Effects of an intervention approach on the medication prescribing behavior of physicians

PDF

Download

Open document

Flip pages

Summarize

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content EFFECTS OF AN INTERVENTION APPROACH ON THE MEDICATION
PRESCRIBING BEHAVIOR OF PHYSICIANS
by
Horace Bryant Williams, Jr.
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Public Administration)
May 1986
UMI Number: DP31178
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMI
Oissiffirtaiion PüWiaNng
UMI DP31178
Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106- 1346
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CAUFORNIA 90089
This dissertation, written hy
HORACE BRYANT WILLIAMS, JR.,
under the direction of h.L?... Dissertation
Committee, and approved by all its members,
has been presented to and accepted by The
Graduate School, in partial fulfillment of re
quirements for the degree of
DO CTO R OF PHILO SO PHY
Dean of Graduate Studies
W723
/ A5
Date
airverson
11
Acknowledgments
Many thanks are due to numerous individuals and
groups of people for their support during the process of my
doctoral studies, research and writing of the Dissertation.
Within the University community, 1 wish to thank
the members of my Guidance and Dissertation Committees.
Professors Lois Friss, David Lopez-Lee, John Peters (faculty
of the School of Medicine), Francine Rabinovitz, and
Michael White, the chairperson, were members of the
Guidance Committee. The Dissertation Committee consisted
of Professors Robert Koda (faculty of the School of
Pharmacy), David Lopez-Lee and Michael White, the chair
person. Professor Koda was most helpful with the writing
style and the organization of the Dissertation. Professor
Lopez-Lee's knowledge of statistics was very valuable.
Professor White's guidance for both committees was above
and beyond the call of duty. He was always available, en
couraging, understanding and nurturing. His encyclopedic
mind was an asset. Jose Serra, a student enrolled in the
Masters degree program. School of Public Administration
was helpful in obtaining the statistics software.
Within the study institution, there was much
support and enthusiasm. Mrs. Lorisse Maxwell, special
assistant to the President, and my immediate supervisor
Ill
during most of the doctoral work was very supportive and
understanding. Dr. Donald DeFazio, the assistant director
of Pharmacy, managed the pharmacy department skillfully and
was instrumental in putting together the Antibiotic Guide
lines, the Measures and getting the cooperation of the
medical staff. Dr. Ian MacMillan, a pharmacist, programmed
the computer with much dedication and patience. Mrs. Joni
Duzy, the secretary, used her expert editing and typing
skills on numerous occasions with no complaints. Mrs.
Angela Finer helped me keep the many details of my life
integrated as I performed as spouse, parent, director of
pharmacy and student. The entire pharmacy and medical
staff members participated in the study phase with much
professionalism.
My support community deserves much credit also.
My wife Perla and our children, Emily, Bryant and Felicia
were very understanding and supportive. They shared both
the agony and the ecstasy. Thanks to my brother Oscar and
my sisters Ruth and Shirley for continuing to be "family"
with me. Our deceased parents, with their fourth grade
education, shared their vision and wisdom with us. Their
love will always be with us. With his expert counseling,
Victor Nicassio, a friend and co-counselor, aided me in
thinking clearly about my objectives and keeping them in
sight.
To the above and many others, my} thanks for being
.a_par-t-of-my_l if e_during_thi s_pruc.es,s_._______________________
IV
ABSTRACT
Previous research on the prescribing of medication
by physicians has identified the problems involved in
medication decision making. Not many studies were devoted
to improving the prescribing behavior of physicians.
Very few of the interventions produced long term effects
on prescribing behavior. This study attempted to demon
strate the effects of an intervention approach on the
medication prescribing behavior of physicians. Careful
collection of data from multiple observations at several
time periods before and after initiation of the interven
tions and analysis of this data were at the center of
this study.
Data was collected in a medium size Southern
California institution with respect to the prescribing
of antibiotics by physicians. The objective was to deter
mine if an educational program could optimize physician
prescribing behavior. Data was retrieved for more than
two years. Eleven types of prescribing errors or measures
were developed and applied to the data to determine if
there were changes in the prescribing behavior. The
three interventions included a lecture by an infectious
disease specialist, a lecture by an infectious disease
specialist with written material on the lecture distributed
to pharmacy and medical staff members, and the direct
V
intervention by a pharmacist when orders fell between
certain parameters. Data was analyzed on 77 physicians
by analysis of variance (ANOVA) with repeated measures.
To avoid low frequency problems, the physicians were
separated into two groups. The groups were low and high
participation in the educational programs.
All changes were in the expected direction.
Significant ones were (1) the two successful measures
were antibiotic selection and dosing of antibiotics ;
variance was shown between the time phases indicating
that there was a response, (2) there was statistically
significant variance among the time phases and (3) the
mean performance on the time phase of the third interven
tion was statistically different from the mean performance
on the time phases relating to no intervention, and the
first and second interventions. After the third interven
tion, a decrease in costs of cephalosporin antibiotics
was demonstrated.
Untapped possibilities of improved health outcomes
and decreased costs of medication are possible when pharma
cists are used to aid in improving prescribing behavior.
VI
TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ..................................... ii
ABSTRACT..................................... iv
LIST OF TABLES ................................. vii
LIST OF FIGURES ............................ . viii
Chapter
I. INTRODUCTION................................... 1
II. REVIEW OF THE LITERATURE...................... 5
III. METHODOLOGY.................................. 45
IV. FINDINGS...................................... 62
V. CONCLUSIONS AND POLICY IMPLICATIONS .... 78
REFERENCES......................................... 86
APPENDICES............................................... 97
vil
LIST OF TABLES
Table Page
I. Description of Studies Based on Intervention
Approaches to Improving Physician
Prescribing Behavior ...................... 41
II. Hospital Fact Sheet 1983-1985 ................ 47
III. Staffing and Services of the Pharmacy
Department, 1983-1985 ....................... 48
IV. Medical Staff (by Specialty and Affiliation)
as of April 1985 49
V. Raw Scores Resulting from the Total Number
of Incidents, for Each Measure, for Each
Time Period................................. 64
VI. Adjusted Scores Resulting from the Total
Number of Incidents, for Each Measure, for
Each Time P e r i o d .......................... 65
VII. Results of ANOVA Performed on Measures 3, 4,
8, and 9 ................................... 70
VIII. Summaries of ANOVAS with Repeated Measures
for Physicians Who Attended Both Lectures
and Physicians Who Attended 1 or Zero
Lectures................................... 71
IX. Tests on Means of Factor B (Time Periods or
T0-T3) Using Newman=Keuls Procedures for
Measure Eight (Antibiotic Selection) . . . 74
X. Tests on Means of Factor B (Time Periods or
T0-T3) Using Newman-KeuIs Procedure for
Measure Nine (Dose of Antibiotic) ......... 75
Vlll
LIST OF FIGURES
Figure Page
1. Bar Graph Depicting the Strength of the
Measures by Showing Which Have the Highest
Cell Frequencies.......................... 66
2. Line Graph Depicting the Strength of the
Measures by Showing Which ones Have the
Highest Cell Frequencies ................... 67
3. Cephalosporin Utilization by Generation--
1985-1986 ................................... 77
CHAPTER I
INTRODUCTION
Hospitals, governmental bodies, hospital accredita
tion agencies, and health insurance companies have
attempted to improve the quality of care of patients in the
hospital environment. The various modalities of evaluation
must be considered when determining the quality of care. A
large majority of hospitalized patients take medication
during some part of their hospital stay. Control or
influence over the medication therapy modality or component
of treatment is therefore highly desirable and necessary.
Physicians have the authority to diagnose and
prescribe medication. The entire process of diagnosing is
managed by physicians with very little participation from
other health care professionals. The process of prescrib
ing, while managed by the physician, has areas where there
can be participation and cooperation by other health care
professionals, particularly the pharmacist. Participation
and cooperation among primary-care physicians, medical
specialists, and medication specialists (pharmacists) can
promote improvement in prescribing behavior (Saya &
Coleman, 1985). This requires some behavioral changes in
the hospital environment and the process must cvonsider the
nature of the medical profession and medical practice.
The physician primarily makes a judgment on the
safety, efficacy, and efficiency of drug use. Determina
tion of these aspects of drug use requires empirical,
intuitive, and rational decision-making. Each kind of
decision-making can result in an inappropriate prescribing
of medication.
Factors contributing to the inappropriate prescrib
ing of medication in the hospital environment are (a) over
sight due to haste, overwork, and carelessness; (b) failure
to review medication orders frequently and critically ;
(c) inability to keep abreast of fast-moving developments
in pharmacology ; (d) insulation of physician and patient
from cost considerations because of third party coverage ;
and (e) lack of communication between physician and
pharmacist (Soumerai & Avorn, 1984). The inappropriate
prescribing of medication may result in the emergence of
drug-resistant organisms, a high frequency of adverse
effects, excessive costs, and therapeutic failure. These
will be discussed in Chapter II.
This research project evaluates the effect of an
educational intervention on the rate of the inappropriate
prescribing of antibiotics. The objectives of reducing the
inappropriate prescribing of antibiotics are listed above.
Achievement of these objectives will improve both patient
outcomes and hospital efficiency and reduce the hospital's
financial risk in providing medical care. Antibiotics were
3
chosen for this study because of sixteen major therapeutic
classes of drugs that account for more than three-fourths
of total drugstore and hospital expenditures, antibiotics
were one of the top classes. Antibiotics and cardio
vascular drugs account for more than one-fourth of all
costs of drugs (Baum et al., 1985). Antibiotics alone
account for one-third of the hospital total. It is
estimated that hospitals spent $3.0 billion on ethical
pharmaceuticals in 1982; therefore, the estimate for
antibiotics would be about one billion dollars.
Two major issues were considered in planning this
study, namely a suitable intervention and the nature of the
study group. Both will be discussed in Chapter II.
Approaches available for improving physician
prescribing behavior include (a) medication audits,^
(b) interaction with prescribers during administration of
2
medication, (c) medication restriction programs, (d) con
trol of contact by pharmaceutical representatives, and
(e) education programs.
Education programs were chosen as the intervention
approach for this study. Education programs include
A medication audit is a formal, periodic examina
tion and checking of medication orders to verify their
conformity, or lack of, to established guidelines.
2
Medication restriction programs are methods of
reducing medication use that rely on permitting the use of
medication that is approved in the formulary or approved
after consultation with designated physician specialists.
4
reports and literature from hospital departments, commit
tees, and pharmaceutical manufacturers; education confer
ences, medical staff meetings, and grand rounds; and
pharmacy consults, drug information services, and a
newsletter. Using educational programs for improving
physician prescribing behavior has the advantage of causing
the least conflict among health care team members.
The approaches for improving physician prescribing
behavior involves changes in behavior and this will be
discussed in Chapter II.
CHAPTER II
LITERATURE REVIEW
The literature review will cover (a) evidence of
the consequences of inappropriate prescribing of medica
tion, (b) variations in medical practice, (c) some theory
supporting planned change, (e) studies utilizing pharma
cists to assist in improving prescribing behavior, and
(e) studies using intervention approaches to improving
physician prescribing behavior. Numerous studies present
evidence of the consequences of inappropriate prescribing.
Other studies suggest approaches for planned change of
physician behavior.
Antibiotics Are Over- and Underused
in the Hospital Environment
Several studies illustrate that antibiotics are
over- and underused in the hospital environment through
inappropriate prescribing. Scheckler and Bennett (1970)
performed twenty-four prevalnce studies of nosocomial
infections and anti-infective usage in seven community
hospitals between November 1967 and June 1969. The medical
records of 5,256 patients were reviewed showing that 30.6
percent of them were receiving antibiotics. These were
used most frequently on the pediatric and surgical ser
vices. Fewer than 30 percent of the patients receiving
6
penicillin, sulfonamides, and streptomycin had recorded
evidence of infection when the antibiotic was first given.
During the first six months of 1969, only 38 percent of 454
patients receiving antibiotics had evidence of infection
when the antibiotic was first given. The use of systematic
antibiotic agents in 1,035 patients consecutively admitted
to a 500-bed nongovernmental community hospital was
evaluated by Roberts and Visconti (1972). Their findings
indicated that of the 223 therapies judged irrational, one-
half were so judged although therapy was warranted, because
the specific therapy used was inappropriate. In the other
cases of irrational therapies, no antibiotic therapy was
necessary. Kunin, Tupasi, and Craig (1973) cited in their
article an unpublished study conducted in 1969 at the
University of Virginia Hospital, Charlottesville, Virginia.
When the data were analyzed, it was learned that in 51.5
percent of patients, therapy was either not indicated or
the choice of drug or dose was inappropriate. Misuse was
more common on surgical than medical services (61.5 percent
versus 41.7 percent). It might also be noted that the
results are generally similar to those reported by
Scheckler and Bennett (1970) and Roberts and Visconti
(1972).
In a study of antibiotic use at Duke University
Medical Centers in June 1973, Castle and others (1977)
demonstrated that 34.2 percent of all patients received
7
antibiotics (43.5 percent surgical, 21.4 percent medical
patients). A retrospective analysis of fifty randomly
selected patients, according to the Kunin categories of use
(1973), showed 64 percent of total antibiotic therapy as
not indicated or inappropriately administered in terms of
drug or dosage.
Kass (1978) reported on the Inter-Society Committee
on Antimicorbial Drug Usage's survey of twenty randomly
selected general hospitals in Pennsylvania. Records
studied were on 5,288 patients who either died or were
discharged on ten randomly selected days throughout 1974.
About 28 percent of these patients received at least one
antimicrobial drug during hospitalization. In 60 percent
of the patients the drugs were given for treatment of
infectious disease ; in 30 percent the medications were
given as prophylaxis for surgical or nonsurgical proce
dures. Prophylactic treatment was given for more than two
days after a procedure in most cases, even though it is not
