International Journal for Quality in Health Care 2004; Volume 16, Number 3: pp. 245–251
10.1093/intqhc/mzh033
Physician knowledge and adherence
to prescribing antibiotic prophylaxis
for sickle cell disease
KEELE E. WURST AND BETSY L. SLEATH
University of North Carolina School of Pharmacy and School of Public Health, University of North Carolina, Chapel Hill, NC, USA
Abstract
Objective. The purpose of this research was to examine how physician characteristics were associated with: (i) physician
knowledge of and adherence to sickle cell guidelines; and (ii) the types of educational programs about sickle cell disease desired
by physicians.
Methods. A survey was developed to assess the research objective. After the survey was pre-tested and an institutional review
board exemption was obtained, it was sent to a systematic random sample of 375 pediatricians and all 125 practicing hematologists
in North Carolina. They were asked to answer a six-item knowledge test relating to the antibiotic prophylaxis guidelines.
Results. The response rate was 57%, of which 61% were pediatricians. Over half (56%) were in a practice with at least one
pediatric sickle cell patient. Fifty-nine percent of physicians answered five or more questions correctly on the knowledge test.
The question most physicians answered correctly (97%) pertained to the necessity of antibiotics for children with sickle cell disease.
The question most frequently answered incorrectly (62%) pertained to prescribing antibiotics to a child with unconfirmed
sickle cell disease. Logistic regression results indicated that the number of sickle cell patients seen in practice influenced the
number of questions answered correctly. Sixty-six percent of physicians prescribed prophylactic antibiotics for 100% of their
patients with sickle cell disease and therefore were 100% adherent. Eighty-one percent of pediatricians compared with 12% of
hematologists were 100% adherent in prescribing antibiotics. Hematologists and those practicing at a medical school or university
were less likely to be 100% adherent in prescribing antibiotic prophylaxis.
Conclusion. The majority of physicians surveyed were relatively knowledgeable about sickle cell guidelines, however there may
be a need for continuing education programs that focus on the issues of prescribing antibiotics to a child with unconfirmed
sickle cell disease and penicillin dosage.
Keywords: antibiotic prophylaxis, continuing education, physician adherence, physician knowledge, sickle cell disease
Sickle cell disease is a potentially fatal disease, striking one in
every 375 African-American children [1]. Children with sickle
cell disease under the age of 5 years are at great risk for fatal
invasive infection with Streptococcus pneumoniae and Haemophilus
influenzae due to the inability of their spleen to protect against
infection [2,3]. Randomized, controlled trials have demonstrated
that a twice-daily dose of penicillin prophylaxis until the age of
5 years reduces the incidence of septicemia by 84% [4]. Studies
have indicated that the rate at which sickle cell patients receive
antibiotic prophylaxis is typically less than optimal [5–7]. There
are two main reasons why patients do not receive antibiotic
prophylaxis: (i) physicians do not prescribe the antibiotics; and
(ii) the patient does not take the antibiotics. This study focuses
on physician prescribing of antibiotic prophylaxis.
Clinical practice guidelines to prescribe antibiotic prophylaxis for patients with sickle cell disease have been created and
disseminated in the hope that physicians will follow them as
stated to ensure best clinical practice. However, many physicians
are unaware of or are not influenced by the guidelines [8]. There
have been many reasons postulated as to why physicians do
not follow guidelines.
