effects of an education program for pharmacists to help them detect drug-related
problems in elderly patients (Kimberlin et al., 1993).
A total of 744 telephone numbers were called. Of those, 251 were either
nonworking numbers or were businesses rather than households and 35 calls were not
completed due to a language barrier. Thus, 485 households were contacted for consent
to participate in the interview. There were 250 refusals (51.5 percent of calls) and 57
households (11.7 percent of calls) were ineligible, meaning there was no one in the
household over the age of 18 who had a prescription dispensed in the previous six
months. Although it is unknown what percent of patients who refused to complete the
questionnaire were eligible, 151 patient interviews were completed yielding a response
rate of 35 percent in the worst case assumption that all refusals were eligible. The
mean length of patient interviews was 7.2 minutes (s.d. =3.0).
Factor analysis was used to establish the construct validity of the pharmacist and
patient scales. The factor loadings for pharmacist and patient scales are presented in
Table 5-4 (page 104) and Table 5-5 (page 105) respectively. Separate analyses of
pharmacists' direct, metaperspective and meta-metaperspective responses (Table 5-4,
p. 105) indicated that at each level of perception, all the items were representative of
one factor. This is inconsistent with the theoretical development of the questionnaire
and will be discussed further in the final chapter.
Results from analysis of patients' direct and metaperspective responses (Table
5-5, p. 106) showed that the items loaded on two factors. Based on direct perspective
results, only one item seems to be representative of Factor 2, while the other six items
loaded on Factor 1. Item G ("Tells me about other drugs which may cause problems
with my medicationss)) loaded more highly on Factor 2--"Discuss problems" than on
Factor 1--"Counseling." This suggested that the item represented a unique factor and
should not be included in the summated scores with the other items. Analysis of the
metaperspective showed that item G loaded on both factors, but more highly on Factor
2--"Discuss problems." Item D ("Make sure I understand the information given to
me") also loaded on Factor 2--"Discuss problems," but loaded more highly on Factor
1--"Counseling." Factor analysis of patients' meta-metaperspective revealed only one
factor for all seven items. Crocker and Algina (1986) suggest that factor loadings less
then 0.30 are usually considered unimportant, thus a loading of 0.30 was considered the
criteria for inclusion in the scale score. All of the loadings for the items in Factor 1
were found to be greater than 0.30 in each of the perspectives, so all items were thus
kept in the scale.
Internal consistency reliability of the summed scales was calculated using
Cronbach's coefficient alpha. The overall reliability estimates for the pharmacists' and
patients' direct perspective scales were 0.87 and 0.79 respectively (see Table 5-6, page
107 and Table 5-7, page 108). Corrected item-total correlations on the pharmacists'
scale ranged from 0.54 to 0.81, while patients' ranged from 0.22 to 0.73. Item G
("Tell me about other drugs which may cause problems with my medicationss))
showed the lowest item to total correlation on the patients' scale (0.22), which is
consistent with the results of the factor analysis. The overall reliability of the scale was
at least 0.80 and was considered high .
Coefficient alpha estimates for the metaperspective and meta-metaperspectives
scales are also presented in Tables 5-6 (p. 107) and 5-7 (p. 108). Coefficient alpha for
pharmacists' metaperspective and meta-metaperspective scales were 0.90 and 0.93
respectively. Coefficient alpha for patients' scales were 0.86 and 0.91 respectively.
All of the item to total correlations were greater than 0.50 with the exception of item G
on patients' metaperspective scale. This was consistent with results from the direct
perspective scale. Thus, item G was eliminated from all summated scales and was
analyzed as an independent item. Coefficient alpha for the revised scales are also
presented in Tables 5-6 (p. 107) and 5-7 (p. 108). There was little change in the
coefficients after the deletion of item G. Coefficient alpha for patients' direct
perspective scales increased from 0.79 to 0.80, metaperspective increased from 0.86 to
0.87, and for the meta-metaperspective coefficient alpha remained the same (0.91).
Coefficient alpha for pharmacists' direct and metaperspective scales decreased slightly
from 0.87 and 0.90 to 0.86 and 0.89 respectively. Pharmacists' meta-metaperspective
scale showed no change in coefficient alpha (0.93).