considered effective after 48 hours. Discontinuing
prophylaxis after 48 hours could have reduced by 20 percent
the use of antimicrobial drugs in these hospitals.
Fry and others (1981) assessed compliance with the
use of prophylaxis with systemic antibiotics in selected
specialty procedures. One hundred three patients were
reviewed. Of all patients, 75 percent received antibiotics
preoperatively and 79 percent received prophylaxis with
8
antibiotics longer than 24 hours post-operatively.
Evaluated against the criteria, only 10 percent of the
patients received appropriate prophylaxis. There was also
a three percent incidence of drug-associated complications;
each patient with complications had received antibiotics
for a prolonged time post-operatively.
Several factors support those who would argue that
appropriate antibiotic prescribing is not a reality. Since
the minimal inhibitory concentrations (MIC) and minimal
bactericidal concentrations (MBC) are determined in an
artificial broth medium at pH 7.4, some would argue that
since infections occur in tissues and body compartments and
body fluids often at markedly different pH, ionic content,
and under different states of aerobic or anerobic physiol
ogy, it is not possible to establish guidelines for
antibiotic use. However, results obtained in the labora
tory generally have correlated well with clinical outcome.
Another controversy centers around the question of
whether drug concentrations measured in serum or urine
adequately reflect antimicrobial activity. There is no
question that assay of antimicrobial activity by high
performance liquid chromatographic or microbiologie methods
alters the protein binding that occurs in humans as well as
the synergy or antagonism of serum proteins for various
agents. Furthermore, host protective elements, such as
white blood cells, complement, and gamma globulins, are not
9
evaluated in tests done in the laboratory. Above all, one
cannot determine the subinhibitory effects of antimicrobial
agents on adherence or growth of bacteria.
Some physicians support their argument for using
combination antimicrobial therapy by citing the inability
of infectious disease specialists to definitively correlate
laboratory studies with clinical outcome. Combination
antimicrobial therapy has relaly few established indica
tions, but the frequency of use of combination therapy is
high. The rationale usually used by the physician is that
the patient is "very sick," and combinations provide a
broader coverage against possible infecting bacteria. Too
often, the major focus of discussion of antimicrobial
pharmacokinetics has centered around the neutropenic
patient; this is a special area within the field of
antibiotic therapy.
Unnecessary Infections and Resistant Varieties
of Bacteria Result from Inappropriate
Prescribing of Antibiotics
What is the case of the emergence of drug-resistant
organisms and nosocomial infections? Over the past two
decades, more than one hundred antibiotics have been
developed by the pharmaceutical industry. Almost as
quickly as new antibiotics have made their appearance on
pharmacy shelves, bacterial resistance to them has devel
oped. The emergence of large numbers of resistant bacteria
has been driven by two major factors. These factors are:
10
organisms' innate ability to adapt in order to survive and,
ironically, improper antibiotic use. Improper use of
antibiotics can breed bacterial resistance. According to
some 150 scientists and physicians from twenty-five
nations, indiscriminate antibiotic use is causing a rapid
spread of antibiotic-resistant micro-organisms (Barrière,
1982). The scientists and physicians signed an antibiotic-
misuse statement at a 1981 meeting of the International
Plasmid Conference. They criticized five practices in
particular. One of the five was, prescribing antibiotics
for conditions for which they are ineffective. This
amounts to overuse and misuse of antibiotics.
Another problem that may arise from the overuse and
misuse of antibiotics is the increase in nosocomial
infections. Nosocomial infections are infections that
occur in an institutional setting, such as a hospital,
convalescent center, or skilled-nursing facility. In other
words, these are institutionally caused infections.
Potential bacterial pathogens are present in or on vir
tually all objects sampled from the institutional environ
ment, including the patient. Aerobic gram-negative bacilli
cause 60 percent or more of all nosocomial infections.
An international conference on nosocomial infec
tions, sponsored by the Centers for Disease Control, was
held in August 1980. Haley and others (1985) reported that
from a random sample of patients and hospitals and
11
extrapolation ratios derived from the best available
sources of data, it was estimated that the nationwide
nosocomial infection rate among the 6,449 acute-care
hospitals in the United States in 1975-67 was 5.7 nosoco- .
mial infections per 100 admissions and that over two
million nosocomial infections occurred in a 12-month period
in these hospitals. Nosocomial urinary tract infections
constituted 42 percent of the infections, surgical wound
infections 24 percent, nosocomial pneumonia 10 percent,
nosocomial bacteremia 5 percent, and nosocomial infections
at all other sites 19 percent. McGowan (1985) reports that
nosocomial infection continues to be a significant cause of
death among hospitalized patients. He also emphasized the
consequence of nosocomial cross-infection for the community
as evidenced by the community spread of strains of Staphyl
ococcus aureus resistant to methicillin. Only a few years
ago in the United States, these strains were predominantly
restricted to hospitals.
The changing spectrum of hospital-acquired infec
tion continues to pose new challenges that necessitate the
development of highly active agents. Despite the introduc
tion of many effective antibiotics and an increased
understanding of appropriate infection control measures,
new pathogens continue to emerge as nosocomial opportunists
(Mayer & Zinner, 1985). Some of these bacteria have
developed increased antibiotic resistance, while others
12
have acquired the ability to survive in antiseptics,
intravenous infusion solutions, or on intravascular
catheters, and still others are ubiquitous in the hospital
and readily colonize in compromised patients.
The improper use of antibiotics accounts for 20 to
25 percent of endemic antibiotic resistance in nosocomial
gram-negative bacilli infections (Weinsteii, 1985). The
role of antibiotics in the occurrence and persistence of
resistant bacteria has been controversial. Antibiotics may
suppress sensitive flora andfeciiitate colonization of
patients by more resistant bacteria from a much larger
sensitive population. For example, many studies have
documented the emergence of aminoglycoside-resistant
strains after use of topical aminoglycoside ointments,
after parenteral aminoglycoside therapy, and during use of
nonabsorbable aminoglycosides in enteral regimens for
suppression of gut bacteria in oncology patients. Emer
gence of resistance also has been correlated with subthera-
peutic dosing (Weinstein, 1985).
These Problems Can Be Mitigated by Changing
the Prescribing Practices of Physicians
What is being done to reduce the number of nosoco
mial infections? Haley and others (1985) reported that in
a representative sample of United States general hospitals,
they found that the establishment of intensive infection
surveillance and control programs was strongly associated
13
with reduction in rates of nosocomial urinary tract
infection, surgical wound infection, pneumonia, and
bacteremia between 1970 and 1975-76, after controlling for
other characteristics of the hospitals and their patients.
Essential components of effective programs included
conducting organized surveillance and control activities
and having a trained, effectual infection control physi
cian, an infection control nurse per 250 beds, and a system
for reporting infection rates to practicing surgeons.
Untoward toxic effects of antibiotics are well
known. They range from death from anaphylaxis or aplastic
anemia, from penicillin and chloramphenicol, respectively,
to severe diarrhea from lincomycin and clindamycin, rash
from ampicillin, and nephrotoxicity, and ototoxicity from
the aminoglycosides. In contrast, the cephalosporins have
fewer side effects, therefore are sometimes prescribed with
abandon.
Iatrogenic disease is the result of a medical
intervention or encounter, which could be in the form of
medication or procedure (medical or surgical) used for
prophylaxis, diagnosis, or treatment, which leads to
adverse effects. The incidence of iatrogenic disease in
the last twenty years has not decreased but probably has
increased as modern technological developments have
produced more potent drugs and a large variety of invasive
and noninvasive procedures. The incidence of drug
14
reactions causing hospitalization varies from three to
seven percent and the incidence of drug reactions during
hospitalization varies from seven to thirty-four percent
(Justiniani, 1984).
Steel and others (1981) found that drugs were the
most common hospital intervention leading to iatrogenic
illnesses (41 percent of all complications) in a general
medical ward. Diagnostic and therapeutic procedures
accounted for 35 percent of all complications and a
miscellaneous group including falls, transfusions, and diet
accounted for 23 percent.
Three problems that relate to iatrogenic infectious
diseases are the continuing saga of super infection, the
related problems of antibiotics and colitis, and the more
serious and far-reaching problem of the continuing upsurge
in the development of antibiotic-resistant strains of
bacteria. The first and third have been discussed pre
viously in this chapter. The second problem, antibiotics
and colitis, is concerned with antibiotics that suppress
normal gut flora thus predisposing the patient to colitis
caused by Clostridium difficile. Normally, this bacterium
is found in the stools of less than two percent of healthy
adults. Clindamycin and lincomycin accounted for 80
percent of all antibiotic-associated colitis cases reported
to the Committee on Safety of Medicines in Britain between
1964 and 1978, although ampicillin, tetracycline,
15
chloramphenicol, co-trimoxazole, cephalosporins, penicil
lin, and metronidazole have been associated with other
cases (D'Arcy & Griffen, 1982).
Inappropriate Prescribing of Antibiotics
Results in Sizable Costs
The case for the inappropriate prescribing of
antibiotics leading to the emergence of drug-resistant
organisms and a high frequency of adverse effects has been
presented. What can be said about excessive costs? Health
care spending in the United States between 1971 and 1981
increased from $83 billion to $287 billion. This growth in
health sector spending substantially outpaced overall
growth in the economy, averaging 13.2 percent per year
compared to 10.5 percent for the gross national product
(GNP). It is predicted that if current trends continue and
if present health care financing arrangements remain
basically unchanged, national health expenditures are
projected to reach approximately $756 billion in 1990 and
consume roughly 12 percent of GNP (Freeland & Schendler,
1983).
The chief factors that are influencing the growth
of health expenditures currently are the aging of the
population, development of new. medical technologies,
increasing competition, and restrained public funding.
Older Americans may require more antibiotics because of a
less than optimally functioning immune system. New
16
antibiotics will be produced as a result of newer medical
technology and competition among drug manufacturers. Will
these antibiotics be under- or overused?
Medical care cost containment was legislated by the
passing of the Omnibus Budget Reconciliation Act of 1981
for Medicaid and the Tax Equity and Fiscal Responsibility
Act of 1982 (TETRA) for Medicare. Basically, reimbursement
was changed from retrospective to prospective with the
health care institution at financial risk. Hospitals find
it imperative to operate within a defined budget to keep
the financial risks in check. The pharmacy budget consti
tutes from 10 to 15 percent of the total hospital budget.
Consequently, if drug costs can be kept at a minimum by
putting into effect mechanisms to reduce the inappropriate
use of drugs, the hospital's financial risk could be
reduced. For years, prices charged for drugs increased at
a much slower rate than the United States' general infla
tion rate of the economy. However, since 1980 that trend
has been reversed and drug prices rose by 37 percent, while
the price for all commodities rose only 13 percent. The
rise in the price paid for drugs cannot be controlled by
the hospital, but the more appropriate use of drugs can.
Variations in medical Practice Suggest a Need
for Educational Programs Directed toward
the Prescribing Behavior of Physicians
Wennberg and Gittelson (1982) were concerned about
variations in medical care among small regions. They
17
learned that the amount and cost of hospital treatment in a
community had more to do with the number of physicians
practicing in the region, their medical specialties, and
their preferences for various medical procedures rather
than the level of health of the residents. Their results
suggested that such variances were due to differences in
the approach to medical practice of local physicians. In
the absence of authoritative standards, differences among
physicians in perceptions of illness and preferences for
treatment modalities appear to be the cause of much
variations in rates of surgery and other kinds of treat
ment. Wennberg and Caper (1985) also observed from other
studies that the lack of objective information on outcome
allows different options to have, basically, equal merit.
One cannot be certain, except in extreme cases, that a
given practice approach is inappropriate. It appears that
the less authoritative the standards of practice, the more
variance will occur in practice approach. The goal then
would be to take the least authoritative and most costly
standards and make them more authoritative.
The authors further observed from their study that
analysis of standards of practice provided a method for
obtaining objective data concerning physician practice
behavior and its relationship to costs. Using this
objective data, a dialog would develop among physicians
rather than an environment where each physician or each
18
group of physicians asserted their opinions. This kind of
dialog among physicians promoted medical decision-making
within the profession rather than outside of the profes
sion. The authors speculated that once discussions were
held, information would be fed back, then guidelines and
protocols could be proposed, discussed, promulgated,
accepted, and implemented, thereby affecting the standard
of practice.
A large percentage of physicians prescribe antibio
tics. Standards for prescribing antibiotics are not very
well followed except by infectious disease specialists.
This makes these standards authoritatively weak and
therefore subject to much variance. Audits are necessary
to assist physicians in discovering antibiotic prescribing
problems. The educational process is used to provide the
same information to all parties so that guidelines may be
agreed upon and a change intervention is used to modify the
behavior so that the final goal may be achieved.
Planned Change Theory Offers Strategy to Affect
the Behavior of Physicians
Change brings with it many undesirable outcomes.
Use of a theoretical framework helps to steel a change
agent from some of the effects of the undesirable outcomes.
Chin and Bennet (1985) posited three theoretical models for
bringing about planned change in human systems. A common
element to their models of change is the conscious
19
utilization and application of knowledge as an instrument
or tool for modifying patterns and institutions of prac
tice. The first model is called empirical-rational and is
based on two assumptions: (a) people are rational and
(b) people will follow their rational self-interest once it
is revealed to them. In this case, a change-agent will
offer a situation (change) that is desirable, effective,
and in line with the self-interest of the person, group,
organization, or community which will be affected by the
change. Because the person (or group) is assumed to be
rational and moved by self-interest, it is assumed that