Factors that have been shown to positively influence physician
adherence to guidelines include female gender, younger age,
awareness and agreement with guidelines, and larger practice
size [9–11]. In a study performed to assess adherence to
guidelines regarding cancer screening, the physician characteristics that influenced adherence to guidelines included:
agreement with guidelines, level of continuing medical education,
and perceived probability of the disease [9]. The characteristics associated with disagreement with guidelines, and consequently non-adherence, included older age, male sex, and not
having completed a postgraduate residency program [9]. Practice
Address reprint requests to Keele E. Wurst, University of North Carolina at Chapel Hill, School of Pharmacy, Beard Hall,
CB #7630, Chapel Hill, NC 27599-7630, USA. E-mail: [email protected] or [email protected]
International Journal for Quality in Health Care vol. 16 no. 3
© International Society for Quality in Health Care and Oxford University Press 2004; all rights reserved
245
K. E. Wurst and B. L. Sleath
characteristics that affected adherence to guidelines included
type of practice (group or solo), payment system, perceived
time to perform the test, access to current information, and
the costs of the testing [9]. Physicians may also not follow the
guidelines by prescribing incorrectly for the disease, such as
prescribing antibiotic prophylaxis therapy for a patient with
sickle cell trait instead of sickle cell disease. There are many
causes of inappropriate prescribing, including the physician’s
knowledge base (the physician’s knowledge on up-to-date
therapeutics) and physician practice patterns (type of practice,
payment schedule, number of physicians in practice) [12].
Some of these causes may contribute to lack of adherence to
the guidelines and thus inadequate prescribing for sickle cell
disease.
Physician awareness of guidelines, agreement with guidelines, and knowledge of the content of the guidelines are also
important in assessment of adherence to guidelines. In a study
of adherence to guidelines for cesarean section, Lomas et al.
demonstrated that awareness and agreement with guidelines
did not lead to adherence to the guideline. Poor knowledge of
the contents of the guidelines was associated with poor guideline
adherence [13].
We are not aware of any studies performed with the intention of understanding physicians’ attitudes toward the guideline for prescribing for sickle cell patients and examining the
physicians’ adherence to the guideline for prescribing sickle
cell prophylaxis. Understanding physician characteristics that
are associated with adherence to guidelines can help to create
better educational programs for physicians. Therefore, the
purpose of this study was to examine: (i) how physician
gender, age, awareness of the North Carolina Sickle Cell Syndrome Program (NCSCSP), and practice size were associated
with physician knowledge of sickle cell guidelines to prescribe
prophylactic antibiotics for sickle cell disease; and (ii) to
describe how physician gender, age, practice specialty, physician knowledge, and number of sickle cell patients contributed
to physician adherence to antibiotic prophylaxis-prescribing
guidelines.
Methods
The study was a cross-sectional survey of North Carolina
hematologists, hematology/oncologists, and pediatricians.
Information on the providers was obtained from the 1999
North Carolina Health Professions Data System [14]. A
knowledge questionnaire, which consisted of six items relating to sickle cell disease and adherence to antibiotic prophylaxis, was included in the survey.
After obtaining an institutional review board (IRB) exemption, all questions included in the survey were pre-tested on
three physicians (a hematologist, a pediatrician, and a general
practitioner) and were subsequently modified for clarity. The
survey questionnaire for the study is available as Supplementary data, available at IJQHC Online. The survey was sent to
500 physicians: all 125 hematologists/oncologists practicing
in North Carolina and a stratified random sample of 375 pediatricians. The sampling frame included all pediatricians who
246
practiced in a North Carolina county where >20% of the
population was non-white in order to target those who most
likely cared for patients with sickle cell disease. The pediatricians
included in the sampling frame were sampled by random
number until a sample of 375 was obtained. Four weeks after
the survey had been mailed, the survey was re-sent to nonresponders. A monetary incentive of US$1 was included with
the survey to improve the response rate [15].
Measures
Physician age, years in practice, number of sickle cell patients
in practice, and number of sickle cell patients <5 years of age
in practice were measured as continuous variables. Physician
gender, race (white and non-white), and practice specialty
(pediatrics and hematology) were measured as dichotomous
variables. Practice setting was measured as a categorical variable (group office, hospital setting, medical school, and other).
Physicians were first asked demographic questions and if
they cared for sickle cell patients under the age of 5 years.
Physicians who did care for sickle cell patients were asked
‘For what percentage of sickle cell disease patients seen in
practice have you prescribed antibiotic prophylaxis?’. Since
the guideline states that a patient with sickle cell disease
should be prescribed antibiotic prophylaxis, this question was
used as a surrogate measure of physician adherence to guidelines [1]. Physicians were asked a series of six questions to
determine their knowledge of the contents of the sickle cell
disease guidelines (Table 1). Physicians were also asked to rate
their method of preference for receiving continuing education
on sickle cell disease.