Distributions of items in the IPM and PCQ were examined. Scores covered the
full range and indicated variability in response to items. Overall, pharmacists rated the
provision of the pharmaceutical care type services as potentially more beneficial than
patients. Means of pharmacists' and patients' responses to each item are presented in
Table 5-8. Their direct perspective responses provide an indication of how potentially
beneficial each group perceives the services to be. On the scale from one to ten, where
ten represented strong agreement with the potential benefit of the service described,
pharmacists consistently rated the items toward strong agreement. The range of mean
pharmacists responses on different items was from 7.7 (s.d. =2.4) to 9.4 (s.d.= 1.3).
This suggests that pharmacists perceive that the services would be of benefit to patients.
Pharmacists believed that the most beneficial services would be to explain to patients
how to use their medications (9.4 1.3), to make sure patients understand the
information given to them (9.2 1.5) and to tell patients about other drugs which may
cause problems with their medications (9.1 1.5). Asking patients questions to find out
if they might be having any problems with their medications (8.6 1.9), carrying out
plans to solve or prevent problems with patients' medications (8.2 2.1) and asking
patients questions to figure out if their goals for the medications are being reached (8.1
2.3) were also seen as very beneficial. Designing and carrying out follow up plans to
measure patients' progress toward their goals for their medications was seen as the least
beneficial service which could be offered to patients (7.7 2.4).
Results from pharmacists' report of their pharmaceutical care activities indicated
that pharmacists provided a wide range of services. Summed scores for the level of
services provided ranged from 0 to 90, covering the entire range of potential scores. The
mean score was 40.0 (s.d.=22.9). Table 5-9 (page 110-111) displays the means and
standard deviations for each service. Compared to Badejogbin's (1994) study, it seems
that pharmacists are providing pharmaceutical care services to more patients. The overall
mean score was reported by Badejogbin (1994) to be 32.9 (s.d.=23.2). The mean score
for the documentation domain (range 0 to 30) was 13.8 (s.d.=12.2), slightly higher than
Badejogbin's finding (1994) of 10.6 (s.d.=7.2). Therapeutic objective implementation
and patient assessment were provided to fewer patients. The mean for the domain of
patient assessment (range 0 to 30) services was 9.2 (s.d.=10.0). Pharmacists reported
providing therapeutic objective implementation (range 0 to 15) for 5.7 (s.d.=6.0) patients.
Badejogbin's findings (1994) showed pharmacists' mean scores in these domains to be 6.9
(s.d.=7.0) and 4.6 (s.d.=4.2), respectively. Pharmacists reported providing screening of
patients' record often, providing it to an average of 3.9 (s.d.=1.8) patients of the last five.
Verification of patient understanding and patient consultation were also provided to more
than half of patients. Pharmacists reported consulting with an average of 3.0 (s.d.=1.9)
patients of the last five, and reported verifying the understanding of an average of 3.2
(s.d.=1 .9) patients which is consistent with the level reported by Badejogbin (1994).
These means are comparable with earlier findings, which reported means of 4.4
(s.d.=1.4), 3.1 (s.d.=1.7) and 3.3 (s.d.=1.8) respectively. It seems that pharmacists report
providing more pharmaceutical care services than in 1994. However, the types of
services being reported seem to be consistent with those reported by the pharmacists
surveyed in 1994 (Badejogbin).
Association Between IPM and PCQ
Each of the items on the IPM-PC had a corresponding item on the PCQ.
Associations between pharmacists' perceptions of the benefits of the services and the
number of patients for whom they provided that service was calculated to examine the
extent to which pharmacists engaged in behaviors they believed would be beneficial for
patients. Table 5-10 (page 112) shows the correlations between these items. Of the seven
items, three were found to be significantly correlated with their corresponding behavior
measure. A significant positive correlation was found between the items "it would be
beneficial for my patients if I were to make sure that they understand the information
given to them" and the number of patients for whom pharmacists reported that they
verified that the patient understands the information presented to him or her (r=0.41,
p<0.001). The item "it would be beneficial for my patients if I were to carry out plans to
solve (or prevent) problems with their medications," was found to be positively
associated with the number of patients for whom pharmacists reported that they
implemented a strategy to resolve (or prevent) the drug related problem (r=0.24, p<0.01).
