Yonatan Shapiro MD, Anthony Heymann MB BS MHA, Varda Shalev MD, Nitzan Maharshak MD, Eyal Kremer, Gabriel Chodick PhD MHA, Joshua Shemer MD, Ehud Kokia MD
E-mail : Jonmidi@netvision.net.il
Disenchik st. 16/9 Tel aviv, Israel. 69353
Background: Influenza vaccination has been shown to be associated with a reduction in risk for hospitalization for pneumonia and death among the elderly. Consequently, a major public health challenge is to increase rates of influenza vaccination and to detect factors affecting the compliance with vaccination. One of these factors maybe the vaccination status of the family physician. Our objective was to assess whether patients treated by family physicians who were vaccinated against influenza have higher rates of vaccination in comparison to patients of non-vaccinated physicians.
Methods: A historical cohort was assembled using computerized records of 528 family physicians who treated 96,445 patients aged 60 years and above in a large HMO. The period studied was from September 2001 through February 2002, which were the months in which vaccination was provided. Socio-demographic and vaccination history data were all obtained from the computerized medical records.
Results: The rate of vaccination among the family physicians was 25.6%. In univariate and multivariate analysis patients compliance with vaccination was significantly (P<0.001) associated with the following characteristics of the family physician; vaccination for influenza, male gender, older age and having a medical specialty. Physicians treating a higher number of patients have a lower rate of vaccination among their patients.
Conclusions: The rate of physician uptake of influenza vaccination has a significant positive effect on rates of compliance among their patients. This is the first study to quantify this effect based on a national data-base. These findings support the implementation of special measures to encourage the vaccination of primary care physicians for influenza.
Key words: influenza, vaccine, compliance, physician-education
Influenza causes hundreds of thousands of hospitalizations and tens of thousands of deaths each year in the USA(1). The vaccine has been shown to reduce rates of hospitalization due to cardiovascular causes (2). Although the influenza vaccine was proven cost-effective(3,4) the number of individuals receiving the vaccination, including those at high risk for complications (e.g. the elderly) could still be substantially improved. In a national survey in the USA amongst patients over 65 years 34.5% were not immunized. (5) In Israel 51.3% of this population were not immunized. (6) Understanding the factors underlying poor influenza uptake is essential for developing vaccination promotion campaigns in the future.
It has long been recommended that healthcare workers (HCW's) receive the influenza vaccination because they are a high-risk group for contracting and transmitting the disease. (7) In Israel the rate of vaccination among HCW's is 66% (8) Evidence supports the positive effect of providers' recommendation on vaccination uptake. (9, 10,11)
A questionnaire-based study held in the US found that female gender of the healthcare provider, the provider having been vaccinated, and the providers' belief about the vaccination effectiveness, in descending order, were most strongly associated with their patient's compliance. (12)
The reported association from questionnaires between the provider having been vaccinated and the effect on patient uptake has not been confirmed on a national scale by examining the patient utilization data. In order to address this problem, data was drilled from the large database of "Maccabi Healthcare Services".
Settings: This whole population
study was held in Maccabi Healthcare Services (MHS), the second largest HMO
in Israel, with 1.6 million members nationwide.
Study population: All primary care physicians (Internists, family medicine specialists and physicians with no specialization) who were members of MHS were selected. In order to reduce statistical variance in compliance rates, physicians with less than twenty patients over fifty-nine years of age during the study period were excluded from study.
Data collection: Demographic data concerning patients and physician's age, gender, statuses of influenza vaccination, as well as physician's specialty were all derived directly from MHS' computerized database which includes all patient information.
Statistical Analysis: comparison of mean ages between male and female physicians was performed using the standard student-t test. Statistical analysis of vaccination rates between different categories was held using the chi-square test. Confidence intervals were calculated according to the Wilson procedure with a correction for continuity (13), using WHATIS 4.12 statistical calculator (PEPI, Copyright JH Abramson 2000-2003) Multivariate linear regression analysis was applied to determine the standardized coefficient related to patient's compliance with vaccination, using SPSS 10.0 (Chicago, Ill).
There are 691 primary care physicians who were themselves members of MHS. These physicians treated 97768 patients aged 60 or more. Our cohort includes 528 of these physicians who treated twenty or more patients. These 528 physicians treated 96657 patients. The physician distribution according to age and gender are described in Table 1.
Rate of vaccination among
Among the physicians, 25.6% were vaccinated for influenza during the study period. Specialists showed a significant (p<0.01) higher rate of vaccination (31.5%) in comparison to non-specialist physicians (19.5%). The lowest rate (5.4%) has been documented among 37 physicians (7%) who treated more than 450 elderly patients (Table 2). This small group of physicians treated 22697 (23.5%) patients.