he/they will adopt the proposed change if it can be
rationally justified and if it can be shown by the change-
agent that he/they will gain by the change. This would be
like providing knowledge, having it accepted, and thus
updating the person or group. There would be no signifi
cant behavior change; therefore, this model would not be
entirely suitable for changing physicians' prescribing
behavior on a long-term basis.
The second model is called normative-re-educative.
The assumption- underlying this model is that patterns of
action and practice are supported by sociocultural norms
and by commitments on the part of individuals to these
norms. Sociocultural norms are supported by the attitude
and value systems of individuals: normative outlooks which
undergird their commitments. Change in a pattern of
20
practice or action, according to this view, will occur only
as the persons involved are brought to change their
normative orientations to old patterns and develop commit
ments to new ones. Changes in normative orientation
involve changes in attitudes, values, skills, and signifi
cant relationships, in addition to changes in knowledge,
information, or intellectual rationales for action and
practice. Changing the environment of practice for
physicians so that changes in prescribing behavior could
take place would be an example of the use of this model.
In the case of prescribing practices, these might include
attitudes and beliefs about medication, diseases, or
patients, views of respected opinion leaders, patient
demands, fear of malpractice, financial incentives, etc.
In this formulation, the greater efficacy of face-to-face
education may result in part from its capacity to identify
individual motivations for existing practice and provide
on-the-spot acceptable alternatives that satisfy these
perceived needs. One of the difficulties in applying this
model is the effort required to accurately identify the
motivations for particular prescribing behaviors.
The third model, power-coercive, is based on the
application of power in some form--political or otherwise.
The influence process involved is basically that of
compliance of those with less power to the plans, direc
tions, and leadership of the most powerful. In the
21
hospital prescribing arena, this would, for example,
consist of removing a drug product from the formulary,
restricting prescribing decisions or selected drugs to
specialists, and imposition of automatic stop-order
policies. This approach does not depend on, nor necessar
ily result in, improving the physician's decision-making
process, thus would not be long-lasting. In the case of
requiring consultation with specialists or justifications
for use of particular medication (although power and
authority are main ingredients of the interventions), the
final decision to prescribe rests with the physician
requesting the medication. This model would cause a threat
to the autonomy and professional prerogative of the medical
profession, however, this is done in many institutions,
e.g., teaching hospitals, and with the Medicaid program.
The medical profession already feels the threat to its
autonomy and professional sovereignty by the very society
that granted it (Starr, 1982).
For purposes of this research project, the second
model is more appropriate because it satisfies some of the
demands of a professional group when planning change in
human systems.
Studies Show that Pharmacists Can Assist
in Improving Prescribing Behavior
Several articles that illustrate the utilization of
pharmacists in assisting to improve physician prescribing
22
will be reviewed.
In the early nineteen sixties, pharmacists began to
be more visible on patient care units in acute care
institutions. Communicating with other health care
practitioners was one of the first tasks necessary for the
pharmacist to learn. Williams and Sykora (1970) reported
successful attempts at communication with physicians.
Certain types of formal notes were designed to determine
which would be most effective at getting physician coopera
tion in reading and assessing the information left.
Physicians preferred notes that were sealed and clipped to
the chart and designed not to be a permanent part of the
medical record. This made the information personal because
only the physician involved was aware of the information
and the note could be destroyed after it was read. There
was no guarantee that any action would be taken as a result
of the information.
Bell and others (1973) measured the frequency of
pharmacists providing drug information in a clinical
pharmacy service and the acceptance and utilization of the
information by physicians. Other objectives of the study
were to (a) characterize the type of information provided,
(b) correlate patient, physician, and pharmacist variables
with provision and acceptance of information, and (c) com
pare acceptance of unsolicited information with that of
requested information. The results indicated that the
23
pharmacist, functioning in the clinical areas of the
hospital, can successfully disseminate information at its
point of use.
Hull and Eckel (1974) went a step further by
looking at a model of a pharmacist providing drug informa
tion to physicians in a university teaching hospital, while
actively participating as a member of a medicine team.
Working rounds were attended by the pharmacist for six
months and he provided drug information upon request. The
pharmacist actively monitored the patients, critically
reviewed their drug therapy, and offered unsolicited
suggestions for changes in their regimen when drug-related
problems were suspected. The pharmacist offered 68
unsolicited drug therapy suggestions during the study with
over one-third of the suggestions related to observed or
potential side effects from drugs. Greater than fifty
percent of the suggestions were related to observed or
potential adverse drug reactions. Two-thirds of all
suggestions were accepted by the physicians as measured by
a written drug order change within 24 hours. The percen
tage of suggestions accepted by physicians declined as the
training level of the physicians increased (from no
residency to completion of a residency program). It was
concluded that the pharmacist could contribute to direct
patient care and physician education through active
participation oh a medical team.
24
Briggs and Smith (1974) showed, in their study of
pharmacist-physician drug consultations in a community
hospital, that most of the consultations were concerned
with the dose or scheduling of medication. The greatest
number of consultations involving a single drug class
occurred for anti-infectives. In this study, a community
teaching hospital was used instead of a university-teaching
hospital and several pharmacists provided consultation
rather than one. It was observed that recommendations made
by pharmacists concerning a patient's drug therapy were
usually accepted by physicians, but no attempt was made by
the authors to verify affirmative replies by observing
changes in the medical chart corresponding to the recommen
dation. Of the 1163 recommendations made, 1026 (88
percent) were recorded as accepted.
A report of the utilization of hospital pharmacists
as a source of drug information for nurses in addition to
physicians was made in the early eighties (Schweigert,
1982). The types and frequency of questions asked of
pharmacists practicing in a decentralized pharmacy service,
and the frequency with which pharmacist-recommended changes
were implemented by the inquirer, were studied. Nearly
one-half of the questions came from nurses, 38 percent from
physicians, 12 percent from pharmacists, and 4 percent from
other hospital personnel. Pharmacists recommended changes
in drug therapy for 62 percent of the cases involving a
25
specific patient (which represented 65 percent of the total
of 3946 questions answered). The most common problem
prompting questions were about drug dose, and method of
administration.
The pharmacists’ recommended changes were imple
mented 80 percent of the time. The authors stated that
their study showed empirically that, over the years, the
roles of pharmacists have evolved from reviewing the orders
of prescribers to becoming involved in the decision-making
process for the selection of drug therapy, and the high
degree of compliance with pharmacists' drug therapy
recommendations also supported the role of the pharmacist
as a drug information source in the community hospital
setting.
An analysis of drug information provided by
clinical pharmacists was completed (Lipman, 1982). As
reported before, the most common information requested was
regarding patients' dosage schedules. Drugs of choice,
adverse reactions, and indications for therapy were the
next three most commonly reported questions. Improvements
were shown in the number of pharmacists' recommendations
adopted, the immediate response to the inquiry and time
required to search for responses. Ninety-six percent of
the pharmacists' recommendations were adopted within
twenty-four hours ; more than half (56 percent) of the
inquiries required no literature search by the pharmacists.
26
and nearly half required less than one minute of the
pharmacists' time.
An attempt to use a protocol to determine the
rationality of physician orders for antimicrobial therapy
(Witte et al., 1980) was successful in analyzing retrospec
tively the medical records for patients admitted to a 30-
bed surgical unit and a 30-bed medical unit before and
after clinical pharmacy services were initiated. The
results showed that exposure to routine clinical pharmacy
services appeared to improve prescribing rationality only
of drugs used to treat infectious conditions. The study
design had no control group, therefore the causal effect of
clinical pharmacy services on the rationality of antimicro
bial therapy could not be statistically determined.
The inappropriate use of serum drug levels can lead
to dangerous and costly therapeutic decisions based on
erroneous interpretations. Appropriately used, serum drug
assays are a valuable diagnostic and monitoring tool. The
effect of pharmacist intervention on the use of serum drug
assays was investigated by Levin (1981). The objectives
were to evaluate prescribers' use of serum drug assays,
study the influence of physician education by a clinical
pharmacist on the use of serum drug assays, and to perform
drug-related patient-care evaluations for the medical audit
committee. The authors concluded that appropriate use, as
defined by compliance with three criteria, increased
27
significantly. Since considerable savings in patient
charges were realized, it was further concluded that
clinical pharmacists could provide a cost-effective service
to improve serum drug assay therapeutic patient monitoring,
A program that monitored antibiotic sensitivity
also monitored for adverse reactions, and for adequate
antibiotic therapy and infection control (Greenlaw, 1977),
utilizing the daily culture and sensitivity reports as a
starting point. The objective was to ensure that all
patients who had culture and sensitivity tests performed
would have the proper antibiotic prescribed for them and at
the proper dose and dosing interval. This was accomplished
but at an extra cost to the institution. The author
reported improved patient care and acceptance of the
program and participation by the medical staff as benefits.
A study that demonstrated the identification of
problems by drug-use review activities (Norris, 1982) also
included the implementation of appropriate solutions and
documentation of desired changes. The results of the study
suggested that gentamicin dosage recommendations by
pharmacists could culminate in more appropriate drug
therapy.
Another study utilizing gentamicin demonstrated the
results of concurrent monitoring of therapy by a clinical
pharmacist (Powell et al., 1981). All patients in a
322-bed teaching hospital who were started on gentamicin
28
therapy during a 72-day period were monitored using
predetermined criteria. Deviation from these criteria
resulted in contact with the patient's physician by the
pharmacist, and specific recommendations were made at that
time. Major criteria deviations were found in 12.5 percent
of the courses of antibiotic therapy, resulting in the
contact of ten physicians. The physicians responded to the
pharmacist's recommendations in 60 percent of the cases.
The authors concluded that a clinical pharmacist could
interact effectively with physicians in monitoring drug
therapy in the hospital.
The foregoing reviewed studies that demonstrated
the role of the pharmacist in providing physicians and
other members of the health care team with information that
could result in improved prescribing behavior. These
pharmacy functions do not exist in isolation, but are
encapsuled with other pharmacy services called clinical
pharmacy services. The evolution of clinical pharmacy
services in a community hospital was demonstrated (Williams
et al., 1979) along with an illustration of how the various
parts are integrated.
An assessment of whether or not a clinical pharmacy
program was meeting its goal was performed by Fink and
others (1982). This assessment provided information about
clinical pharmacy's goals, activities, outcome, impact, and
costs. Questionnaire surveys were used to gather data on
29
three of the foremost goals of the clinical program from
nurses, pharmacists, and physicians. Generally, the
clinical pharmacy was found to be meeting its goals.
Another study on documenting effective clinical
pharmacy services (Gibson et al., 1982) defined the impact
of these services as one of the following outcomes:
decreased morbidity (side effects, toxicity, or adverse
drug reactions), decreased mortality, improved prescribing
or monitoring practices, improved therapeutic outcomes, and
increased compliance.
Measuring physicians' attitudes toward the clinical
pharmacist's role is another method of measuring the
effectiveness of clinical pharmacy services. An instrument
was developed and validated for this purpose (Geussing et
al., 1984). Based on physician-clinical pharmacist
interactions, statements were written and edited into
tentative subscales. Measuring the attitudes of physicians
toward the clinical pharmacist's role can yield important
information, but cannot substitute for an evaluation of
clinical pharmacy services including patient outcomes.
Intervention Approaches to Improving Physician
Prescribing Behavior
Reports on using intervention approaches to improve
physician prescribing behavior were reviewed for method(s)
used and outcome. The interventions or methods used were
education and restriction. In most cases, a medication
30
utilization audit preceded the intervention. Most of the
audit systems were based on the guideline audits of
antimicrobial use (Kunin, 1977). Audits were accepted well
by the medical staff, were cost-effective and educational
(Latorraca & Martins, 1979).
The restriction interventions relied on formulary
restrictions, requiring consultation by an infectious
disease consultant or consultation by a clinical pharmacist
or department chairperson. An early attempt at applying
mild restraint by requiring physicians to justify the
choice of antibiotics resulted in substantially limiting
the use of certain potentially toxic or expensive agents
(McGowan & Finland, 1974). When the restriction was
removed, an increase in the use of potentially toxic or
expensive agents was observed.
Nineteen hospitals were surveyed for costs and use
of antimicrobial agents (Craig et al., 1978). Total costs
of antimicrobial drugs per patient day accounted for 16
percent to 41 percent of total pharmacy drug costs. An
eighteen-month antimicrobial drug control program at one of
the hospitals resulted in a decrease of 31 percent in
antimicrobial drug costs. One study reported on the effect
of three controlled prescribing programs on the number and
appropriateness of initial cefazolin prescriptions written
in a Veterans Administration Hospital. The three programs
were (a) no restrictions on the prescriber, (b) strict
31
control whereby all cefazolin prescriptions had to be
approved by an infectious disease and pharmacy consultant,
and (c) limited control whereby the prescription could be
approved by infectious disease or pharmacy consultants or
section chiefs, but not by individual physicians alone.
Strict control, especially when compared with no control,
was most effective in reducing overall use and assuring
