Physicians' current and future expectations of and
current experiences with pharmacists were studied.
A three-part questionnaire was mailed to 2600 practicing physicians in
California, including office-based practitioners, hospital-employed
(nonhouse staff) physicians, and medical residents. A 5-point Likert scale
was used to measure respondents' level of agreement with statements in
each part; part 1 dealt with current expectations, part 2 with current
experiences, and part 3 with future expectations. Respondents were asked
to confirm their practice setting (office-based, hospital-based, medical
resident, or other) and to indicate the number of years since graduation
from medical school.
The mean values of physician agreement with each statement were
reported. There was neither strong agreement nor strong disagreement with
any statement regarding physicians' current and future expectations of and
current experiences with pharmacists. There was no correlation between
practice setting and level of agreement with statements in any part of the
questionnaire. There was a correlation between the number of years since
graduation from medical school and the level of agreement with statements
about current and future expectations; in both cases those physicians who
graduated from medical school less than 10 years ago had higher
expectations of pharmacists than those who graduated 10 or more years ago.
Overall, physicians do not know what to expect from pharmacists.
Physicians expectations of pharmacists were most strongly correlated with
the number of years since graduation from medical school and least
strongly correlated with physicians' practice settings.
For centuries, pharmacy has supported medicine --
pharmacists have provided services for physicians' patients. In the
historical model, physicians diagnosed and prescribed, while pharmacists
compounded and dispensed medications. This model continues to exist today,
although with some notable variations, and undoubtedly serves as the basis
for what many physicians, and a large segment of the general public,
expect of pharmacists today.
Given that prescribing is still almost exclusively in the physician's
domain, the utility of pharmacists' input about managing drug therapy
ultimately depends on physicians' willingness to accept it. It is
important, therefore, that pharmacists understand what physicians expect
of them and how receptive physicians are to the contributions that
pharmacists seek to make to patients' pharmaceutical care.
Hynniman and Lamy and Williamson and Kabat
found that, although a majority of physicians are receptive to an expanded
role for pharmacists as drug information experts, most physicians'
believed that very few pharmacists were performing this service in any
meaningful way. Their expectations of pharmacists were therefore quite
In addition, several investigators found that physicians are receptive
to specific clinical services provided by pharmacists, such as therapeutic
drug monitoring, patient counseling, and recommendations for therapeutic
agents.[3-12] The level of physician acceptance of these
services was found, in several of these studies, to be related to the
degree of exposure physicians have to pharmacist-provided services.
Several other investigators found that physicians are generally receptive
to a wide range of clinical pharmacy services provided in an inpatient
setting, if provided in a consultative or supportive
role.[13-18] Most physicians were found to oppose autonomous
decision-making responsibility for pharmacists. Sulick and
Pathak found that, although physicians are receptive to the
consultative and educational services provided by pharmacists in an
inpatient setting, the receptivity by individual physicians to those
services is related to the value the physician attaches to the service and
the physician's perception of the pharmacist's competence. Bradshaw and
Doucette found that there is good support among physicians
for the provision of certain patient-oriented services by community
pharmacists, including drug therapy recommendations, patient monitoring,
and patient education; physicians were less favorable toward independent
decision-making roles for pharmacists, such as adjusting drug therapies
under protocols and changing drug dosage forms. Similar findings were
reported by Ranelli and Biss.
Bailie and Romeo found that most physicians favor
patient counseling by pharmacists but are opposed to pharmacist
prescribing, even limited prescribing under protocol. They also found
little support among physicians for the provision of health-screening
services by pharmacists. Two studies in the United Kingdom found similar
information. [23,24] Segal and Grines found that
opposition to pharmacist prescribing is reflected in the official policy
of state medical associations. However, according to a newspaper report,
the American Medical Association (AMA) supports pharmacy's efforts to
provide disease management services.