Finally, a significant positive correlation was found between the item "it would be
beneficial for my patients if I were to tell them about other drugs which may cause
problems with their medications," and the number of patients for whom pharmacists
reported that they discussed the patient's drug therapy with him or her (r = 0.27, p<0.001).
Although none of the correlations was large, they were in the direction expected, in that
pharmacists who rated the item as more beneficial for patients did in fact report that they
provided the service to more patients. The remaining four items were not found to be
significantly associated with pharmacists' reported behavior.
The three measures of behavior found to be associated with the corresponding
PCQ item were also found to be associated with other IPM-PC items (see Table 5-10).
The item "discuss the patient's drug therapy with him or her" was found to be
significantly correlated with five of the items (items C (r=0.22, p<0.01), D (r=0.33,
p<0.001) E (r=0.22, p<0.01), F (r=0.21, p<0.01), G (r=0.27, p<0.001)) representing the
domains of assessment, verification of understanding and one of two items representing
consultation and implementation of therapeutic plans. The item "verify that the patient
understands the information presented to him or her" was found to be associated with its
PCQ counterpart (r-=.41, p<0.001) and also the item "tell me about other drugs which may
cause problems with my medicationss" (r=0.21, p<0.01) from the domain of
consultation. Finally, the item "implementing a strategy to resolve (or prevent) the drug
related problems" was found to be correlated with the items representing the domain of
consultation (items A (r=0.25, p<0.01) and G (r=0.27, p<0.01)), the one item representing
verification of understanding (item D, r=-0.25, p<0.01) in addition to its corresponding
PCQ item. The summated score of pharmacists' direct perspective scores were found to
be positively associated with overall score on the PCQ (r=0.22, p<0.01).
Correlations between pharmacists' metaperspective scores for individual items on
the IPM-PC and reported behavior on the PCQ were calculated. As can be seen in Table
5-10, none of these correlation was found to be significant. Additionally, correlations
between pharmacists' meta-metaperspective scores and reported behavior were calculated
(Table 5-10, p. 112). The item in which pharmacists responded to the extent that their
patients would say "my pharmacist would say 'I think it would be beneficial for my
patients if I were to make sure that they understand the information given to them'" was
found to be correlated with the number of patients for whom pharmacists reported
performing this service (r=0.24, p<0.01).
Summary of Difference Scores
Difference scores for each pharmacist were calculated to determine the
pharmacists' agreement with patients, pharmacists' understanding of patients, pharmacists'
realization that patients misunderstand them and pharmacists' feeling understood. The
mean difference score for pharmacists' agreement with patients was 5.2 (s.d.=8.9)
Approximately 20 percent (N=30) of pharmacists had negative difference scores, while
80 percent had positive difference scores (N=l 17). This suggests that most pharmacists
perceived the pharmaceutical care services to be more beneficial than patients did. The
difference scores for pharmacists' understanding of patients showed that 30 percent
(N=44) of pharmacists thought patients would perceive the services as less beneficial than
they did. On the other hand, 70 percent (N=103) believed that patients rated the services
as more beneficial than they actually did. The mean understanding difference score was
3.0 (s.d.= 10.7). The results of pharmacists' realization that patients misunderstand them
were similar to those of pharmacists' understanding. Negative difference scores, which
indicated that pharmacists' meta-metaperspective scores were lower than patients'
metaperspective scores were found for 29 percent of pharmacists (N=43). Positive scores
were found for the remaining 71 percent (N= 104), which indicated that most pharmacists
believed patients rated pharmacists' perceptions higher than patients did. Overall, the
mean difference score for pharmacists' realization was 3.2 (s.d.=10.2). Finally, the mean
difference score for pharmacists' feeling understood was -1.0 (s. d.=9.1). Negative scores
accounted for 42 percent (N=62) of the scores and 58 percent (N=85) were positive.
Thus, more than half of pharmacists thought patients would rate pharmacists' perceptions
higher than pharmacists' actual perceptions.
Patients' direct perspective responses on the potential benefits to them of
pharmaceutical care services were consistently lower than pharmacists'. Means ranged
from 6.1 (s.d.=2.4) to 9.4 (s.d.=1.6). Patients believed that being told about other drugs