proper use of cefazolin (Ma et al., 1979).
A university teaching hospital was used to demon
strate the success of cost containment through restriction
of cephalosporins (Britton et al., 1981). Strict guide
lines were written and clinical pharmacists were used to
monitor them. There was a substantial reduction in costs
over the period of time that the guidelines were enforced.
The university teaching hospital has the advantage of house
staff with established chiefs of staff with some control.
Community, non-university teaching hospitals do not enjoy
this advantage. Medical staff members possess greater
autonomy and less allegiance to an elected chief of
service. The chief of the service is elected for one year
usually, therefore has very little influence. Most of the
hospitalized patients in this country are hospitalized in
community-type hospitals.
A similar study in a federal hospital was success
ful in decreasing the number of dosage units of benzodiaze
pines and cephalosporins dispensed (Huber, Patry, & Hudson,
32
1982. Federal hospitals are similar to university-teaching
hospitals; their medical staff members are paid by the
facility and therefore they have less autonomy.
The success of the federal hospital was duplicated
by others. Seventy Veterans Administration Medical Centers
throughout the country were surveyed to determine the
formulary status and the methods used to control the use of
the semisynthetic penicillins, the cephalosporins, and the
aminoglycosides (Cynamon & Kallet, 1983). More than half
of the hospitals surveyed restricted the antibiotics by
exclusion from the formulary. It was concluded that the
simplification of hospital formularies to avoid duplication
of antibiotics in the same class with almost identical
antibacterial activity would seem to be an appropriate and
attainable goal.
The 112 medical-school affiliated teaching hospi
tals were surveyed by mail questionnaire as to methods used
to restrict antibiotic prescribing (Klapp & Ramphal, 1983).
Direct control (an infectious disease specialist authoriza
tion or restricted indications for use) was used in 62 (57
percent) of 108 institutions responding. Nonformulary
status indirectly controlled prescribing in 35 institu
tions. Most of the physicians (85 percent) agreed with
restriction practices. Cost was the reason given most
frequently for restriction of the penicillins and cephalo
sporins while aminoglycosides were most frequently
33
restricted because of increased bacterial resistance. The
most common method of approval for use of restricted
antibiotics was by verbal authorization by the infectious
disease specialist. No best method for antibiotic restric
tion was evident.
The support of the medical staff in ensuring
physician compliance with restrictions on antibiotic
prescribing is very important. In a report of enforcement
by medical staff committees of a policy limiting prophylac
tic antibiotic use, a high degree of success was attained
(Nickman, Blissenbach, & Herrick, 1984). Audits of
antibiotic use were performed with no intervention, then
followed by audits performed with the pharmacy department
authorized to discontinue prophylactic antibiotic orders at
72 hours post-operatively for patients in whom there were
no signs of infection. For 153 prophylactic antibiotic
orders audited before the program was implemented, approxi
mately 50 percent were discontinued within 72 hours by the
prescribing physician. During a same period of time when
the pharmacists were discontinuing orders, the percentage
increased to 85 percent. Physicians who did not comply
were reported to the quality assurance, medical credential-
ing, and surgery committees.
Guidelines for the use of antibiotics prophylactic-
ally with surgical procedures are accepted by most physi
cians. A physician auxiliary antibiotic form was used to
34
restrict the use of nonformulary antibiotics (Pastel,
1985). To prescribe an antibiotic not approved in the
formulary, the physician had to complete the form, includ
ing his/her rationale for not prescribing an alternate
antibiotic listed in the formulary. Results of the
restriction were determined by â reduction in costs of
antibiotics not listed in the formulary. The antibiotic
utilization review committee initiated this project. A
similar study was initiated by a pharmacy and therapeutic
committee with good results (Hayman & Sbravati, 1985).
Guidelines for a restrictive formulary system are
usually written very flexibly to assure cooperation of the
medical staff. As mentioned previously, there is a
mechanism for prescribing a nonformulary antibiotic. There
can be a danger if the physicians use the mechanism too
often. Green, Chawls, and Fong (1985) reported in their
study that, of all nonformulary requests, 65 percent were
for medications previously evaluated by the pharmacy and
therapeutic committee and denied admission to the formu
lary. Excluding 22 percent of nonformulary items that were
requested for the continuation of preadmission drug
therapy, only 13 percent of the rationales for the remain
ing requests were appropriate. This information resulted
from a telephone survey of eight hospitals.
An early attempt to improve physician prescribing
behavior by using education as a method of intervention was
35
performed at Stanford University (Rubenstein, 1973). An
extramural program of continuing education based on a
federation of community hospitals in northern California
was established by the Stanford University School of
Medicine. This group education led to significant changes
in medical practice through lectures and rounds. One
measurement used to determine change in physician behavior
was that of the change in the prescribing of heparin by the
subcutaneous route to the intravenous route. After the
educational program, there was a sharp decline in the
prescribing of subcutaneous heparin and a steep rise in the
prescribing of intravenous heparin.
A team approach was used to educate physicians in a
State mental institution (Ellenor & Frisk, 1977). This
team, labeled a behavioral review committee, consisted of a
physician, pharmacists, nurses, psychologists, sociolo
gists, and therapists. The pharmacist completed drug
histories on each patient and made his/her recommendations
to the committee. The committee would contact the physi
cian to seek compliance in changing orders. During the
study period of two years, there was a decrease in the use
of antipsychotics, antianxiety-antidepressants, sedative-
hypnotics, and miscellaneous agents.
Forty-two Veterans Administration hospitals
participated in an evaluation of educational techniques for
physicians on the use of psychotherapeutic drugs. The
36
educational techniques used were journal articles and
videotapes. The findings were that there was a decrease in
polypharmacy and the use of antiparkinson drugs, but no
statistically significant differences between the educa
tional groups (Schroeder et al., 1979).
Sohn, Wolter, and McSweeney added cost information
to their education program (1980). They attempted to
encourage physicians to use a single parenteral first
generation cephalosporin since they could purchase it on
contract at a low price. The education process was one to
one by a clinical pharmacist and with printed material. A
drug use review after one month after the institution of
the education program showed marked changes in physician
prescribing habits for the first generation cephalosporins.
A study combining education and restriction proved
successful for changing prescribing habits for prophylactic
antibiotics but not therapeutic antibiotics (Durbin,
Lapidas, & Goldmann, 1981). An antibiotic prescription
form required the physician to categorize antibiotic use as
prophylactic, empirical (culture and sensitivity results
not available), or therapeutic. Over a period of two
months while the intervention was being applied, the mean
duration of prophylaxis was reduced and more first pro
phylactic doses were administered at the proper time
interval.
37
Eisenberg and Williams (1981) reviewed six differ
ent strategies that had been employed to improve physi
cians' awareness of the costs of health care and to reduce
medical expenditures. These strategies were: (a) educa
tion, (b) peer review and feedback, (c) administration
changes, (d) participation, (e) penalties, and (f) rewards.
They emphasized physicians' use of ancillary services such
as laboratory tests and radiological examinations. They
concluded that those strategies using individualized
instruction had been the most effective, and while pen
alties and direct rewards may be effective, alterations in
the then current reimbursement system could offer financial
incentives to physicians who practice in a cost-effective
manner.
The effectiveness of a structured educational
program was evaluated (Johnson et al., 1982). The program
presented information to physicians on one antibiotic,
namely, gentamicin, the specific indications for initiating
the therapy, the dosage regimen, and the precautions to be
taken to avoid toxicity. The information was written and
posted at all nursing stations throughout the hospital.
The incidence of excessive doses declined from 21 percent
to 7 percent following the educational program. This is
significant since the toxicity of gentamicin increases with
excessive doses.
38
A prospective investigation of the prescribing of
oral and parenteral cephalosporins over a four-month period
demonstrated only 16 percent compliance with criteria (Moss
et al., 1982). A nationally recognized set of evaluation
criteria for cephalosporin use was applied. A collabora
tive educational effort between the infectious disease
service and the department of pharmacy was the method of
intervention. Information on the proper use of this
antibiotic class was disseminated through a series of
lectures and by the publishing of a series of Drug Informa
tion Newsletters. One problem that the investigators
pointed out was the fact that medical staff members
practice at other hospitals in the area that do not
restrict the availability of the agents. Consequently, the
gains are slow or short-lived and require repeating the
educational program.
A time-series design using oral analgesics was
reported by Berbatis and others (1982). The hospital's
drug bulletin, which contained information on the treatment
of minor, moderate, and severe pain, was distributed to all
members of the medical staff. Prescriptions for oral
analgesics for all inpatients were surveyed one, three, and
five weeks before and one, three, five, and seven weeks
after the distribution of the bulletin. For the medica
tion, propoxyphene, the drug bulletin significantly reduced
its prescribing across both the minor and moderate pain
39
categories. An increased use of aspirin and acetaminophen
was significant three weeks after the release of the drug
bulletin. Aspirin and acetaminophen are less costly than
propoxyphene.
Physician prescribing patterns of postoperative
prophylactic antibiotics on a university teaching hospital
orthopedic unit before, during, and after one-to-one
teaching by a clinical pharmacist were monitored and
compared with patterns on a similar unit without a clinical
pharmacist intervention. There were consistent reductions
in overall and antibiotic drug costs in the experimental
group, but the differences were not signifient. There was
a significant reduction during phase 2 in number of hours
of antibiotic prophylaxis after removal of surgical drains,
but prescribers' overall compliance with guidelines for
postoperative antibiotic prophylaxis did not change
significantly over time (Herfindal, Bernstein, & Kishi,
1983).
The same authors (1985) conducted a study using a
cardiothoracic/vascular surgical unit. The purpose of this
study was to study the impact of clinical pharmacy services
on the cost of drug therapy on this unit. Physician and
nursing attitudes about the usefulness and likely effect of
clinical pharmacist recommendations were also assessed. A
cross-sectional design with a temporal factor was used to
study physician prescribing of all pharmacologic classes.
40
and particularly of antibiotics. Measurements were taken
for nine months before the institution of clinical pharmacy
services, twelve months during a clinical pharmacy service
period, and for six months after the cessation of the
services. Costs of drugs per patient day and total number
of doses per patient day showed a reduction during phase 2
and a return to phase 1 levels with cessation of clinical
pharmacy services.
Table 1 offers a tabulated description of the
studies based on interventions.
In conclusion, the various interventions did effect
changes in physician prescribing behavior. None were
continued on a long-term basis; this needs to be investi
gated. The measures of the behavioral changes appeared to
be effective; standardization of the measures would be
helpful for future studies.
0
H
CO
s
§
I—I
g
1
S .
M
s o
o s
M
O pq
W M
CO p a
< U
PQ CO
M
CO p £ :
M
M
IS
CO U
M
k CO
o > - •
gS
o
M O
H g
Ph I —I
H >
P < o
U Pü
CO p L ,
gs
5
CO
I
u
H
S
CO
I
CO
s
A
g
pa
s
ë
>4
g
H
CO
.
O
00 C O 1 —1
c +J C O C O
•H c 0) p X
rO C U X
•H A
•H 00 4J J-I J4 O
J-i C O o o o
O PI C O A A
C O C D •H 4J o rn TO
Ü) >
C O J-I O P
J-I •H O O A 1-4 C O
PL C O C O O C O
C C O T3 X C O
J-I cu m 00 PI A P
o PL J -I T3 3 C O O P!
y-4 . X C J O J -4 O •H
C O O J PI |J » "O ' C J A
C O •H J -4 C O CM P
•H 'H 1 + -4 C J r-4
P o N
o P
C O C O P
C O o O o P
C O 0) o 15
•H > P
x: X I X s e o C X
4J PL 4-J C J C O •H •H
•H C U J4 X 4J T3 CM
13 U T3 p PI O P o
0) C O o C O O O
O J 1 —* 4-J C O 00 P •H C O
C J C O "H
3 C O C T ) 4-J J4
C J-I B
•r-( O O P TO
C O a •H PI o J4 p P X P
•H 4-1 f —< o >
•r4 T3 B
C O
r— 1 Ç 3 •H •H 1 —1 1 —1 1-4 O P p
PL C l) 4J C O CO O o J4 P p
e Ü PI
J4 N A
O C O
T —1 •H o M -4 4-1 C O 1-4
P C O
C J PL C O J-I PL O P!
CM P •H •H
•H 4-1 X O C J •H TO
IS T3 4-1 C O C J C J o O 4 - 1 p
o p: S C J C O P C O C O
r-4 C O C O J4 M -4 C O 4-1 «44 C O p p
4 -S o C J o o 4J p •H
1 —C C O P J4 •H C O J4 J4
J-I C O X o C J o J-I o X o O
Ü) J-I P C O C J 4-1 ( 0 P p A
> o CO P A CO p CO A C O
< <3 X o A W
41
CM
00
J4 4-1 p C O C O N
P P p 6
r— i
4-1 P X P
•H
B
•H P 1 —1
J4 P p bO 00 p
u a O O CO ON o
o CM C O P r ~ - 1-4 C O
T) TO 0 0 p A ON TO
P o\ p TO 1-4 ON 0
X TO
1-4
A P P
r ~ - C O X
C O P P O P ON p
•r4 P P o •H Î—( C O 1 - 4 p TO
1 - 4 C O X •H 4-J P P X P
X P p H 4-J O •H P 44 P
p 4-1 p P •H A X C O O
4-1 P X X P P 4-1 P >N
C O TO p 4-1 4-J 4-J 4-J TO o P TO P
P p o •H C O m P X P
4-1
CM X [2
•H p P TO 4-J P P
O p TO P pa P o A
C O 4-J P P •H P P P
4-1 P P O B p P TO o
•H O p
•H TO o 1 2 bO P
4-1 P
TO •i4 o C O 4-J <3 •H O
•H P 4-1 P
P 4-1 •H C O O 4-1 o p •H X
<3 O 4-1 o P bO P X
p
P
P P
P Ü s P C Ü s O S X X
bO 1-4 P P P •H
p
1-4 4J p P TO
J 4 o <44 P p O P
A o O < 1-4 A s s <3- X u ■ A M
0
H
CO
g
1
A
<
§
M
H
I
g
X O
o X
K -!
A X
X M
X pa
X X
M
X p s S
X A
M
t s
X C J
K-!
X X
A
g
H S
A M
M t>
A O
C J A
X A
g S
T3
S
O i-H
•p C O
p >
1 1 o p
P P p p
O P t p p O
O > > N p "p
C O p P P 44
X
»p p
•p X X P O A p
A P P P O p a)
P p p B A
P P P
•P o A •p P
P •P •P •p p P
P P p P P
|S A A X p C O p X
P P p p P p P
C O P P p -P A
p C O « p A A Ü O Ü
p •p •P X p p
A )—1 ï P •p
P P A •p
P X P P p
O p P . ^ P g P ü
o O B B X p
O P P
P A A
X •p C O p A A P
X
4-1 C O C O X X P p P P
S O o O p p P P O
O •p •p •p P P P
u X p p O A A O A
H •p o o B A P P •P A
A 4-J •P •P P O P > P
O p X X o *r4 P P P
p •p -p 1 5
i-H A P X P
C O p p P A P C O p O
X o P p p A P •p X P
H •p P p p 44 P p
X 4-J O O C O •p bû C O
A O X p p bû P B P p
X P •p •p A P :s A •P
:s
X
Tp CM O •H P X
A r" p p
•P p C O •p C O
X o o P X -p P p o
A A p •p > N •p P X P o X
X O p p U p P A C O P P C O p
H p ü p ü P p X
> P o
A A A P r-4 p X P 1 5 > N P •H
O p A p A p X
C O A X
A P p P B P 6
P P A p •p
X P p U u P P P O O C O P p
A O •p o O > ”P P o X p
S
p <
44 C
44 O P A A H p
A CJ A X
42
>4
g
g
c s |
T3
P
P
X
o
p
g
p
C O /—N
C O s t
•H X
1-4 ON
X I — !
P A
P ü
•H
P
P
•P P
Xi X
A
a i
P
O
p A
p A 1 P
p O •P P
p P
p bO P C O
o P P o
X •r4 •p
p X
p
C O •r4 C O o
p o p
A ü •p A
O C O p >4
p o X
bû p X
1
P A O P
•p >
X P C O •P
•P X A P
P P •p P
O P A
C O P P P
p •r4 P C O
p
A p A
C O O C O
P •p p
X p P p
p C O p C O
A P p C O p
P P •p w p
•p P P p p p
P P X p bû
P C O A p A P
P P P
"p p X P A C O
1-4 •p- -P p
U A C O P o
P P P P P p
A P O C O P p
P p 1 p
A P P p >1 X
P •P > p p î— 4
P P P o p P
X P P p -p O
C O A P A X C O
P P P 'p
<3 X
•P
<3 P B