Hirsch et al. found that physicians expect pharmacists
to provide patient medication counseling, but pharmacists are not
providing this service to the extent assumed by physicians. Campbell and
Saulie and Sardinha found that there is
strong acceptance of pharmacists by physicians in group practice settings
in which pharmacists participate in collaborative patient care.
Physicians' attitudes toward expanded roles for pharmacists are
reflected from time to time in the medical literature. Several such
reflections have openly questioned pharmacists' competence to provide
direct patient care services, such as patient education and drug therapy
recommendations. [30-33] Although such comments may be viewed
as isolated opinions, they cannot be entirely discounted.
In a historical review of the relationship between pharmacists and
physicians, Cowen notes that, although conflict has existed
between the two professions throughout the centuries, considerable
progress toward harmony has been made. He attributes this, in part, to the
increasing dependence of physicians on pharmacists for drug-related
These studies and commentaries suggest that physicians' attitudes
toward and acceptance of pharmacists' clinical services have gradually
improved over the past 30 years, physician acceptance of pharmacists'
clinical services is related to the level of exposure physicians have had
to those services, and physicians are generally receptive to consultative
and educational services provided by pharmacists but are opposed to
pharmacists having independent decisionmaking authority.
Questions not answered by these studies include the following:
What do physicians expect of pharmacists? Most of the studies
cited address physicians' attitude toward, or acceptance of, specific
services provided by pharmacists. They do not help us understand what
physicians generally expect pharmacists to do.
Does the current level of pharmacy practice match physicians'
expectations? We assume that many physicians have developed certain
expectations of pharmacists on the basis of their experiences with
specific pharmacists; however, these expectations may not be met by all
What do physicians expect pharmacists to do in the future? It
is not known if physicians have any particular expectations of what
pharmacists will do in the future. It is conceivable that physicians
could have fairly limited expectations of pharmacists today while
holding higher expectations of what pharmacists will do in the future.
It is also conceivable that physicians have no expectations about
pharmacists' duties in the future beyond those they have today.
Our study attempts to answer these questions.
A three-part questionnaire was developed for
administration to a randomly selected group of physicians in California.
Each part included a set of statements for which respondents were asked to
indicate their level of agreement using a 5- point Likert scale where 1 =
strongly disagree, 2 = disagree, 3 = neutral, 4 = agree, and 5 = strongly
agree. In addition, one question required a "yes" or "no" response. Part 1
measured physicians' current expectations of pharmacists, part 2 measured
their actual experience with pharmacists, and part 3 measured their future
expectations of pharmacists. Demographic information collected included
the respondents' practice category (office-based, hospital-based [nonhouse
staff], or medical resident [house staff]) and the number of years since
graduation from medical school. The questionnaire was validated for
content by a panel of five practicing physicians.
This study was conducted from March 1998 through June
1998. The questionnaire was mailed to 2600 physicians in California who
were randomly selected from lists maintained by the AMA. The practicing
physicians on these lists were not limited to AMA members. The breakdown
of questionnaire distribution was as follows: 2000 to physicians listed as
office-based practitioners, 400 to physicians listed as hospital-based
practitioners (excluding radiologists, pathologists, and other nondirect
patient care specialists), and 200 to medical residents. The questionnaire
allowed respondents to confirm their practice category; one additional
choice was provided for "other."
A postcard was sent to all physicians 10 days before the first
questionnaire was mailed to acquaint them with the nature of the study and
to let them know to expect the questionnaire. Three separate mailings of
the questionnaire, each with an appropriate cover letter, were sent out at
three-week intervals. The second and third mailings were sent only to
those who had not responded to previous mailings. Control numbers on the
postage-paid return envelopes allowed us to track those who had responded.
Returns were accumulated in batches, and the envelopes in a particular
batch were opened by a secretary once they had been checked off, thus
protecting the anonymity of individual responses.