Analysis
The survey data were analyzed using SPSS statistical software
(SPSS version 10, 2000; SPSS, Chicago, IL). Bivariate analyses, including χ2 and t-tests, were conducted to assess the
relationships between the physician demographic and practice
characteristics, physician knowledge, and physician adherence to the guidelines. Multivariate logistic regressions were
conducted to: (i) predict if the number of sickle cell patients
seen, practice specialty, race, years in practice, gender, and
awareness of the sickle cell program influenced physician
knowledge of the sickle cell guidelines; and (ii) determine if
age, practice setting, practice specialty, years in practice, and
physician knowledge were associated with physician adherence to the sickle cell guidelines.
Results
Of the 500 surveys mailed, 64 (50 pediatricians and 14 hematologists) were returned because of incorrect addresses and
were discarded. Two hundred and forty-eight responses were
returned (overall response rate = 57%). Among pediatricians
the response rate was 59% (n = 191), and among hematologists the response rate was 51% (n = 57). Table 2 presents
respondent characteristics. The majority of respondents were
Knowledge of sickle cell disease
Table 1 The percent and number of physicians that answered each question correctly (n = 142)
Question
Physicians answering correctly % (n)
Is it necessary for children <5 years of age with SCD to be on antibiotic prophylaxis?
Correct answer: yes
Should a child be on antibiotic prophylaxis if they have been diagnosed with SCD?
Correct answer: yes
Should a child with sickle cell trait be on antibiotics?
Correct answer: no
Should a child with suspected SCD be on antibiotics?
Correct answer: yes
The dose of penicillin under the age of 3 years is 125 mg b.i.d.
Correct answer: true
The dose of penicillin over the age of 3 years is 250 mg b.i.d.
Correct answer: true
Answered all questions correctly
Answered five questions correctly
Answered four questions correctly
Answered three questions correctly
Answered two questions correctly
Answered one question correctly
Answered no questions correctly
97.1 (138)
.........................................................................................................................................................................................................................
78.8 (112)
96.4 (137)
49.2 (70)
76.7 (109)
73.9 (105)
26.7 (38)
37.3 (53)
12.7 (18)
9.2 (13)
5.6 (8)
1.4 (2)
7.0 (10)
SCD, sickle cell disease.
white, male, aged from 30 to 76 years (mean 47 years), and
specialized in pediatrics. The characteristics of gender, race,
age, practice setting, and practice specialty of the respondents
and non-respondents were not significantly different. Over
half (137 out of 248) of the responding physicians cared for
sickle cell patients under the age of 5 years. In the six-item
test, a physician was considered to have complete knowledge
of the sickle cell guidelines if he/she answered five or six of
the six questions correctly. Table 1 demonstrates the number
of questions physicians answered correctly. Sixty-four percent
of physicians answered five or more questions correctly. The
question most physicians answered correctly (96.4%)
addressed the necessity of antibiotics for children with sickle
cell disease. Ninety-seven percent of physicians knew that
children with sickle cell disease trait should not be on antibiotic prophylaxis. The question most frequently answered
incorrectly (50.8%) pertained to prescribing antibiotics to a
child with unconfirmed sickle cell disease. Questions about
penicillin dosages were also frequently answered incorrectly.
For all questions answered, a greater percentage of physicians
answered the question correctly than answered incorrectly.
The results of the logistic regression indicated that pediatric
specialty was associated with answering five or six out of six
questions correctly. Table 3 presents the multivariate regression
statistics. Pediatric specialty was associated with whether the
physician was knowledgeable about the sickle cell guidelines
in both the bivariate and the multivariate analysis. Pediatricians were more likely than hematologists to answer five or
six out of six questions correctly. Gender was significantly
associated with the number of questions answered correctly
in the bivariate analysis, but was not significantly associated
when other variables were added to the model. An interaction
term of practice specialty and number of sickle cell disease
patients seen in practice was statistically significant in the
multivariate model. This term indicates that a hematologist
who saw more patients with sickle cell disease was more likely
to answer five or six out of six questions correctly than hematologists who saw fewer patients with sickle cell disease.