CO 1 -4
ON P
X O
•P
P
p P
p P
m
B
O
oo A p
ON
r-4
B
r 4 o p
o P
X
“ P A A
P P
P P >N
> « en
P C O
P p P C O C 3 N
X A ü B
1 —1
I T ) CO •p P P
00 A p C O p
0\ A P P P bû P
T — 4
P X o > O •p
P CJ •P P p
P A p
r-4 P P
A P C O
p P •P O P P p
p B X P O p
C O >N p X P *r4 X
p p 3 o C O p P
A . s Q p p p A
p p P ü
p A P P
o P O A
O X
X CJ A S W <3
> >
X
r-
ON
A
r a
*p
p
A
TO
P
P
P
P
P
P
X
X
43
0
H
CO
1
A
<
g
X
I
I
g
X
t X 1 t X
p • H , 1 P p
p P P >
P p •H X P O X o
1 P •H A X o •H P •H P
e B
•H •H O X
t X P A A P •H P X
P p p
P X •H P p O O A U O p P p p
C p G
•H u X P y
•H p •H
>
P X p g i p bû A P A p X p p p
p p u P
P P o •H p P •H p
A bO p P P p p P P P •H P X P p
O P P p O p X X O U P P P P CM A
•H •H bû P J 4 p O •H P A P P P X P
P X P P p o P A O O P A EH P
C O •H O A •H A o P O •H
X A P P P
3 P A O
t X A p bû e bû A A P O P •H
O P A A o
t X
P P >N P J 4 X X
P P O P e P p P A •H •H X bû p A p
X P t X P o P p A p. P X o t X P
t X
p P P P o o e P P •H P A
t X t X P bû >N
A X O
t X
p p P e O p P O P X
A A A bû p p p P U O p J 4 u p •H •H A
CO P P P P J 4 p p X O P p A P P
A P O
P P O P •H O t X p p P P > P •H P P
•H > P A J 4 •H J 4 p P O e P A A
o >N O P P •H P bû P p O A
e A X P •H bû
CJ O •H O > P P P •H o •H O p X P P •H P O
H p P •H O P bû•H P > p A P •H P P O A
X e P X P X O P bû X P u A •H P
O p X P p O O P •H P A X P p P O t n P X P P O
p A P X P •H J 4 •H p •H P A r X O P O O •H
X P p O P A X X X P P •H P P X
CO A P P X P P P
•H p J 4 P P P P A •H
H >N •H P A X P P A P bû U o O P O O X P P P
X
X P bû O CM • H P P P O X P X o P p P P P
X o bO P X P A P X • H p A A A P
P
CO A P P P X p e bû P P X p p O p P p P
M P bû p X o o U • H • H J 4 p P p J 4 • H P P
A P A P P p J 4 A O J 4 P J 4 A o • H p P P X
• H P • H p p A P A O P O u P
e
i X p P P O p
O P X
A p A > A J 4 P A o P p p P O • H
M P O O p • H A • H P P P P P p • H p p p P p
A p
H C O P P Ü P t X • H O J 4 P J 4 p u p O p • H p
A P P A X p P P 1 P t X P A P O o p p o p P p
O P • H P p p • H • H P P P X P p bû P • H p bû o
A P A A p X U P X p p
J 4 p P t n P bûA • H X
M O P P p p P O O • H A p O p P A p P P
A P
A p P P o t X
X p X J 4 P p • H P X P P X O P • H A
Q A S
A A H A CO O O • H O P A O P E H CJ A A P o
!a o
o !a
A
C J o M A O X A
o m
w A
C O Pi
C O
< C J 00
pq C O
M
t X
O N
1 —4
C O A o 00
M A 00 C 7N
A ON i X • H
O B
1 —4 X
p) <
p
E H A p
• H
C O C J C 7N t n p A
A r - - p B
C M
A C O C 7N P A C M 00 A
O
A p 1 —4 00 C 7N c
X p o C 7N
1 —4 P
B A o
1 —4 C M
P O P C O 00
A O P o A ON C O • H
E H X P s P C O 1 —4 p P
A A X P p p P
A
> P A p X P
A O O p p
X
C O p P
C J A p P p p o P
C O A A A o p ■ P
M S P p • H
X
A pq
O A P p A A p P
p P P o P
P t X
X P
1 —4
P p A P p
A 12 P P
• H A
A P P O P P P
O
• H p P • H
P P P X P
P ■ X
A
P X X p X
P p P
O o O p O O O ) P
P Q C O C / J A > “) S
pq
X
X
M E H
C O
S
C J A M A C J X
A
g
g
C O
I
A
<
B
M
g
g
A >
ea
X w
M A
A
O PQ
W A
C O A
C O
m C O
A
C O A
M A
A
C O u
A
A C O
o > >
X
X A
O
A O
H X
A A
M >
A O
U A
C O A
g S
A
P
C
O
O
M
a
> N
rA
P bû
P P
P p A
O p P
A
X P P
! — t P A
6 A O
r —1
>
A P P •H
P A
P P P O P
P P bû C O A P
P P A A p > C O
> P P P p P C O
O P P p P p
A P O o C O o
O bû P p
p P P A X
p m P A U p
p O X P P P
• H
X A O P Ê >
• H A P
P P p P o P P C O
U O o p > X lA C O
p A p A O A P P
C O p X > O
w p P A P C O *A P A
X A P A P P r A
>
o P A P > N U P
U P A A P • H A P
H P lA P P A P CA
X X P A P A P
O A X A P lA C O P > N
A p P O P O
P P
lA p P X P
C O P b û > N p A X A g
H
A
X A P p P
X P X A P p O P
X P P O C O A o P X
C O A A P b û e A
w 1 5 A P P P
A A P P C M P
tA
P P U A A A P P
X P O A P P U
M
• H P P
e
A C O P A
Ê - l bO P O A O p C M P P
A P A A P C O A
O P
A X C O o P P A
A p A A P p A C O P
M e P P C O p A
A A o P > o A
X P A
O U P O U O A P O
44
k H
g
A
C O
p
C
O
o
m
00
ON
• H
X
p
• H
A
A
P
P
P
• H
P
P
P
P
P
P
X
p
A
P
• H
A
P
P
X
45
CHAPTER III
METHODOLOGY
Introduction
The previous chapter has identified the conse
quences of the Inappropriate prescribing of medication, and
methods that were used to prevent such. This study
measured the effects of an intervention approach on the
medication prescribing behavior of physicians. The
intervention used was an education program because, as
stated in the literature review, it has been successful in
the past and caused less conflict among health team
members. The drug class chosen for this study was the
antibiotics because they constitute a large share of the
pharmacy budget for medication costs.
The general thrust of the research effort was to
demonstrate ;
1. that nonphysicians could monitor the prescribing
behavior of physicians in an acute care institution;
2. that the prescribing behavior of physicians could be
modified or changed in response to the proper interven
tions;
3. that this modification or change could be measured;
4. that this modification or change could be sustained
over time.
46
This chapter details the research site and sub
jects, procedures, collection of data and instrumentation,
research design and hypotheses.
Research Site and Subjects
The institution selected for this study is located
in an area of Southern California near Los Angeles,
California. The subjects were members of the medical staff
of the institution. Demographics of the institution, the
pharmacy department, and the medical staff are shown in
Tables II, III, and IV.
The three areas of the institution used for the
study were medical nursing units, surgical nursing units
(including obstetrics and gynecology), and the critical
care nursing units (including coronary care, surgical and
medical intensive care units).
Antibiotic orders for all four hundred one physi
cians were daily monitored by pharmacists. There was
little change in the medical staff members during the study
period and pertinent information, such as number of
patients admitted by each physician during the study period
and their attendance at the lectures, was made available to
the investigator.
47
TABLE II
ADMISSIONS
SURGICAL PROCEDURES
NUMBER OF EMPLOYEES
NUMBER OF BEDS
HOSPITAL FACT SHEET 1983-1985
1983 1984 1985
I2,!217
6,677
1,350
345
12,197
6,677
1,350
345
11,325
6,222
48
TABLE III STAFFING AND SERVICES OF THE PHARMACY
DEPARTMENT, 1983-1985
*NUMBER OF FULL TIME
EQUIVALENTS
1983 1984 1985
Pharmacists 22.75 22.75 2 0 . .25
Technicians 22.75 22.50 2 0 . .25
Secretary 1.00 1.00 1 . .00
Pharmacy Students 6.00 4.00 4..00
SERVICES (the same for
all three years) 24 hour unit dose drug distribution,
IV admixtures, satellite services, dis
charge prescriptions, 24 hour service,
newsletter, pharmacokinetic dosing,
nutritional support, USC clerkship and
information
*Each Full Time Equivalent (F.T.E.) represents 40 hours worked
per week or 2,080 hours worked per year.
49
TABLE IV MEDICAL STAFF (BY SPECIALTY AND
AS OF APRIL 1985
AFFILIATION)
PRIVILEGES: ACTIVE ASSOCIATE
— « issr
DEPARTMENTS
GENERAL PRACTICE 25 3 11 1
MEDICINE 60 4 38 23
Dermatology 3 2
Neurology 3 1
Pediatrics 5 1 14 5
OBSTETRICS/GYNECOLOGY 16 5
PATHOLOGY 3 2
PSYCHIATRY 11 1
. 7 7
RADIOLOGY 9
3 ,
2
SURGERY
Anesthesiology 8 8 2
General Surgery 15 1 12 7
Neurosurgery 3 1 2 1
Ophthalmology 6 1 3 3
Oral.Surgery (dentistry) 1 4
Orthopedic Surgery 11 6
Otolaryngology Head &
Neck 6 3 4
Plastic Surgery 4 5 1
Thoracic Surgery 4 1 5
Urology 5 2 1
TOTALS 197 13 124 67 401
50
Procedures
In January 1983, the Pharmacy and Therapeutics
Committee and the Medical Executive Committee (the highest
authority for the medical staff members) approved a method
for antibiotic auditing. The audit was based on seven of
the eighteen guideline-audits of antimicrobial usage
developed by the Veterans Administration ^ Hoc Interdis
ciplinary Advisory Committee on Antimicrobial Drug Usage
with Calvin Kunin as project leader (Kunin, 1977), and the
antibiotic review was performed within the study institu
tion in 1982. The other twelve guideline-audits were not
relevant to this study.
The guideline-audits were presented in a tripartite
format consisting of (a) element, (b) standard, and
(c) exceptions. The seven guideline-audits that were used
were based on the following areas:
1. Prophylaxis in surgery: Use of antimicrobial
agents to prevent infectious complications during and
immediately following surgery.
2. Nonsurgical prophylaxis or indications for
antimicrobial prophylaxis in nonsurgical adult patients.
3. Oral Cephalosporins: These are expensive oral
antibiotics and there are few instances where they are the
primary drugs of choice for any infection. Less expensive
and equally effective drugs should be used unless the
organism is resistant to the antibiotic or the patient
51
exhibits an untoward reaction to the agent of first choice.
4. Parenteral Cephalosporins: These are very
useful antibiotics but because of their low toxicity, they
are sometimes prescribed in excessively high doses for
prolonged periods of time.
5. Tests to be ordered in Conjunction with
Antibiotic Use (listed as #12 by Kunin). Laboratory test
monitoring is necessary for some nephrotoxic antibiotics;
the white blood count (WBC) is important in evaluating the
response to treatment and the appearance of adverse
reactions in patients with infections. Tests for anemia,
azotemia, and hypokalemia are also important for some
potentially toxic antibiotics.
6. Ordering Antimicrobial Agents for Hospitalized
Patients (listed as #16 by Kunin): A profile of antibiotic
use within the hospital is maintained to determine if it is
changing for the hospital. If so, it could indicate an
overuse of certain antibiotics so that a selection of more
virulent strains is occurring.
7. Antimicrobial Sensitivity Tests and Therapy
Monitoring: Are the indications for sensitivity testing
appropriate? Are the reports of the results of sensitivity
testing timely?
The antibiotic review included use of data from the
pharmacy and review of routine orders for prophylaxis in
surgery and for specific infectious diseases. Audits of
52
individual agents were based on standardized guidelines for
use. Since this was a sensitive and complex process, much
time was spent in trying to understand the constraints of
medical practice and social forces exerted on the physi
cian. It would have been counterproductive to single out
the physician or the pharmaceutical industry for blame for
inappropriate antibiotic prescribing or to add new con
straints to medical practice in the institution. The
entire effort was designed to reduce and manage conflict
rather than escalate it. As an example, we learned that
prophylactic antibiotics were frequently used in patients
undergoing clean surgery. Most infectious disease special
ists state that prophylactic antibiotics should be used on
a very limited basis (Kunin, 1977). Some physicians
justified the use of prophylactic antibiotics in patients
undergoing clean surgery based on the surgical procedure's
association with a substantial risk of bacterial contamina
tion. The justification was that even though the risk was
small, development of infection, in some cases, was
associated with substantial morbidity or mortality. Some
but not all of these situations were accepted as appro
priate for the study. In other words, some "shades of
gray" were expected, respected, and allowed.
Clean surgery is a term that refers to surgery where
there is no break in technique, no entry into gastro
intestinal, genitourinary, respiratory tracts, non-
traumatic and no evidence of an inflammatory process.
(Kerstein, 1980)
53
It is also difficult to carefully define an
appropriate duration for preoperative antibiotic adminis
tration. Although most guidelines recommend limiting
antibiotic administration to twenty-four hours or less, in
some cases seventy-two hours or less was considered
appropriate. We also respected some of the rationales for
empiric^ antibiotic prescribing that was expressed by some
physicians. Rather than attack these rationales, we
accepted some of them. Less conflict-generating strategies
for influencing medication prescribing should be directed
by changing the prescribers' response to the stimuli to
prescribe and beliefs regarding the perceived outcome of
medication therapy (Helpler, 1982). Keeping this in mind,
we recognized that the ultimate outcome of the research
project would help to correct some of the inappropriate
prescribing of antibiotics due to empiric rationales.
The result of the antibiotic audit demonstrated
that there could be improvements in antibiotic prescribing
in the institution. The Pharmacy and Therapeutics commit
tee recommended to the Medical Executive committee that
concurrent antibiotic review be instituted with corrective
action. The pharmacy department responded by designing a
research project of a quasi-experiment of the interrupted
Empiric antibiotic prescribing--initiation of
antibiotic therapy when there are signs and symptoms of
infection present ; there is no evidence of a prophylactic
motive ; and the results of culture and sensitivity are
unknown.
54
time-series. Pharmacists and physicians were informed that
these activities were carried out to improve patient care
and satisfy the requirements for antibiotic review of the
Joint Commission on the Accreditation of Hospitals (JCAH).
As part of their daily monitoring of medication
therapy, the pharmacists were asked to follow the Concur
rent Antibiotic Review Form Guidelines (see Appendix A) in
documenting total antibiotic orders written and indicating
which of those orders exhibited an uncommon, inappropriate,
or questionable use of an antibiotic. Those orders that
were judged to be inappropriate orders by literature
standards were documented using eleven codes (or measures).
The eleven codes were based on questions - which could be
answered either affirmatively or negatively. The codes
were as follows:
1. A11ergy-Contraindication. Was the patient
allergic to the prescribed antibiotic or was the antibiotic
contraindicated? This code was not expected to be used
frequently, because physicians have already learned over
time to consider allergies and contraindications that
patients have to antibiotics.
2. Route of Administration. Was the route of
administration prescribed the best route (oran or paren
teral) ? In most cases, the site and severity of the
infection dictate the route of administration, therefore
this code was not expected to be used frequently.
55
3. Frequency of Administration. Were the inter
vals between doses adequate to maintain proper serum levels
of the antibiotic? This code was expected to be used
frequently because frequency varies from antibiotic to
antibiotic, especially the cephalosporins. Frequency can
also depend on the severity of the infection.
4. Duration. Was there underuse or overuse in
relation to length of therapy? This code was expected to
be used frequently because of the unfamiliarity that
physicians have with guidelines for prophylactic antibiotic
therapy.
5. Laboratory Tests (culture and sensitivity).
Was a culture and sensitivity requested and, if so, did the
physician use the reported information as a guide to
prescribing the proper antibiotic? This code was not
expected to be used as frequently as four and five.
Results from laboratory tests are received after the
patient has been on antibiotic therapy for twenty-four or
more hours. If the patient is responding well, some
physicians are reluctant to change the order from one that
the identified organism(s) is/are resistant to, to one that
it/they are sensitive to.
6. Laboratory tests (excluding culture and
sensitivity). Were other necessary laboratory tests
ordered, such as serum levels when appropriate?. This code
was not expected to be used much, because pharmacists are
56
involved in ordering laboratory tests for serum levels.
The guidelines that were developed by the medical staff and
pharmacy and therapeutics committee are specific and are
followed closely by the pharmacists for purposes of
monitor ing the response of the patient to the antibiotic
therapy and any untoward reactions, such as ototoxicity,
renal toxicity, etc.
7. Timing. Was the order written in a timely
manner? This code was not expected to be used frequently
because ordering the antibiotic closely follows diagnosis.
This is procedural for all physicians and it has become
rather automatic.
8. Antibiotic Selection. Was the proper antibio
tic selected in reference to the diagnosis or complaint?
This code was expected to be used frequently and have the
highest frequency because of the overuse of antibiotics as
mentioned in the literature review.
9. Dose. Was the proper dose prescribed based on