The data were first analyzed descriptively,
computing frequencies, means, and standard deviations for each item on the
questionnaire. Second, internal consistency reliability analysis and
confirmatory factor analysis were used to evaluate the psychometric
properties of the three measurement scales. Finally, multivariate analysis
of variance (MANOVA) was used to examine differences among physician
practice settings and years since graduation for the three dependent
variables (i.e., current expectations, current experiences, and future
expectations). Internal reliability was measured for each part of the
The first MANOVA was conducted between the independent variable (years
since graduation from medical school) and the dependent variables (current
expectations, current experiences, and future expectations). A second
MANOVA was conducted between the independent variable (practice setting)
and the dependent variables (current expectation, current experiences, and
future expectations). The dependent variables were the harmonic means of
the items in part 1 (current expectations), part 2 (current experiences),
and part 3 (future expectations). Significance was determined by the F
test with analysis of variance tests. Post hoc comparisons were made with
Tukey's test. The level of significance was set at 0.05. Statistical
analysis was performed with SPSS version 9.0 (SPSS, Chicago, IL).
Sixty-five (2.5%) questionnaires were returned as
undeliverable, leaving 2535 that were assumed to have been delivered.
Another 130 (5.1% of those assumed delivered) were returned by recipients
who did not complete them, either because they had retired or were in
areas of practice that did not involve direct patient care. The remainder
(2405) was used as the denominator when calculating percent response; 463
usable questionnaires were returned, representing a response rate of
Reliability (coefficient alpha) was estimated for each of the three
dependent variables. The highest coefficient (0.903) was found for the
"current expectations" scale. The remaining two scales ("current
experiences" and "future expectations") resulted in reliability estimates
of 0.877 and 0.820, respectively.
The results of part 1 of the questionnaire are shown in Table 1. There
were only three items (4, 12, and 16) in part 1 for which the mean
response was greater than 4 (agree). For one item (item 11), the mean
response was less than 3.
The results of part 2 of the questionnaire (current experiences) are
shown in Table 2. There was one item in this set (item 1) with a mean
value greater than 4. For three items, the means were between 3 and 4. The
means for five items were less than 3. Less than one fourth (23.2%) of
respondents indicated that they had a prior-approval relationship with
pharmacists for adjustments in patients' medications.
The results of part 3 of the questionnaire (future expectations) are
shown in Table 3. Responses to each of the five items in this part were
close to 3, and there was neither strong agreement nor strong disagreement
with any of these statements regarding physicians' future expectations of
Table 4 shows the breakdown of respondents by practice setting and by
years since graduation from medical school and the MANOVA results to
determine significance of the three dependent variables. No significant
difference between the physicians' practice settings and their current
expectations, current experiences, or future expectations was revealed
(Wilks' lambda = 0.975, p = 0.364). However, a significant
difference did exist between the years since graduation from medical
school and both current and future expectations of pharmacists (Wilks'
lambda = 0.965, p = 0.030).
We now return to our original questions that were not
answered by previous studies.
What do Physicians Expect of Pharmacists?
Most of the mean values
in Table 1 were between 3 and 4, suggesting that (1) physicians have no
strong expectations that pharmacists will provide any of the services or
reflect any of the behaviors described in this part of the questionnaire,
and (2) they have no strong negative expectations. This might be
interpreted to suggest that physicians are not sure what they expect of
There was good agreement with item 4 ("I expect pharmacists to be
knowledgeable drug therapy experts"). Considering the responses to other
items, however, it appears that physicians have no clear expectations of
how pharmacists apply that knowledge in practice.
Respondents differed in their expectations of hospital pharmacists and
community pharmacists in designing drug treatment plans. The mean level of
agreement with item 5 (pertaining specifically to hospital pharmacists)
was 3.60; the level of agreement with item 11 (pertaining specifically to
community pharmacists) was 2.57. On the other hand, they indicated a lower
degree of expectation that hospital pharmacists would provide patient
medication education than would community pharmacists (means, 3.81 and
There was a slightly higher expectation that hospital pharmacists would
monitor patients' response to drug therapy (items 7 and 13) and that they
would know the specific indications for each drug prescribed (items 8 and
14). In all cases, however, the mean values were between 3 and 4.