Number of sickle cell disease patients in practice, years in
practice, race, and awareness of the North Carolina sickle cell
syndrome program were not associated with physician knowledge of the guidelines.
Sixty-six percent of physicians stated that they prescribed
prophylactic antibiotics for 100% of their patients with sickle
cell disease and were therefore 100% adherent. Five percent
were 75–99% adherent, 3% were 25–49% adherent, 2% were
1–24% adherent, and 21% reported that the question was not
applicable because they did not prescribe antibiotic prophylaxis. Eighty-one percent of pediatricians compared with 12%
of hematologists were 100% adherent in prescribing antibiotics
(χ2 = 46.12; P = 0.000). Seventy-five percent of physicians in
group practice, 70% of those in hospital practice, 12.5% of
those in medical school practice, and 69.2% of those in other
practice settings were 100% adherent in prescribing antibiotics
(χ2 = 12.94; P = 0.005). Eighty-four percent of females and
56.9% of males were 100% adherent in prescribing antibiotics
(χ2 = 11.34; P = 0.001). Younger physicians were significantly
more likely to be 100% adherent in prescribing antibiotics
compared with older physicians (t = −2.16).
Table 4 presents the knowledge questions alongside the percentage of physicians who answered the questions correctly and
were also 100% adherent in prescribing antibiotic prophylaxis.
247
K. E. Wurst and B. L. Sleath
Table 2 Respondent characteristics (n = 248)
Characteristic
Percent (n)
..........................................................................................................
Gender
Male
Female
Unknown
Race
White
Black
Asian
Other
Unknown
Age
Mean
Range
Practice setting
Group practice
Medical school or university
Hospital
Other
Unknown
Practice specialty
Pediatricians
Hematology/oncology
Other
Unknown
Number of patients <5 years of age
with SCD seen in practice
Zero
1–2
3–5
6–15
≥16
Missing
Mean
Range
Years in practice
Mean
Range
56.9 (141)
40.7 (101)
2.4 (6)
81.1 (201)
4.4 (11)
6.0 (15)
6.0 (15)
2.4 (6)
47 years
30–76 years
30.3 (75)
11.7 (29)
9.3 (23)
12.1 (30)
36.7 (91)
60.9 (151)
19.3 (48)
11.3 (28)
8.5 (21)
30.6 (76)
14.1 (35)
13.7 (34)
12.5 (31)
16.1 (40)
12.9 (32)
14.9 patients
(0–500)
13.8 years
(1–48 years)
SCD, sickle cell disease.
Table 5 presents the results of the multivariate logistic
regression predicting physician adherence to guidelines.
Pediatricians were significantly more likely to be 100% adherent in prescribing antibiotic prophylaxis than hematologists.
Physician knowledge of antibiotic prophylaxis-prescribing
guidelines was also shown to be associated with physician
adherence to prescribing antibiotic prophylaxis. The number
of questions answered correctly was significantly associated
with physician adherence to guidelines in the bivariate and
multivariate analyses.
Those that practice in a medical school or university were
significantly less likely to be 100% adherent in prescribing
antibiotic prophylaxis. Gender was significantly associated
248
with adherence in prescribing antibiotic prophylaxis in the
bivariate analysis, but not in the multivariate analysis. Years in
practice was not associated with a physician being 100%
adherent in prescribing antibiotic prophylaxis.
Thirty-eight percent of physicians thought that continuing
education programs about sickle cell disease would be most
useful in helping them learn more about sickle cell disease,
32% thought mailings on sickle cell disease would be most
useful, 13% thought that an e-mail address where the physician could ask questions would be most useful, and <1%
thought that it would be useful to have a sickle cell educator
contact them at their practice. Thirteen percent had no need
for further education in the area of sickle cell disease.