the diagnosis or complaint? This code was expected to be
used frequently, because of the overuse of antibiotics as
mentioned in the literature review.
10. Side Effects. Did the patient show side
effects to the antibiotic; were the effects documented by
the physician and was the antibiotic order managed accord
ingly? This code was not expected to be used frequently,
because the patient can be uncomfortable from the side
57
effects. Physicians will usually change the antibiotic
when side effects are reported. The reporting mechanism at
this institution involves the physician, nurse, and
pharmacist. With such a check and balance system, most
side effects are documented.
11. Indication Undetermined. This is a miscella
neous category and was used when inappropriate antibiotic
orders did not fit into any of the above categories. This
code was not expected to be used frequently, because from
experience, most orders could be fitted into the above ten
categor ies.
It was expected that, as the study moved through
the various time phases, the frequency in use of all of the
codes would decrease from the time when there was no
intervention to the time period of the last intervention.
In addition to the codes, a determination was made
as to whether the orders were for therapeutic, empiric, or
prophylactic purposes. These determinations affected some
of the codes, such as duration, culture, and sensitivity
and timing.
Collection of Data and Instrumentation
Data were collected on the form shown in Appendix
B. Each form had the capacity for two weeks of data. At
the end of each two-week period, the forms were delivered
to the pharmacy office where they were organized, physi
cians' names were transformed into codes, and the
58
information disseminated to the various medical staff
committees for review.
The pharmacists began documenting this information
in 1983 to test the instrument. The project documentation
formally began on January 1, 1984 and is continuing. Data
were used from January 8, 1984 up to and including March 1,
1986.
Research Design
This research study is of a.quasi-experiment,
interrupted time-ser ies design (Cook & Campbell, 1979).
There were four observations over time on the same individ
uals. The four observations (interruptions) and the
specific times in the series for the observations were:
1. November 4, 1984, a lecture on antibiotics was
given at the institution by an internationally recognized
infectious disease specialist. The lecture was open to
physicians and pharmacists.
2. Six months later, on May 9, 1985, another
lecture on antibiotics was given at the institution by
another internationally recognized infectious disease
specialist. The lecture was again open to physicians and
pharmacists. Within two weeks after this lecture, written
information was distributed to the physicians and pharma
cists containing pertinent information from the lecture.
An audiotape was made of this lecture and made available
for staff members. Both lecturers were asked to stress the
59
p r o p e r u s e o f f i r s t , s e c o n d , and t h i r d g e n e r a t i o n c e p h a l o
s p o r i n a n t i b i o t i c s ( s e e A p p e n d i x C f o r l i s t i n g ) , e m p h a s i z
i n g t h e c o s t d i f f e r e n c e s and t h e f a c t t h a t c e f a z o l i n , a
f i r s t g e n e r a t i o n c e p h a l o s p o r i n a n t i b i o t i c , c o u l d be d o s e d
e v e r y e i g h t h o u r s i n s t e a d o f e v e r y s i x h o u r s i n m o s t
i n s t a n c e s . B o t h l e c t u r e r s c o m p l i e d w i t h t h e r e q u e s t s .
3. In December 1985, pharmacists began actively
seeking the cooperation of physicians in changing antibio
tic orders in two instances. First, if a second or third
generation cephalosporin were prescribed, but a first
generation cephalosporin, which is less costly, would
suffice, the pharmacist would seek to have the order
changed. Secondly, if cefazolin (a first generation
cephalosporin) were ordered to be administered every six
hours, the pharmacist was instructed to get the cooperation
from the physician in modifying the dosing interval to
every eight hours if that dose would provide adequate
coverage.
DiPiro and others (1984) compared seventeen
published studies of parenteral prophylactic cephalosporins
used in surgery. Examination of these studies revealed
little justificatino for preference of one cephalosporin
over another. There was no evidence that the use of a
second or third generation cephalosporin resulted in
postoperative infection rates lower than with first
generation cephalosporins like cefazolin.
60
Two pharmacists who specialize in antibiotic
therapy, in addition to their other responsibilities, were
responsible for providing other staff pharmacists with
information on antibiotic therapy so that their skills were
evenly maintained in this area of practice. Effective
judgment on the use of any medication is not based solely
on a distillation of literature about a given medication,
but also on knowledge of the patient’s disease state and
its dynamics. All physicians and 90 percent of the
pharmacy staff shared this approach to training. Up-to-
date information on drug therapy was available through the
institution's drug information service for any member of
the hospital staff. This included access by computer to
various medical libraries.
C o n t a c t i n g t h e p h y s i c i a n and g e t t i n g h i s / h e r
i m m e d i a t e r e s p o n s e t o c h a n g i n g an a n t i b i o t i c o r d e r c o u l d be
r a t h e r r i s k y and t h r e a t e n i n g t o t h e p h y s i c i a n . S i n c e t h e
m e d i c a l s t a f f m em bers w e r e i n v o l v e d w i t h p h a r m a c i s t s i n t h e
p r o c e s s o f s e l e c t i n g and a p p r o v i n g t h e m e t h o d o f s u r v e i l
l a n c e f o r a n t i b i o t i c u s e , a f i r m f o u n d a t i o n o f t r u s t and
r e s p e c t was e s t a b l i s h e d and b u i l t u p o n .
P e r i o d i c r e p o r t s o f a p p r o p r i a t e a n t i b i o t i c p r e
s c r i b i n g w e r e s h a r e d w i t h t h e m e d i c a l s t a f f d u r i n g t h e t i m e
o f t h e s t u d y . T h i s p o s i t i v e f e e d b a c k h e l p e d t o m a i n t a i n
t h e m e d i c a l s t a f f ' s i n v o l v e m e n t , s u p p o r t , and c o o p e r a t i o n .
61
A value (p<0.01) will be interpreted as highly
significant; a value (p<0.05) as very significant, and a
value (p<0.10) as probably significant.
62
CHAPTER IV
FINDINGS
This chapter will present the results of the
research. This will be accomplished by using frequency
distribution to demonstrate the strength of the measures,
and analysis of variance to demonstrate the statistical
significance of the treatments or interventions.
Data were recorded daily, collected every two
weeks, and tabulated every six weeks. Data were recorded
for each physician by listing the name of the antibiotic
involved and the corresponding number of the measure(s)
depending on the violation.
Seventy-seven physicians were included in the
study. They were randomly plced into groups of seven, thus
forming eleven super-groups that served for statistical
purposes as a unit of one. This modification is sometimes
used when the potential for cells with zero or very small
frequency arises (Lopez-Lee, 1986). This has the advantage
of still capturing data variability to assess treatment
effects via analysis of variance (ANOVA), and gives
variability more stability.
Regarding attendance at lectures, physicians were
separated into two groups: those who attended both
lectures and those who attended no lectures or one lecture.
63
They were again randomly assigned to groups of seven, thus
forming five super-groups for the physicians who attended
both lectures and six super-groups for the physicians who
attended one or zero lectures.
Analysis of variance was performed on the eleven
super-groups and the two super-groups resulting from
attendance or lack of attendance at the lectures.
Strengths of the Measures
The scores for all physicians were aggregated for
the corresponding measure and time period. The scores for
each measure were normalized by multiplying them by a
factor. The factor consisted of the total number of
incidents (violations or errors) divided by the total
number of antibiotic orders written for that time period.
The formula developed was:
AQ = Ç _
^^m ^ TAB
where AS^ = Adjusted score for the measures ; S = raw score;
TI = total number of incidents; and TAB = total antibiotic
orders written for a particular time period.
Raw scores are shown in Table V and adjusted scores
are shown in Table VI.
By Inspection of Tables V, VI, and Figures 1 and 2,
an observation is made that measures three, four, eight,
and nine have the highest cell frequencies. These were
frequency of administration, duration of therapy.
TABLE V RAW SCORES RESULTING FROM THE TOTAL NUMBER OF
INCIDENTS, FOR EACH MEASURE, FOR EACH TIME PERIOD
64
MEASURES
To
No intervention
TIME PHASES
Ti T2
T3
Intervent ions
1 . Allergy/contra
indications 1 0 1 1
2 . Route of admin
istration 6 7 2 1
3. Frequency 65 61 26 23
4. Duration 42 50 35 10
5. Lab (C&S) 31 16 17 6
6 . Lab (exclude C&S) 9 0 1 0
7. Time 2 1 1 0
8 . Antibiotic selection 92 59 37 22
9. Dose 38 42 25 19
1 0 . Side effect 2 1 1 0
1 1 . Undetermined 18 18 18 2
TOTAL ANTIBIOTIC ORDERS 3823 3078 2562 1128
65
TABLE VI ADJUSTED SCORES RESULTING FROM THE TOTAL NUMBER OF
INCIDENTS, FOR EACH MEASURE, FOR EACH TIME PERIOD
MEASURES
TIME PHASES
Tq Ti T2 T3
1. Allergy/contra
indications 0.10 0.00 0.10 0.10
2. Route 0.50 0.60 0.10 0.10
3. Frequency 5.20 5.00 1.50 1.80
4. Duration 3.40 4.10 2.00 0.70
5. Lab (C&S) 2.50 1.30 1.00 0.40
6 . Lab (exclude C&S) 0.70 0.00 0.10 0.00
7. Time 0.20 0.10 0.10 0.00
8 . Antibiotic
selection 7.40 4.90 2.10 1.60
9. Dose 3.00 3.50 1.40 1.40
10. Side effect 0.20 0.10 0.10 0.00
11. Undetermined 1.40 1.50 1.00 0.10
66
JOj iO m.Q In
Figure 1. Bar Graph Depicting the Strength of the Measures
by Showing Which Have the Highest Cell Free
uencxes.
57
lOr
Legend ;
Figure 2. Line Graph Depicting the Ftrength of the
Measures by Showing Which Ones Have the
Highest Ceil Frequencies.
68
antibiotic selection, and dose prescribed. These four
measures were identified in Chapter III as the ones
expected to have the greatest frequencies.
Upon further inspection of Tables V, VI and Figures
1 and 2, an observation is made that there is a decrease in
the number of violations as you move from Tq (no interven
tion) through T-| , T2, and T3 (interventions), except for
measure four (duration) and nine (dose). There was a
slight increase in errors after the first intervention
(lecture). After examining the recordings and notations,
no explanation can be offered from examining the records,
and this increase will here be attributable to chance.
A one-way analysis of variance (ANOVA) with
repeated measures was used to determine to what degree
differences in the population of physicians contributed to
the overall variation in the experimental data. The ANOVA
was performed with the assistance of an IBM-PC, using
Statpak as the statistical software package.
A d j u s t e d s c o r e s f o r e a c h p h y s i c i a n w e r e d e r i v e d by
d i v i d i n g t h e raw s c o r e by t h e p r o d u c t o f t o t a l a n t i b i o t i c
o r d e r s and number o f a d m i s s i o n s f o r e a c h p h y s i c i a n . The
f o r m u l a f o l l o w s :
S
" (TAB) (A)
where ASp is the adjusted score for each physician; S is
the raw score; TAB is total antibiotic orders; and A is
number of admissions for each physician for the concerned
time period.
Each of the four measures showing the highest cell
frequencies were subjected to analysis of variance. The
results are shown in Table VII. We observe that the F
ratio for measures three and four support the hypothesis
that the population means are equal. The F ratio for
measures eight and ine reject the hypothesis that the
population means are equal. Measures eight and nine are
highly significant (£<.05).
R e c o r d s s h o w i n g w h ic h o f t h e s e v e n t y - t w o p h y s i c i a n s
a t t e n d e d t h e l e c t u r e s w e r e u s e d t o s e p a r a t e them i n t o two
g r o u p s . The f i r s t g r o u p c o n s i s t e d o f t h i r t y - f i v e p h y s i
c i a n s who a t t e n d e d b o t h l e c t u r e s ; t h e s e c o n d g r o u p c o n
s i s t e d o f f o r t y - t w o p h y s i c i a n s who a t t e n d e d n o n e o f t h e
l e c t u r e s o r o n e o f th e m . B o t h g r o u p s w e r e s i m u l t a n e o u s l y
s u b j e c t e d t o ANOVA w i t h r e p e a t e d m e a s u r e s . R e s u l t s a r e
show n i n T a b l e VIII.
Factors A (two groups of physicians) and B (time
periods or TQ-T3) are fixed, therefore mean squares within
cell (MS^g) was the denominator for all tests. Using the
.05 level of significance, the critical value for the test
on interaction is F,g5 (3,36)=2.87. The observed F ratios
for measures three, four, eight, and nine (F=0.29, 1.55,
0.54, and 0.962 respectively) are smaller than the critical
TABLE VII RESULTS OF ANOVA PERFORMED ON MEASURES 3, 4,
8 AND 9
70
MEASURES
SUM OF
SQUARES
DEGREES
FREEDOM
OF
MEAN SQUARE
3. Frequency 2.17679E-08 3 7.255968E-09
1.048122E-07 30 3.493742E-09 .
F-Test Ratio: 2.08 F 9^(3,30) = 2.92
4. Duration 8.879738E-09 3 2.959913E-09
9.409102E-08 30 3.136368E-09
F-test Ratio: .94 F g^(3,30) = 2.92
8 . Antibiotic 4.149352E-08 3 1.383117E-08
selection 8.17919E-08 30 2. 726397E-09
F-test Ratio: 5.07 F gg(3,30) = 4.51
9. Dose 5.613515E-08 3 1.871172E-08
8.121532E-08 30 2.707177E-09
F-test Ratio: 6.91 F 9 9 (3 ,3 0 ) = 4.51
71
TABLE VIII SUMMARIES OF ANOVAS WITH REPEATED MEASURES FOR
PHYSICIANS WHO ATTENDED BOTH LECTURES AND PHYSICIANS
WHO ATTENDED 1 OR ZERO LECTURES ■
Measure #3 - Frequency of Administration
Source of
Variation
Sum
of
Squares
df Mean
Squares
F
Ratio
A (two groups
physicians)
of
2.63 1 2.63
-
B (repeated
measures or
Time Phases)
23.39 3 7.79 0.87
AB 7.75 3 2.59 0.29
Within Cell 320.70 36 8.91
Measure #4 - Duration of Therapy
'
A (two groups
of physicians) 16.18 1 16.19
B (repeated
measures or
Time Phases)
9.64 3 3.21 1.04
AB 13.42 3 4.77 1.55
Within Cell 111.15 36 3.088
72
TABLE VIII cont'd.
Measure #8 - Antibiotic; Selection
A (two groups
physicians)
of
13.44 1 13.44 3.29*
B (repeated
measures or
Time Phases)
46.64 3 15.55 3.81*
AB 11.79 3 3.93 0.96
Within Cell 147.03 36 4.08
*p<.05
Measure #9 - Dose of Antibiotic
A (two groups
physicians)
of
7.67 1 7.67 1.30
B (repeated
measures or
Time Phases)
60.12 3 20.04 3.41*
AB 9.49 3 3.16 0.54
Within Cell 211.52 36 5.88
*p<.05
73
value, therefore the data tend to accept the hypothesis
that there is zero interaction.
The test on the main effects for factor B has the
critical value of 2.87. The observed F ratios for measures
three and four (F=.087 and 1.04 respectively) are smaller
than the critical value, therefore the data tend to accept
the hypothesis that there is zero interaction.
The observed F ratios for measures eight and nine
(F=3.81 and 3.41 respectively) are larger than the critical
value, therefore the data tend to contradict the hypothesis
that there is zero interaction. In other words, there is
no variance between the two groups of physicians, but there
is variance between the time phases for mesures eight and
nine. This was also the case with the previous ANOVAS.
For factor A, using the .05 level of significance,
the critical value for the test on interaction is Fg^
(1,36)=4.12. The observed F ratios for measures eight and
nine are F=3.29 and 1.30 respectively. For measure nine,
it is very significant and for measure eight, it is in the
anticipated direction, i.e., the group of physicians who
attended the most lectures tended to benefit most.
Newman-Keuls tests on the difference between all
possible pairs of means were performed on measures number
eight and nine. Tables IX and X show the results.
The mean performance on time phase b^ is shown to
be statistically different from the mean performance on
TABLE IX TESTS ON MEANS OF FACTOR B (TIME PERIODS OR
TQ-T3) USING NEWMAN-KEULS PROCEDURES FOR
MEASURE EIGHT (ANTIBIOTIC SELECTION)
74
ans
bl
b2
bo b3
8.05 9.03 15.38 24.32
bl
b2
bo b3
0.98 7.33 16.27
b2 "6 .35 15.29
bo
''8.94
r SËq .95 (r,36)
4--—----3.12
■3------- 2.84
2------- 2.35
b2
t>3
hi bo b3
** **
** **
**
**Pairs of means considered different.
75
TABLE X TESTS ON MEANS OF FACTOR B (TIME PERIODS OR T0-T3 )
USING NEWMAN-KEULS PROCEDURE FOR MEASURE NINE (DOSE