Respondents indicated that they do expect community pharmacists to
assist patients in selecting appropriate nonprescription drugs (mean,
4.17). There was no corresponding statement in the instrument concerning
physicians' expectations of hospital pharmacists.
The physicians' practice settings did not seem to have a bearing on
their current expectations of pharmacists. However, younger physicians
(those who graduated within the past 10 years) appeared to have higher
expectations of pharmacists than those who graduated more than 10 years
ago. This may reflect the changing level of pharmacy practice in more
recent years in those settings where physicians do their residency
Does the Current Level of Pharmacy Practice Match Physicians'
Part 2 of the questionnaire measured physicians' actual
experiences with pharmacists, regardless of what they may expect of
pharmacists. Responses to these items provided an indication of the
current level of pharmacy practice.
Item 1 was the only item in this set with a mean value greater than 4
(agree). This undoubtedly accounts for the respondents' strong expectation
that pharmacists are knowledgeable drug therapy experts (Table 1).
However, the mean value for item 2 (related to clinical drug information)
was somewhat lower. Respondents' experience, then, is that pharmacists do
a better job providing nonpatientspecific drug information than providing
information tailored to individual patients' clinical situations.
Items 3 (patient counseling) and 4 (problems with prescriptions) had a
mean value between 3 (neutral) and 4 (agree). With regard to patient
counseling, pharmacists appear not to be measuring up to physicians'
expectations. The mean value for respondents' experience with pharmacists
in this area is 3.47. This is somewhat lower than their expectation
(means, 3.81 and 4.12 for hospital pharmacists and community pharmacists,
The mean value for items 5-8 and 10 were between 2 (disagree) and 3
(neutral). This suggests that, based on respondents' experiences,
pharmacists perform these activities less than half of the time. This
finding is somewhat surprising with respect to the response to items 5 and
6 (Table 2). It may suggest a lack of alignment of incentives between
pharmacists and payers. The finding may also suggest that pharmacists'
workload does not permit them to routinely follow up with prescribers on
such matters. In either event, the finding was surprising, given our
impression that most pharmacists devote inordinate amounts of time to
contacting physicians to change prescriptions to less costly alternatives.
The responses to item 7 (clarification of drug therapy objectives) were
difficult to interpret. They may suggest that drug therapy objectives are
usually sufficiently clear to pharmacists, or that pharmacists neglect to
clarify the physician's intent.
For item 8 (problems with patients' medications), respondents'
experience with pharmacists may not fulfill the expectations they have of
them in this area. The mean value for actual experience in this area was
2.48; the mean value for the expectation that pharmacists will provide
such information was 3.51 for hospital pharmacists and 3.15 for community
pharmacists (Table 1).
Less than one fourth (23.2%) of respondents indicated that they had a
prior-approval relationship with pharmacists for adjustments in patients'
medications (Table 2). The fact that this percentage is so small may mean
that there is an opportunity for greater participation by pharmacists in
collaborative drug therapy management. It may also be related, in a number
of ways, to the relatively low level of expectations physicians have of
pharmacists, as reflected in Table 1.
Perhaps one of the more damaging findings of our study is the response
to item 10 (regarding pharmacists taking responsibility for resolving
problems). The mean response was 2.94. This was considerably below the
mean for physicians' expectations (mean, 3.53) (Table 1). We think it is
likely that physicians' experience with pharmacists' lack of willingness
to take "ownership" of patients' drug-related problems lies at the base of
their limited expectations of pharmacists.
What Do Physicians Expect Pharmacists to Do in the Future?
in part 3 measured respondents' support for having pharmacists routinely
provide three specific services in the future: "adjust the drug therapy
for some of my patients, using protocols that I have approved"; "monitor
and document my patients' response to drug therapy"; and "be on-call to my
patients for questions about their medications." The response to the first
two items was slightly below 3 (neutral), indicating that physicians are
leaning away from having pharmacists perform these services in the future.