Discussion
Sixty-four percent of the physicians sampled answered five or
six out of six questions correctly. This shows that the majority
of physicians surveyed were relatively knowledgeable about
sickle cell guidelines. The current guidelines state that a child
with suspected sickle cell disease should be placed on antibiotics. Many physicians disagreed. This issue needs to be
examined further to assess the reasons behind this finding.
The emergence of penicillin-resistant pneumococcal infection
in children with sickle cell disease may play a role in physician
disagreement with guidelines. There has been concern that
prophylaxis may favor the development of resistant organisms, therefore some physicians may decide to wait until
sickle cell disease is confirmed before prescribing antibiotic
prophylaxis [16].
A frequently missed question addressed correct dosages of
penicillin for prophylaxis for children under the ages of 5 and
3 years. Many physicians did not know the correct dosage or
disagreed with the current recommendations. Pediatric specialty and hematology practices with a greater number of
sickle cell disease patients were both significantly associated
with answering five or six out of six questions correctly. In
this sample, pediatricians were more likely to see patients with
sickle cell disease. Therefore this would explain why those in
pediatric practice were more likely to be knowledgeable about
the guidelines. It is also likely that a hematologist with more
patients with sickle cell disease would likely be more knowledgeable about the guidelines. Studies assessing adherence to
guidelines showed that older age, male gender, and practice
size contributed to non-adherence to guidelines [9]. These
factors were not associated with knowledge of guidelines.
This may have been due to the way knowledge was measured,
as knowledge was only measured by six specific questions.
The questions on the survey may not have reflected the true
knowledge level of the physician.
This study found that gender was significantly associated
with adherence to guidelines in the bivariate model. This
study found a significant relationship between practice specialty and adherence to guidelines, as 81% of pediatricians
compared with 12% of other practice specialties indicated
that they were 100% adherent in prescribing antibiotics. In
this subset of physicians surveyed, pediatricians cared for most
Knowledge of sickle cell disease
Table 3 Multivariate logistic regression results predicting the physician characteristics associated with answering five or more
questions correctly (n = 142)1
Beta
Significance
Odds ratio
Lower 95% CI
Upper 95% CI
0.001
−3.80
−0.299
−0.046
0.017
0.575
0.860
0.000 2
0.586
0.096
0.970
0.209
1.00
0.022
0.742
0.955
1.02
1.78
0.992
0.003
0.253
0.905
0.411
0.725
1.01
0.147
2.17
1.01
2.52
4.36
0.024
0.039 2
1.03
1.00
1.05
.........................................................................................................................................................................................................................
Number of SCD patients
Pediatricians
Race
Years in practice
Gender
Aware of the North Carolina Sickle Cell
Syndrome Program
Interaction term of pediatricians and
number of SCD patients
CI, confidence interval; SCD, sickle cell disease.
1
Cox and Snell, r 2 = 0.214.
2
Significant at P < 0.05 in the multivariate model.
Table 4 Number of physicians who were 100% adherent to prescribing guidelines and answered each knowledge question
correctly (n = 141)
Question
Physicians with correct knowledge and 100%
adherence to prescribing guidelines, % (n)
.........................................................................................................................................................................................................................
Is it necessary for children <5 years of age with SCD to be on
antibiotic prophylaxis?
Should a child be on antibiotic prophylaxis if they have been
diagnosed with SCD?
Should a child with sickle cell trait be on antibiotics?
Should a child with suspected SCD be on antibiotics?
The dose of penicillin for children <5 years of age is 125 mg b.i.d.
The dose of penicillin for children >3 years of age is 250 mg b.i.d.
Answered all questions correctly
68.7 (97)
53 (75)
68.7 (97)
35.4 (50)
59.5 (84)
56.0 (79)
21.2 (30)
SCD, sickle cell disease.