OF ANTIBIOTIC)
Ordered
bl bo b2 b3
Means
23.35 4.75 6.02 12.15
bl bo b2 b3
r
bl
1 .27^ 7.40 18.60-- --4--
bo
"'6.13 ""17.33--
b2
'"11.20--
bl
bo b2 b3
bl
** **
bo
** **
b2
^Bq .95 (r,36)
3.77
--3.42
2.84
**Pairs of means considered different
76
time phases b] and bp. The mean performance on time phase
b2 is also statistically different from the mean perfor
mance on time phases b] and bp.
The main contributions of these findings are that
there was statistically significant variance among the time
periods and the mean performance on time phase 3 (T3) is
statistically different from the mean performance on time
phases Tq , T-| , and T2.
From December 8, 1985 to March 1, 1986 when
pharmacists directly intervened when the first generation
cephalosporin, cefazolin, was prescribed every six hours
rather than every eight hours, results were favorable. Out
of eleven such orders, physicians changed eight of the
dosing intervals from every six hours to every eight hours.
Pharmacists were also active during this period in
getting physicians to change from prescribing the more
expensive second generation cephalosporins to the less
expensive first generation cephalosporins. As the percen
tage of first generation cephalosporins increased, the
percentage of second generation cephalosporins decreased.
Results are shown in Table IX and Figure 3.
77
Figure 3
per c en t
CEPHALOSPORIN UTIUZATION
BY GENERATION - 1 9 8 5 -1 9 8 6
lOOr
60
40
20
JAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOY DEC JAN
MONTHS
O---- o FIRST GENERATION ❖------ O SECOND GENERATION
Q---- B THIRD GENERATION
78
CHAPTER V
CONCLUSIONS AND POLICY IMPLICATIONS
Through an audit of antibiotic prescribing, it was
confirmed that inappropriate prescribing of antibiotics was
occurring in the study institution. The literature
reported the same in other institutions. Since the
inappropriate precribing of antibiotics may result in the
emergence of drug-resistant organisms, a high frequency of
adverse effects, and excessive costs, a study was designed
to determine the effects of an intervention approach on the
drug prescribing behavior of physicians.
A review of the literature demonstrated that two
interventions had been successfully utilized; they are
medication restriction programs and education programs. An
education intervention was chosen for this study because it
causes less conflict than the medication restriction
intervention and it could have an instaneous effect because
all participants could be exposed at the same time.
Most of the previous studies designated one or more
pharmacists as active intervenors with the prescribing
physicians during the course of the study. Implementation
on a large scale would require an increase in numbers of
pharmacists on the staff of the institutions. In the
present cost-containment environment, this is not a viable
79
option. The studies also showed that the prescribing
behavior changes were not permanent. There was a discon
tinuous effect, that is, one which did not persist over
time.
This study utilized a passive, unobtrusive inter
vention. The presentation of educational programs at the
study institution is ongoing and a pharmacist is a member
of the medical education committee. Designing the educa
tion intervention was a part of this committee's activ
ities. The monitoring of medication therapy has been a
part of the pharmacists' daily activities for many years at
the study institution, therefore the monitoring of antibio
tic therapy became a natural part of the daily routine
without an increase in staffing, nor the necessity of
designating a particular or certain pharmacists to perform
the monitoring and documentation.
The Measures
The eleven measures of physician noncompliance were
determined from guideline-audits of antimicrobial usage
(Kunin, 1977) and an audit of physicians prescribing
behavior at the study institution. The four measures with
the highest cell frequencies were subjected to analysis of
variance (ANOVA) with repeated measures. Measures three
and four did not demonstrate statistical significance in
variation within the time periods. Measures eight and nine
did show variance.
''SO
From the discussion in Chapter III, measures three
and four were expected to show statistical variance, but
they did not. The raw data shows that, though these
measures were frequently used, the change over time was
less than in measures eight and nine. The lack of statis
tical significance in variation within the time periods was
consistent for both ANOVAS for measures three and four.
Likewise, measures eight and nine showed consistency in
their being highly statistically significant in variation
within the time periods.
Success of the Intervention
The variation shown in physician prescribing
behavior over time may be attributable to the intervention.
During the study period, no major threats to validity were
identified. There were few new antibiotics introduced
during the study period. There was little change in the
membership of the medical staff. The change in pharmacists
was minimal. Only one pharmacist was replaced on the staff
during the study and that occurred six months before the
end of the study. The replacement pharmacist had trained
at the institution and was familiar with the procedures for
monitoring medication therapy. No new methods of prescrib
ing were introduced into the medical community during the
time of the study. The interventions began at known points
in time and diffused throughout the relevant population; in
81
other words, exposure time was the same for the study
population.
There was no statistical significance in the
variation between the two physician groups. Differences
were not expected because of the manner in which medicine
is practiced. The physicians constantly refer patients to
each other, they congregate at the institution and share
knowledge on an informal basis during many meetings,
luncheons, etc. Those physicians who attend education
programs communicate the information gained from the
programs to those physicians who were not in attendance.
This is practiced among the pharmacists also. Any new
information gained at conferences, etc. is readily and
promptly shared.
The last intervention, direct intervention by a
pharmacist, demonstrated the acceptance of the pharmacist
as a member of the patient-care team and the pharmacist's
ability to monitor the prescribing behavior of physicians
in an acute care institution without generating conflict.
When pharmacists actively intervened in getting antibiotic
orders changed, there was a decrease in the use of second
generation cephalosporins and a desired increase in the use
of first generation cephalosporins. Also, physicians
cooperated in reducing the dosing frequency of cefazolin.
This active intervention has now become a part of the daily
182
work of the pharmacists and therefore requires no addition
of pharmacists to the staff.
Policy Implications
In the practice of community medicine, physicians
have privileges in several community hospitals. From
hospital to hospital; prescribing practices vary. If
prescribing practices were standardized regionally, there
might be a reduction in inappropriate prescribing. This
study could have far-reaching implications if used on a
regional, state, or national basis.
The policy-making process consists of six basic
phases. They are (a) initiation, (b) estimation,
(c) selection, (d) implementation, (e) evaluation, and - ,
(f) termination (Brewer & deLeon, 1983).
The initiation phase begins with problem recogni
tion or identification. Antibiotic audits are used for
this purpose. These have already been performed, and
reported in the literature. The problem has been iden
tified as the inappropriate prescribing of medication in
the institutional environment. Further studies may be
performed to assist in conceptualizing and sketching out
the rough outlines of the problem, putting together the
information necessary to present a range of possible
options and then beginning to specify policy choices among
the options.
83
The estimation phase involves assigning risks,
costs, and benefits to the various options. The pertinent
question during this phase is, what are the risks, costs,
and benefits attached to an intervention approach to
positively changing physician prescribing behavior? This
present study and the literature have reported that an
education intervention has low risks, can decrease the
costs of medication, and can result in benefits to the
patient.
If the initiation and estimation phases were well
executed, the selection or decision phase will decrease the
tendencies to have multiple, changing, and sometimes
conflicting goals that may be held by those interested in
solving the problem. From the results of this study, the
intervention is almost self-selective. The education
intervention with concurrent monitoring of medication
therapy seems to be the preferred method of sustained,
prescribing behavior change.
Implementation consists of executing the selected
option. The person(s) who performed the various studies
are not necessarily the implementers, therefore communica
tion between them must be attended to carefully.
The evaluation phase is retrospective and is
concerned with measuring outcomes. So far, none of the
studies reported in the literature have had long-lasting
results. Long-term studies are necessary for measuring and
84
evaluating sustained outcomes. For this reason, the
present study will continue.
The terminal phase denotes finality, but this does
not have to be the case. In this phase, one should use the
results of the evaluation phase for adjusting policies and
programs that have become dysfunctional, redundant,
outmoded, or unnecessary.
It is important that studies of this nature
consider the phases of policy-making so that when com
pleted, policy makers and other participants will be able
to make a smooth and successful transition from study to
implementat ion.
Recommendations
This study has demonstrated that physician pre
scribing behavior can be changed or modified, and there are
measurements that were successful in measuring the change.
I recommend that future studies consider the correlation of
improved physician prescribing behavior with a decrease in
nosocomial infections, a decrease in adverse medication
reactions, and a decrease in the cost of medical care.
There have been some studies reported in the
literature regarding reducing the costs of medication.
Ellenor and Frisk (1977) showed a cost savings of $10,000
per year when pharmacists were involved on a team developed
to monitor medication use in an institution for the
mentally retarded. Weintraub (1979) states that medication
85
costs can be reduced by rational prescribing, and educating
the patient (compliance). Gleckman and Gantz (1983) cited
specific examples of how cost-effective antibiotic pre
scribing practices could realize substantial cost savings
without any diminished quality in patient care. Among the
examples were, using guidelines to prescribe prophylactic-
ally, appropriate antibiotic usage, use of therapeutic
equivalents, single agent therapy, oral versus parenteral
route, and duration of treatment. Lastly, Hayman and
Sbravati (1985) demonstrated a $193,172 decrease in
injectable antibiotic expenses over a period of twelve
months. The method used to achieve this goal was anti
biotic restriction.
Health care institutions are logical sites for
studies on patient safety and cost-effective health care to
be performed. The cooperation among physicians and other
members of the patient-care team is continuing to increase.
This study has provided some insight and tools. The future
for health care research can be promising.
REFERENCES
87
Avorn, J., Chen, M. and Hartley, R. 1982. Scientific
Versus Commercial Sources of Influence on the
Prescribing Behavior of Physicians. Am. J. Med.,
73, pp. 4-8 ■ u
Barrière, S.L. 1982. Worldwide Antibiotic Misuse,
an Editorial. Clin. Pharm., 1, pp. 63-4.
Baum, C., Kennedy, D.L., Forbes, M.B. and Jones, J.K.
1985. Drug Use and Expenditures in 1982. JAMA,
253, pp. 382-86.
Bearce, W.C., Coleman, L.T., Kardasinski, D., Kinoshita,
K.E., Louie, R.A., MacMillan, I.R. and Saya, F.
Achieving Pharmacist Proficiency in Pharmacokinetic
Dosing. Paper presented at the 17th Annual American
Society of Hospital Pharmacists Midyear Clinical
Meeting, Los Angeles, December 9, 1982.
Bell, J.E., Bouchard, V.E., South, J.C., and Duffy,
Sister M. G. 1973. A New Approach to Delivering
Drug Information to the Physician Through a Pharmacy
Consultation Program, Part IV: Evaluation Results.
Am. J. Hosp. Pharm, 30, pp. 300-310.
Bennett, S.W. and Scott, A.C. 1980. Computer-assisted
Customized Antimicrobial Dosages. Am. J. Hosp.
Pharm, 37, pp. 523-9.
Berbatis, C.G., Maher, M.J., Plumridge, R.J., Stoelwinder,
J.U. and Zurbrick, S.R. 1982. Impact of a Drug
Bulletin on Prescribing Oral Analgesics in a Teach
ing Hospital. Am. J. Hosp. Pharm., 39, pp. 98-100.
Brewer, G.D. and deLeon, P. 1983. The Foundations
of Policy Analysis. Homewood, IL. The Dorsey
Press.
Briggs, G.G. and Smith', W.E. 1974. Pharmacist-Phys ici an
Drug Consultations in a Community Hospital. Am.
J. Hosp. Pharm., pp. 247-253.
Britton, H.L., Schwinghammer, T.L. and Ramano, M.J.
1981. Cost Containment Through Restriction of
Cephalosporins. Am. J. Hosp. Pharm., 38, pp.
1897-1900.
Brown, J. M., Helling, D.K., Alexander, M.M. and Bur-
meister, L.F. 1982. Comparative Evaluation of
Clinical Pharmacists and Physicians in the Manage-
88
ment of Medication-Related Telephone Calls in
Family Practice Offices. Am. J. Hosp. Pharm.,
39, pp. 437-443.
Burkle, W.S., Lucarotti, R.L. and Matzke, G.R. 1982.
Documenting the Influence of Clinical Pharmacists.
Am. J. Hosp. Pharm., 39, pp. 481-82.
Cook, T.D. and Campbell, D.T. 1979. Quasi-Experimen
tation, Design and Analysis Issues for Field Set-
tings. Boston. Houghton Mifflin Company, pp.
207-209.
Carter, M.J. 1978. Microbiology Laboratories; Noso
comial Infections ; Antibiotics. Ann Int Med,
89, pp. 862-66.
Castle, M., Wilfert, C.M., Thomas, R. and Osterhout,
S. 1977. Antibiotic Use at Duke University Medical
Center. JAMA, 237, pp. 2819-2822.
Chin, R., and Benne, K.D. General Strategies for Effect
ing Changes in Human Systems. 1985. In The Plan
ning of Change, 4th ed., ed. Bennis, W.G., Benne,
K.D., Chin, R. and Corey, K.E. New York: Holt,
Rinehart and Winston, pp. 22-43.
Cluff, L.E. 1985. Controlling Adverse Drug Reactions.
Hosp. Therapy, 10, pp. 41-5.
Cohen, M.S. 1985. Antibiotic Therapy 1985: What
Are The Challenges? Hosp. Formul. December,
pp. 1247-8.
Craig, W.A., Uman, S.J., Shaw, W.R., Ramgopal, V.,
Eagan, L.L. and Leopold, E.T. 1978. Hospital
Use of Antimicrobial Drug. Ann Int Med., 89,
pp. 793-5.
Crossley, K., Gardner, L.C. and the Task Force on Prophy
lactic Antibiotics in Surgery. 1981. Antimicrobial
Prophylaxis in Surgical Patients. JAMA, 245,
pp. 722-726.
Cynamon, M.H. and Kallet, K.K. 1983. Antibiotic Control
in Veterans Administration Medical Centers. Hosp.
Pharm., 18, pp. 1036-8.
D'Arcy, P.F. and Griffin, J.P. 1982. Iatrogenic Dis
eases , 2nd edition, update, 1982. New York :
89
Oxford University Press.
Deutsch, M. 1973. The Resolution of Conflict, Construc
tive and Destructive Processes. New Haven: Yale
University Press.
DiPiro, J.T., Bowden, Jr., Talmadge, A. and Hooks,
III, V.H. 1984. Prophylactic Parenteral Cephalo
sporins in Surgery - Are the New Agents Better?
JAMA, 252, pp. 3277-79.
Durbin, W.A., Lapidas, B. and Goldmann, D.A. 1981.
Improved Antibiotic Usage Following Introduction
of a Novel Prescription System. JAMA, 246, pp.
1796-1800.
Edwards, R. and Adams, D.W. 1982. Clinical Pharmacy
Services in a Pediatric Ambulatory Care Clinic.
Drug Intell. Clin. Pharm., 16, pp. 939-944.
Eickhoff, T.C. 1982. Nosocomial Infections : An Edito
rial. N. Engl. J. Med., 306, pp. 1545-46.
Eisenberg, J.M. and Williams, S.V. 1981. Cost Contain
ment and Changing Physicians' Practice Behavior
- Can the Fox Learn to Guard the Chicken Coop?
JAMA, 246, pp. 2195-2201.
Ellenor, G.L. and Frisk, P.A. 1977. Pharmacist Impact
on Drug Use in an Institution for the Mentally
Retarded. Am. J. Hosp. Pharm., 34, pp. 604-08.
Fink, A., Kosecoff, J., Oppenheimer, P.R., Smith, W.E.
and Brodie, D.C. 1982. Am. J. Hosp. Pharm.,
39, pp. 806-810.
Finkel, M. 1978. Magnitude of Antibiotic Use. Ann.
Int. Med., 89, pp. 791-2.
Freeland, M.S. and Schendler, C.E. 1983. National
Health Expenditure Growth in the 1980's: An Aging
Population, New Technologies and Increasing Competi-