The response to the third item was more positive (mean, 3.42). However, in
all three cases the responses indicated neither strong opposition to nor
strong support for these particular pharmacist-provided services.
It was interesting to note that respondents did not favor pharmacists
having a greater share of financial or legal responsibility for drug
therapy in the future (items 2 and 3; means, 2.68 and 2.85, respectively).
One possible interpretation of these results is that physicians are
opposed to pharmacists having any greater degree of personal relationship
with patients than they have currently.
It appears that physicians do not know what to expect of pharmacists.
It is clear that there is a gap between the theory of practice espoused by
pharmacy and the current reality of practice, especially when viewed
through the eyes of physicians. It appears that pharmacists need to
improve their credibility with physicians by accepting greater personal
responsibility for patients' drug therapy outcomes.
The most encouraging finding of our study is that physicians are not
strongly opposed to having pharmacists more directly involved in patient
care. For pharmacists who have encountered strong physician opposition in
the past, this seems to be a very positive statement. On the other hand,
our study shows that physicians are not yet clamoring for greater
contributions by pharmacists.
The fact that younger physicians have higher expectations of
pharmacists than do their older colleagues suggests an encouraging
prospect for pharmacy's future. We point out again, however, that
respondents' current experience with pharmacists does not, in most cases,
meet their expectations, suggesting that pharmacists will need to close
the gap between what physicians perceive pharmacists do and what they
expect them to do in order to gain future physician support for a more
responsible role in patient drug therapy management.
Limitations of Study
The results of this study cannot be
generalized to all physicians, since the sample was limited to physicians
in California. Also, there may be a potential response bias because of the
19.3% response rate. Despite these limitations, the large sample size (n =
2600) leads us to believe that these results provide useful insight into
physicians' experiences with and expectations of pharmacists.
Overall, physicians do not know what to expect of
pharmacists. Physicians' expectations of pharmacists were most strongly
correlated with the number of years since graduation from medical school
and least strongly correlated with physicians' practice setting.
Hynniman CE, Lamy PP. Physicians view the pharmacist. J Am Pharm
Assoc. 1971; NS11:158-63,180.
Williamson RE, Kabat HF. Pharmacistphysician drug communications.
J AmPharm Assoc. 1971; NS11:164-7.
Silverman HM, Simon GI. Physician reaction to a clinical pharmacist.
Am JHosp Pharm. 1971; 28:361-3.
Wallace D, Kradjan W. Physicians' opinions of pharmacists as
dispensers of patient medication information. J AmPharm Assoc.
Nelson AA Jr, Meinhold JM, Hutchinson RA. Changes in physicians'
attitudes toward pharmacists as drug information consultants following
implementation of clinical pharmaceutical services. Am JHosp
Pharm. 1978; 35:1201-6.
Taylor AT, Martell PH. Physician acceptance of clinical pharmacy
service in a skilled nursing home. J Am Geriatr Soc. 1980;
Moss RL, Garnett WR, Steiner KC. Physician attitudes toward
pharmacists counseling patients on adverse drug reactions. Am J Hosp
Pharm. 1980; 37:243- 7.
Haxby DG, Weart CW, Goodman BW Jr. Family practice physicians'
perceptions of the usefulness of drug therapy recommendations from
clinical pharmacists. Am J Hosp Pharm. 1988; 45: 824-7.
Kirking DM, Bickley SK, Wasserman ML. Physician opinion of
pharmacistinitiated change from injectable to oral administration of
histamine H2-receptor antagonists. Am J Hosp Pharm. 1991; 48:
Smith WE. The economic feasibility of clinical pharmacy in the
hospital setting: personnel costs. Pharmacy Related Branch, National
Center for Health Services Research and Development HSMA, DHEW, Contract
HSM-110-71-208, Document PB-220-616, U.S. Department of Commerce,
National Technical Information Service; 1972.