Table 5 Multivariate logistic regression predicting factors associated with physician adherence to antibiotic-prescribing
guidelines (n = 138) 1
Beta
Significance
Odds ratio
Lower 95% CI
Upper 95% CI
−0.004
−2.60
0.898
0.0012
0.966
0.074
0.940
0.015
1.06
0.361
0.191
0.921
−2.70
0.667
0.403
0.723
0.414
0.0332
0.249
0.0312
1.21
2.51
0.067
1.95
1.49
0.422
0.278
0.006
0.626
1.04
3.47
22.69
0.803
6.06
2.16
.........................................................................................................................................................................................................................
Years in practice
Pediatricians
Type of practice
Group practice
Hospital
Medical school
Gender
Knowledge
CI, confidence interval.
1
Cox and Snell, r 2 = 0.335.
2
Significant at P < 0.05 in the multivariate model.
249
K. E. Wurst and B. L. Sleath
of the sickle cell disease patients under 5 years of age, which
could explain why they were most likely to adhere to the
guidelines.
Physician knowledge was also significantly associated with
physician adherence to antibiotic prophylaxis-prescribing guidelines. This finding is intuitive as it is to be expected that the
more a physician knows about a guideline, the more adherent
that physician would be to the guideline.
Those that practiced in a medical school or university setting were significantly less likely to be adherent to guidelines
than those in other practice specialties. The opposite would
be expected as these physicians work in an environment
where guidelines are often developed. Future research could
determine the reasons behind this and whether in fact these
physicians are actually less adherent to the sickle cell disease
guidelines in practice than others.
Less than half of physicians (38%) indicated that continuing
education in sickle cell disease would be useful to them in
their practice. This could have occurred for three reasons: (i) the
physicians surveyed do not see pediatric patients with sickle
cell disease and therefore do not have use for a continuing
education program; (ii) those in group practices run by
HMOs, and those in hospital or medical school practice may
receive continuing education programs provided by their
employers; and (iii) the physicians surveyed may have much
experience with sickle cell disease and do not feel as though
they need continuing education. Based on the practice characteristics of the survey respondents, the first reason is the most
plausible. The majority of hematology/oncology specialists
saw only adult patients (only 18% had a pediatric patient with
sickle cell disease in their practice) and only two pediatric
hematology/oncology specialists were captured in the sample,
therefore adult hematology/oncology specialists would see less
need to learn more about pediatric sickle cell disease. Further
research may be necessary to understand the true causes for the
lack of desire for continuing education in sickle cell disease.
The study had several limitations. One limitation is that the
pediatrician survey population only consisted of those that
practice in an area where the population is >20% non-white.
In limiting the sample population, some of the providers that
see patients with sickle cell disease may have been missed.
These providers may have different characteristics from the
ones surveyed. However, this strategy is also a strength
because it is likely that more providers that see patients with
sickle cell disease were captured due to the higher percentage
of non-whites in the county population.
Although the response rate of 57% in this study is less than
ideal, it is above the mean response rate of published physician surveys (54%) [17]. It is likely that those physicians who
responded saw more patients with sickle cell disease, although
this could not be measured and therefore introduces possible
non-response bias into the study. However, factors that could
be assessed showed no significant difference between
respondents and non-respondents.
In using a cross-sectional study design it is difficult to
assess causal relationships between factors in the study.
Actual practice and relationships between factors cannot be
assessed. Even though this study design is limited, it provides
250
us with an estimate of the physician characteristics that are
associated with adherence and knowledge of the sickle cell
disease guidelines.
In this study, the assessment of adherence to guidelines is a
surrogate measure. It has been shown that self-reported guideline adherence rates exceed those that are objectively measured [18]. Social acceptability bias may also contribute to an
increased rate of self-reported adherence. Although we did not
have an objective measure of adherence, we did receive an
estimate of physician adherence and knowledge of the sickle
cell disease guidelines. Future research can build upon this
study and use objective measures of guideline adherence from
pharmacy records, physician charts, and patient assessments.
Conclusion
The majority of physicians surveyed were relatively knowledgeable about sickle cell guidelines. Physician knowledge of
antibiotic prophylaxis-prescribing guidelines was shown to be
associated with physician adherence to the guideline. To
increase adherence to the guidelines, specific issues of prescribing antibiotics to a child with unconfirmed sickle cell
disease and penicillin dosage may need to be addressed, as
physicians surveyed were less knowledgeable about these areas.