tion. Health Care Financing Review, 4, pp. 1-38.
Friedson, E. 1970. Profession of Medicine, A Study
of The Sociology of Applied Knowledge. New York :
Dodd, Mead & Co.
90
Fry, D.E., Harbrect, P.J. and Polk, Jr., H.C. 1981.
Systemic Prophylactic Antibiotics, Need the Cost
Be So High? Arch. Surg., 116, pp. 466-69.
Gibson, F.M., Hyneck, M.L. and Scherrer, J.J. 1982.
Documented Effectiveness Clinical Pharmacy Serv
ices. Am. J. Hosp. Pharm., 39, pp. 1902-03.
Gleckman, R. and Gantz, N.M. 1983. Cost-Effective
Antibiotic Prescribing. Pharmacotherapy, 3, pp.
239-248.
Green, J.A., Chawla, A.K. and Fong, P.A. 1985. Evalu
ating a Restrictive Formulary System by Assessing
Non-formulary Drug Requests. Am. J. Hosp. Pharm.,
42, pp. 1537-41.
Greenlaw, C.W. 1977. Antimicrobial Drug Use Monitoring
by a Hospital Pharmacy. Am. J. Hosp. Pharm.,
34, pp. 835-38.
Grossing, P.G., Goff, D.A., Kraus, D.M. and Meuller,
C.E. 1984. Development and Validation of an
Instrument to Measure Physicians' Attitudes Toward
the Clinical Pharmacist's Role. Drug Intell.
Cl in. Pharm., 18, pp. 635-640.
Haley, R.W., Culver, D.H., White, J.W., Morgan, W.M.
and Emori, T.G. 1985. The Nationwide Nosocomial
Infection Rate, A New Need for Vital Statistics.
Amer. J. of Epidemiology, 121, pp. 159-67.
Haley, R.W., Culver, D.H., Morgan, W.M., White, J.W.,
Emori, T.G. and Hooton, T.M. 1985. Increased
Recognition of Infectious Diseases in U.S. Hospitals
Through Increased Use of Diagnostic Tests, 1970-
1976. Amer. J. of Epidemiology, 121, pp. 168-81.
Hayman, J.N. and Sbravati, E.C. 1985. Controlling
Cephalosporin and Aminoglycoside Costs Through
Pharmacy and Therapeutics Committee Restrictions.
Am. J. Hosp. Pharm, 42, pp. 1343-7.
Hepler, C.D., Clyne, K.E. and Donta, S.T. 1982. Ration
ales Expressed by Empiric Antibiotic Prescribers.
Am. J. Hosp. Pharm, 39, pp. 1647-55.
Herfindal, E.T., Bernstein, L.R. and Kishi, D.T. 1983.
91
Services
J. Hosp.
on Prescribing
Pharm, 40,
Effect of Clinical Pharmacy
on an Orthopedic Unit. Am.
pp. 1945-51.
Herfindal, E.T., Bernstein, L.R. and Kishi, D.T. 1985.
Impact of Clinical Pharmacy Services on Prescribing
on a Cardiothoracic/Vascular Surgical Unit. Drug
Intell. Clin. Pharm.., 19, pp. 440-444.
Howrie, D.L. and Schwinghammer, T.L. 1985. Cephalo
sporin Use and Restriction in Pediatric Hospitals.
Am. J. Hosp. Pharm., 42, pp. 1576-1578.
Huber, S.L., Patry, R.A. and Hudson, H.D. 1982. Influ
encing Drug Use Through Prescribing Restrictions.
Am. J. Hosp; Pharm., 39, pp. 1898-1901.
Hull, J. H., 111, and Eckel, P.M. 1973. Evaluation
of the Pharmacist as a Drug Therapy Advisor on
Ward Rounds. Am. J. Hosp. Pharm., 30, pp. 687-694.
Johnson, M.W., Mitch, W.E., Heller, A.H. and Spector,
R. 1982. The Impact of an Educational Program
on Gentamicin Use in a Teaching Hospital. Amer,
of J. Med 73, pp. 9-14.
Justiniani, F.R. 1984. Iatrogenic Disease: An Over
view. Mt. Sinai J. of Med., 51, pp. 210-214.
Kass, E.H. 1978. Antimicrobial
Hospitals in Pennsylvania.
89, pp. 800-801.
Drug Usage in General
Ann. of Int. Med.,
Kerstein, M.D. 1980.
Mount Kisco, NY:
Management of Surgical Infections
Futura Publishing Company, "
Klapp, D.L. and Ramphal, R. 1983. Antibiotic Restric
tion in Hospitals Associated with Medical Schools.
Am. J. Hosp. Pharm., 40, pp. 1957-60.
Kunin, C.M. Tupasi, T. and Craig, W.A. 1973. Use
of Antibiotics, A Brief Exposition of the Problem
and Some Tentative Solutions. Ann. Int. Med.,
79, pp. 555-560.
Kunin, C.M., Efron, H.Y. 1977.
Usage : Guidelines for Peer
pp. 1001.
Audits of Antimicrobial
Review. JAMA, 237,
92
_________________________ 1977. Audits of Antimicrobial
Usage: 1. Prophylaxis in Surgery. JAMA, 237,
pp. 1003-8.
Kunin, C.M. 1977. Audits of Animicrobial Usage:
2. Nonsurgical Prophylaxis. JAMA, 237.
1977. Audits of Antimicrobial
Usage: 3. Oral Cephalosporins. JAMA, 237, pp.
1241-42.
1977. Audits of Antimicrobial
Usage: 4. Parenteral Cephalosporins. JAMA,
237, pp. 1243-45.
1977. Audits of Animicrobial
Usage: 5. Oral Carbenicillin. JAMA, 237, pp.
1366-67.
1977. Audits of Antimicrobial
Usage: 6. Parenteral Carbenicillin. JAMA, 237,
pp. 1367-69.
Kunin, C.M. 1977. Audits of Antimicrobial Usage,
Parts 7 & 8. JAMA, 237, pp. 1481-1484.
1977. Audits of Antimicrobial
Usage, Parts 9 & 10. JAMA, 237, pp. 1605-08.
1977. Audits of Antimicrobial
Usage, Parts 11, 12 & 13. JAMA, 237, pp. 1723-1725.
1977. Audits of Antimicrobial
Usage, Parts 14 & 15. JAMA, 237, pp. 1859-1860.
1977. Audits of Antimicrobial
Usage, Parts 16, 17 & 18. JAMA, 237, pp. 1967-1970.
Kunin, C.M. 1978. Problems of Antibiotic Usage:
Definitions, Causes, and Proposed Solutions. Ann.
Int. Med., 89, pp. 802-805.
Kunin, C.M. 1985. The Responsibility of the Infectious
Disease Community for the Optimal Use of Antimicro
bial Agents. J. Infect. Dis., 3, pp. 388-398.
Kunin, C.M. and Chambers, S. 1985. Responsibility
of the Infectious Disease Community for Optimal
Use of Antibiotics : Views of the Membership
of the Infectious Diseases Society of America.
93
Reviews of Infect. Pis., 7, pp. 547-559.
Latorraca, R. and Martins, R. 1979. Surveillance
of Antibiotic Use in a Community Hospital. JAMA,
242, pp. 2585-2587.
Levin, B., Cohen, S.S. and Birmingham, P.H. 1981.
Effect of Pharmacist Intervention on the Use of
Serum Drug Assays. Am. J. Hosp. Pharm., 38, pp.
845-851.
Lipman, A.G., Devenport, J.K. and Page, B.C. 1982.
Analysis of Information Provided by Clinical Pharma
cists. Am. J. Hosp. Pharm., 39, pp. 71-3.
Lopez-Lee, D. University of Southern California, School
of Public Administration, Los Angeles, California,
Personal Communication, 7 April 1986.
Ma, M.Y., Goldstein, E.J.C. and Meyer, R.D. 1979.
Effect of Control Programs on Cefazolin Prescribing
in a Teaching Hospital. Am. J. Hosp. Pharm.,
36, pp. 1055-58.
MacCosbe, P.E. and Gartenberg, G. 1985. Modifying
Empiric Antibiotic Prescribing: Experience With
One Strategy in a Medical Residency Program. Hosp.
Form., 20, pp. 986-999.
Mayer, K.H. and Zinner, S.H. 1985. Bacterial Pathogens
of Increasing Significance in Hospital-Acquired
Infections. Reviews of Infect. Dis., Suppl. #3,
pp. S371- S379T"
Mehl, B. 1986. Collaborative Efforts Succeed in Restric
ting Antibiotic Use - An Interview. Drug Utiliza
tion and Review, February, pp. 25-27.
McDonald, C.J. 1976. Protocol-Based Computer Reminders,
The Quality of Care and the Non-Perfectability
of Man. New Engl. J. Med., 295, pp. 1351-55.
McGowan, Jr., J.E. and Finland, M. 1974. Usage of
Antibiotics in a General Hospital: Effect of
Requiring Justification. J. Infect. Dis., 130,
pp. 165-168.
McGowan, Jr., J.E. 1982. Whence Come Nosocomial Infec
tions? New Engl. J. Med., 307, pp. 1576-77.
94
McGowan, Jr., J.E. 1985. Changing Etiology of Nosocomial
Bacteremia and Fungemia and Other Hospital-Acquired
Infections. Rev. Inf. Pis., Suppl., pp. S357-S370.
Moss, J., Wyatt, G., Christopherson, D. and Routh,
S. 1982. A Prospective Drug Utilization Review
on the Prescribing of Oral and Parenteral Cephalo
sporins ,^Io£p_^_Form. , 17, pp. 1589-1601.
Neu, H.C. 1985. Antimicrobial Pharmacology : Is It
Possible to Use Antibiotics Cost-Effectively with
Attention to Antimicrobial Activity, Bacterial
Resistance and the Pharmacologic Properties of
the Agents? An editorial, J. Clin. Pharmacol.,
25, pp. 65-66.
Nickman, N.A., Blissenbach, H.F. and Herrick, J.D.
1984. Medical Committee Enforcement of Policy
Limiting Postsurgical Antibiotic Use. Am. J.
Hosp. Pharm., 41, pp. 2053-6.
Norris, S.M. 1982. Evaluation of Gentamicin Prescribing
After Drug-Use Review. Am. J. Hosp. Pharm., 39,
pp. 1592-1530.
Pastel, D.A. 1985. The Pharmacist in Antibiotic use
Review and Cost Reduction. Calif. Soc. of Hosp.
Pharm. Voice, 12, pp. 11-13.
Peterson, C.D. and Lake, K.D. 1986. Reducing Prophylactic
Antibiotic Costs in Cardiovascular Surgery: The
Role of the Clinical Pharmacist. Drug Intel1.
Clin. Pharm., 20, pp. 134-7.
Porter, J. and Jick, H. 1977. Drug Related Deaths
Among Medical Inpatients. JAMA, 237, pp. 879-881.
Powell, S.H. and Lantos, R.L. 1981. Concurrent Pharmacist
Monitoring of Gentamicin Therapy. Am. J. Hosp.
Pharm., 38, pp. 77-78.
Rapp, R.P., Bannon, C.L. and Bivins, B.A. 1982. The
Influence of Dose Frequency and Agent Toxicity
on the Cost of Parenteral Antibiotic Therapy.
Drug Intel1. Clin. Pharm., 16, pp. 935-44.
Roberts, A.W. and Visconti, J.A. 1972. The Rational and
Irrational Use of Systemic Antimicrobial Drugs.
Amer. J. Hosp. Pharm., 29, pp. 828-834.
95
Rubenstein, E, 1973, Continuing Medical Education
at Stanford: The Back-to-Medical-School Program.
J♦ Med. Educ., 48, pp. 911-918,
Ruhright, R. 1984. Persuading Physicians, A Guide
for Hospital Executives. Rockville : Aspen Systems
Corp,
Saya, F. , Coleman, L. and Martinoff, J.T. 1985. Pharmacist!
Directed Heparin Therapy Using a Standard Dosing
and Monitoring Protocol. Am. J, Hosp, Pharm.,
42, pp. 1965-1969,
Scheckler, W.E, and Bennett, J.V, 1970. Antibiotic
Usage in Seven Community Hospitals. JAMA, 213,
pp. 264-267,
Schroeder, N.H., Caffey, Jr., E.M. and Lorei, T.W.
1979. Antipsychotic Drugs; Can Education Change
Prescribing Practices? J. Clin. Psych., 40, pp.
186-89.
Schweigert, B.F., Oppenheimer, P.R. and Smith, W.E.
1982. Hospital Pharmacists as a Source of Drug
Information for Physicians and Nurses, Am. J.
Hosp. Pharm., 39, pp. 74-77.
Sesin, G.P. and Baptista, R.J, 1984, Utilizing the
Clinical Pharmacist to Modify Tobramycin Dosage
Regimens: Accuracy of Prediction and Potential
Cost Effectiveness. Hosp. Pharm., 19, pp. 806-810.
Sohn, C.A., Wolter, H.A. and McSweeney, G.W. 1980
Effectiveness of a Cephalosporin Education Program
- A Pharmacy Education Program. Drug Intel1,
Cl in. Pharm., 14, pp. 272-277.
Soumerai, S.B. and Avorn, J. 1984. Efficacy and Cost-Con
tainment in Hospital Pharmacotherapy : State of
the Art and Future Directions. Milbank Memorial
Fund Quarterly, 62, pp. 447-474,
Starr, P. 1982, The Social Transformation of American
Medicine. New York : Basic Books, Inc.
Steel, K., Gertman, P.M., Crescenzi, C. and Anderson, J.
1981. Iatrogenic Illness on a General Medical
Service at a University Hospital. New Engl. J.
Med., 304, pp. 638-42,
96
Strahl, D.A. and Wightkin, W.T. 1985. Improvement
in Antibiotic Use Secondary to a Small Hospital
Antibiotic Review. Hosp. Form., 20, pp. 310-12.
Tenney, J.H. 1978. Nosocomial Infections and the
Regulated Hospital. Ann. Int. Med., 89, p. 862.
Washington, 11, J.A. 1985. Criteria Sought for Ordering
Antibiotics Without Running Culture and Sensitivities.
Drug Util. Rev., pp. 83-84.
Weinstein, R.A. 1985. Gram-Negative Bacteria in Hospitals
Trends in Resistance and Control. J. Nosocomial
Infect., pp. 2-5.
Wennberg, J. and Gittelson, A. 1982. Variations in
Medical Care Among Small Areas. Scientific Amer.,
246, pp. 120-134.
Wennberg, J. and Caper, P. 1985. An Interview: Medical
Practice: Why Does It Vary So Much? Hospitals,
March 1, pp. 88-90.
White, J.P. 1985. Bacterial Resistance on the Rise,
But New Drugs Are Countering It. Drug Topics,
129, pp. 60-66.
Williams, Jr., H.B. and Sykora, J. 1972. Communicating
in the Patient Care Arena. Paper presented at
the American Society of Hospital Pharmacists Annual
Meeting, New Orleans, March.
Williams, Jr., H.B., DeFazio, D.J. and Kinoshita, K.E.
1979. The Evolution of Comprehensive Clinical
Pharmacy Services in a Community Hospital. Paper
presented at the annual meeting of the American
Pharmaceutical Association, Anaheim, CA, April.
Williams, Jr., H.B., DeFazio, D.J. and Kinoshita, KE.
1980. Quality Assurance Through Peer Review.
Paper presented at the annual meeting of the American
Pharmaceutical Association, Washington, DC, April.
Witte, K.W., Nelson, Jr., A.A. and Hutchinson, R.A.
1980. Effect of Pharmacist Consultation on Rational
Antimicrobial Therapy. Amer. J. Hosp. Pharm.,
37, pp. 829-32.
'97'
Wraith, S.M., Aikins, J.S., Buchanan, B.C., Clancey,
W.J., Davis, R., Fagan, L.M., Hannigan, J.F,,
Scott, A.C., Shortliffe, E.H., van Melle, W.J,,
Yu, V.L., Axline, S.G. and Cohen, S.N. 1976.
Computerized Consultation System for Selection
of Antimicrobial Therapy. Amer. J. Hosp. Pharm.,
33, pp. 1304-08.
98
APPENDIX A
CONCURRENT ANTIBIOTIC REVIEW FORM GUIDELINES
99
CONCURRENT ANTIBIOTIC REVIEW FORM GUIDELINES
OBJECTIVE ; To identify uncommon usage of antibiotics for pharmacy
and medical staff review.
PROCEDURE ;
1. The Concurrent Antibiotic Review form will be maintained in
each satellite area with the CASH Report notebook.
2. On Monday (time card day) the cumulative data from the preceding
two weeks should be turned in with your CASH reports.
3. Monthly or bimonthly the data will be reviewed and sent to
the affected medical staff committees for additional input.
4. The affected medical staff committees will be responsible
for reporting to the P&T Committee and Infection Control Com
mittee their findings and action taken.
5. The medical staff will be responsible for educational programs
or guidelines which will alter the uncommon prescribing habits
identified.
GUIDELINES FOR USE OF FORM;
1. A line is left blank at the beginning of each day to record
all new antibiotic orders. A new antibiotic order is defined
as a new drug order. DO NOT record;
a. change in frequency of dose.
b. change in dose.
c. resumed antibiotic post-op.
d. reorder antibiotic after transfer to another unit.
Each time you dispense a "new" antibiotic order, make a pencil
mark (see A). If you dispense a "new" order from an area
that does not normally service the affected area, a note should
be left for the affected satellite so that the daily dispensing
total can be adjusted accordingly. Example:
Main pharmacy evening shift dispenses new order for oncology
area. Leave note for Oncology Satellite with date and number
of new orders. No additional information is needed.
2. At the beginning of the next day record the total number of
new antibiotic orders from the previous day (see B).
100
CONCURRENT ANTIBIOTIC REVIEW FORM GUIDELINES cotit'd.
3. Each time you determine an uncommon, inappropriate, or question
able use of an antibiotic you should record;
a. the date (see C) you determined the problem.
b. patient's name (see D).
c. room # (see E).
d. prescribing physician (see F).
e. questionable antibiotic (see G).
f. the code which identifies your concern (see H).
g. "T", if antibiotic considered therapeutic or empiric.
"P", if antibiotic considered prophylactic (see I).
h. initials of pharmacist recording information (see J).
i. comments regarding the therapy or any interaction you
you may have had with the physician. example (see K):
1) dose too high, MD reduced.
2) frequency should be q8° vs q4°.
3) duration two weeks post-op, no infection noted by
MD.
101
APPENDIX B
DATA COLLECTION FORM
102
cc
c
< ï
z
0 £
O
u .
_ l
« t 3
« _ >
z >
0 £ LU
LU o:
3 :
H - o
O
O H "
( / ) O
ü _ OÙ
O
) —
z
e t <
t —
O . z
( / )
o o :
I z
3
t —
L T i z
o
O
L U
O
3 :
h —
Z:
00
XJ
§
LU « Qj
u * - >
II) OJ -r- QJ
"O -D X»
a
o
c U
1°
Z Oi
u. o
u
z;
t
V
iJ
\ ^
T z r
g
103
APPENDIX C
INTRAVENOUS CEPHALOSPORINS APPROVED
FOR LISTING IN THE FORMULARY
TOT
INTRAVENOUS CEPHALOSPORINS IN THE FORMULARY
First Generation: Cephalothin (Keflin, Seffin)
Cefazolin (Ancef, Kefzol)
Second Generation : Cefoxitin (Mefoxin)
Cefuroxime (Zinacef)
Third Generation: Ceftizoxime (Cefizox)
Cefoperazone (Cefobid)
Ceftazidime (Fortaz)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

An experimental study of the effect of "human interest" factors on listenability

PDF

The effects of centralization and board membership reform on health occupational licensing policies

PDF

A comparison of attitudes of university seniors and sales executives toward outside selling, as revealed by clinical interview and questionnaire methods

PDF

Realism in the fiction of Frances Burney

PDF

The effects of various factors on the academic achievement of Vietnamese students at Gold Coast College

PDF

Earnings differences among Black, White and Hispanic males and females: the impact of overeducation, undereducation and discrimination

PDF

A generalized scale for the measurement of attitudes toward international issues

PDF

A critical analysis of goodwill arising from consolidation and the varying effects on financial statements

PDF

Student authoring in the American grain

PDF

The functional gene expression dynamics of Shewanella oneidensis MR-1 under various external voltage potentials

PDF

Making it to the top: a study of Black partners in major law firms

PDF

Development, validation and testing of a new sensor array for intra-articular pressure measurement: in-vitro human lumbar spine intra-articular facet testing