Fink A, Kosecoff J, Oppenheimer PR et al. Assessing whether a
clinical pharmacy program is meeting its goals. Am J
HospPharm. 1982; 39:806-10.
Ritchey FJ, Raney MR. Medical role-task boundary maintenance:
physicians' opinions on clinical pharmacy. Med Care. 1981;
Ritchey FJ, Raney MR. Effect of exposure on physicians' attitudes
toward clinical pharmacists. Am J Hosp Pharm. 1981; 38:1459-63.
Hardin TC, Reed WE Jr, Talbert RL et al. Physician perceptions of a
clinical pharmacy program. Am J Hosp Pharm. 1982; 39:125-6.
Ritchey FJ, Raney MR, Keith TD. Physicians' opinions of expanded
clinical pharmacy services. Am J Public Health. 1983; 73:96-101.
Thomas P. Professional relationships between general practitioners
and pharmacists in health centres. Brit J Gen Pract. 1991;
Lobas NH, Lepinski PW, Woller TW. Satisfaction of physicians and
nurses with clinical pharmacy services. Am J HospPharm.
Sulick JA, Pathak DS. The perceived influence on clinical pharmacy
services on physician prescribing behavior: a matched-pair comparison of
pharmacists and physicians. Pharmacotherapy. 1996; 16:1133-41.
Bradshaw SJ, Doucette WR. Community pharmacists as patient
advocates: physician attitudes. J Am Pharm Assoc. 1998;
Ranelli PL, Biss J. Physicians' perceptions of communication with
and responsibilities of pharmacists. J Am Pharm Assoc. 2000;
Bailie GR, Romeo B. New York State primary care physicians'
attitudes to community pharmacists' clinical services. Arch Intern
Med. 1996; 156:1437-41.
Spencer JA, Edwards C. Pharmacy beyond the dispensary: general
practitioners' views. BMJ. 1992; 304:1670-2.
Nathan A, Sutters CA. A comparison of community pharmacists' and
general practitioners' opinions on rational prescribing, formularies and
other prescribing related matters. J R Soc Health. 1993;
Segal R, Grines LL. Prescribing authority for pharmacists as viewed
by organized pharmacy, organized medicine, and the pharmaceutical
industry. Drug Intell ClinPharm. 1988; 22:241-6.
Berner R. Pharmacists are starting to move in on doctors' turf.
Wall St J. 1999; Jan 28.
Hirsch JD, Gagnon JP, Camp R. Value of pharmacy services:
perceptions of consumers, physicians, and third party prescription plan
administrators. AmPharm. 1990; NS30:20-5.
Campbell RK, Saulie BA. Providing pharmaceutical care in a physician
office. J AmPharm Assoc. 1998; 38:495-9.
Sardinha C. Physician groups embrace pharmacists: collaborations
that work. JManag Care Pharm. 1997; 3:526-30.
"Clinical" pharmacy a misnomer. J IndianaState Med Assoc.
Snodgrass W. Clinical pharmacy. CalifGen Practitioner.
1972; 23:13. Letter.
Griffin JF. More physician-pharmacist cooperation needed. N Engl
J Med. 1986; 315:1097. Letter.
Nesbit F, Chaplin DC, Edwards JA Jr et al. Counseling of patients by
pharmacists. Blessing or curse? NC Med J. 1995; 56:115-8.
Cowen DL. Changing relationship between pharmacists and physicians.
Am JHosp Pharm. 1992; 49:2715-21.
William E. Smith, PH.D., is Associate Dean and
Associate Professor, School of Pharmacy, Virginia Commonwealth University,
Richmond, VA. Max D. Ray, Pharm.D., is Dean and Professor, College
of Pharmacy, Western University of Health Sciences, Pomona, CA. David
M. Shannon, PH.D., is Professor, College of Education, Auburn
University, Auburn, AL.