References
1. Sickle Cell Disease Guideline Panel. Sickle Cell Disease: Screening,
Diagnosis, Management, and Counseling in Newborns and Infants.
Clinical Practice Guideline No. 6. Agency for Health Care Policy
and Research (AHCPR) publication No. 93-0562. Rockville,
MD: AHCPR, Public Health Service, US Department of Health
and Human Services, 1993.
2. Stern KS, Davis JG. Newborn Screening for Sickle Cell Disease: Issues
and Implications. New York, NY: Council of Regional Networks
for Genetic Services, Cornell University Medical College, 1984.
3. Overturf GD. Infections and immunizations of children with
sickle cell disease. Adv Pediatr Infect Dis 1999; 14: 191–218.
4. Gaston MH, Veter JL, Woods G et al. Prophylaxis with oral
penicillin in children with sickle cell anemia: a randomized trial.
N Engl J Med 1986; 314: 1593–1599.
5. Teach SJ, Lillis KA, Grossi M. Compliance with penicillin
prophylaxis in patients with sickle cell disease. Arch Ped Adoles
Med 1998; 152: 274.
6. Berkovitch M, Papadouris D, Shaw D, Onuaha N, Dias C,
Olivieri NF. Trying to improve compliance with prophylactic
penicillin therapy in children with sickle cell disease. Br J Clin
Pharmacol 1998; 45: 605–607.
7. Davis H. Use of computerized health claims data to monitor
compliance with antibiotic prophylaxis in sickle cell disease.
Pharmacoepidemiol Drug Saf 1998; 7: 107–112.
8. Pathman D, Konrad T, Freed G, Freeman V, Koch G. The
Awareness-to-Adherence model of the steps to clinical practice
guideline compliance: the case of pediatric vaccine recommendations. Med Care 1996; 34: 873–889.
Knowledge of sickle cell disease
9. Tudiver F, Herbert C, Goel V. Why don’t family physicians
follow clinical practice guidelines for cancer screening? Can Med
Assoc J 1998; 159: 797–798.
15. VanGeest J, Wynia M, Cummins D, Wilson I. Effects of different
monetary incentives on the return rate of a national mail survey
of physicians. Med Care 2000; 39: 197–201.
10. Ely JW, Goerdt CJ, Bergus GR, West CP, Dawson JD,
Doebbeling BN. The effect of physician characteristics on
compliance with adult preventative care guidelines. Fam Med
1998; 30: 34–39.
16. Wang WC, Wong W, Rogers ZR, Williams JA, Buchanan GR,
Powars DR. Antibiotic-resistant pneumococcal infection in
children with sickle cell disease in the United States. J Pediatr
Hematol Oncol 1996; 18: 140–144.
11. Cabana MD, Rand CS, Powe NR et al. Why don’t physicians
follow clinical practice guidelines?: a framework for improvement. J Am Med Assoc 1999; 282: 1458–1465.
17. Asch DA, Jedrzieski MK, Christakis NA. Response rates to mail
surveys published in medical journals. J Clin Epidemiol 1997; 50:
1129–1136.
12. Lexchin J. Improving the appropriateness of physician prescribing.
Int J Health Serv 1998; 28: 253–267.
18. Adams AS, Soumerai SB, Lomas J, Ross-Degnan D. Evidence
of self-report bias in assessing adherence to guidelines. Int J Qual
Health Care 1999; 11: 187–192.
13. Lomas J, Anderson GM, Dominick-Pierre K, Vayda E,
Enkin MW, Hannah WJ. Do practice guidelines guide practice? The effect of a consensus statement on the practice of
physicians. N Engl J Med 1989; 321: 1306–1311.
14. Sheps Center Data Sources. North Carolina Health Professions Data
Book. Sheps Center Data Sources, 1999 (http://www.shepscenter.
unc.edu/data/datatoc.html).
Accepted for publication 10 December 2003
251