Rate of vaccination among patients:
A positive significant (P=0.02) linear trend was calculated between physician's age and rate of patient vaccination. An opposite significant (P<0.01) linear trend was observed between the number of physician's patients aged 60 years and above and their rate of vaccination. A significant (P<0.01) higher rate of vaccination was found for patient's treated by male physicians and specialist physicians in comparison to female and non-specialist ones. The crude vaccination rate of patients among vaccinated physicians was significantly (P<0.01) higher (46.6%) in comparison to non-vaccinated physicians (40.1%) (Table 3).
The multivariate analysis shows the following positive associations with patient's vaccination status: Compliance of physician with vaccination (5.1%), Physician's age (3.6% increase in patients' rate of vaccination for every 10 years of physicians age), medical specialty (5.6%) and male gender (2.8%). No difference was found between specialists in family medicine and internal medicine.
The number of elderly patients treated by the physician is negatively associated with the rate of patients vaccination with a mean reduction of 1.4% for every 100 patients.
Increasing rate of vaccinated subjects is one of the main goals of healthcare organizations worldwide. The vaccination reduces mortality and morbidity rates as well as the national health costs due to hospitalization and medical care of influenza complications. Only one vaccination is needed to prevent one day of hospitalization in high risk patients 
The current study shows that physician's uptake of influenza vaccination is associated with higher rates of vaccination among their patients. Older physicians have higher rates of vaccination uptake as well as higher rates of vaccinated patients.
In our study, rate of vaccination among male physicians and their patients was higher then female physicians. This is contrast to the findings presented by Nichol .
In order to increase rates of vaccinated patients, effort should be put not only in patient education but also in physician's education. This is especially true for young physicians, with no medical specialty.
Another sub group of physicians with low rates of patient vaccination are physicians treating a high number of elderly patients. Special efforts should be put into educating these physicians concerning the effectiveness of the influenza vaccination. This small group of physicians (5.4%) of all 691 physicians treats nearly one quarter of all patients (23.2%). A limited targeted intervention among these physicians may lead to a change in vaccination practice among a relatively large group of elderly patients. From a management perspective, this is a classical Pareto Effect.
Study Limitation: Physicians may have received the vaccination elsewhere which would not be documented in the MHS database.
Our study based on administrative data shows that physician vaccination status is positively correlated with patient vaccination status. From the study data we suggest a possible strategy from increasing the rates of vaccination amongst physicians and their patients.
Table 1 Age and gender distribution of the physician in study
|</=40||30 (9.6%)||53 (24.5%)||83 (15.7%)|
|41-50||139 (44.6%)||105 (48.6%)||244 (46.2%)|
|51-60||108 (34.6%)||46 (21.3%)||154 (29.2%)|
|>/=61||35 (11.2%)||12 (5.6%)||47 (8.9%)|
|Total||312 (100.0%)||216 (100.0%)||528 (100.0%)|
Table 2 Rate of vaccination among physicians and their characteristics
|Number of phsyisician's patients aged 65 and over**|
|Non-specialist physician (n=261)||19.5%||14.9%||24.9%||P<0.01|
|Specialist Physician (n=267)||31.5%||25.9%||37.4%|
Table 3 Vaccination rates among patients of study physicians
|Compliance with vaccination|
|Physician's gender **|
|Number of physician's patients aged =65 ***|
|Non-specialist physician (n=261)||38.5%|
|Specialist physician (n=267)||44.9%|
|Physician's vaccination status**|
|Not vaccinated (n=393)||40.1%|
* F-test for linearity =0.02
** Chi-square P<0.01
*** F-test for linearity =0.01
Table 4: Multivariate linear regression of patients aged 65 compliance with influenza vaccine according to their physician's characteristics
|Variable||Coefficient||95% Confidence Interval for B||P|
(female vs. male)
|Physician's age (per 10 years)||0.036||0.012||0.060||0.004|
|Number of patients (per 100)||-0.001||-0.0017||-0.0003||0.004|
* Adjusted R square =0.91
1. Centers for Disease control and Prevention. Prevention and control of Influenza: recommendations of the Advisory Committee on Immunization Practices (ACIP). MMWR Morb Wkly Rep.2001;50(RR-04):1-44
2. Nichol KL, Nordin J, Mullooly J, Lask R, Fillbrandt K, Iwane M. Influenza vaccination and reduction in hospitalizations for cardiac disease and stroke among the elderly.N Engl J Med. 2003 Apr 3;348(14):1322-32
3. Nichol KL.The efficacy, effectiveness and cost-effectiveness of inactivated influenza virus vaccines. Vaccine. 2003 May 1;21(16):1769-75.
4. Nichol KL, Wuorenma J, von Sternberg T. Benefits of influenza vaccination for low-, intermediate-, and high-risk senior citizens. Arch Intern Med. 1998 Sep 14;158(16):1769-76.
5. Ambrosch F, Fedson DS. Influenza vaccination in 29 countries. An update to 1997. Pharmacoeconomics. 1999;16 Suppl 1:47-54.
6. Rabinovitch G, Porat A, Ruskin-Segal A. Quality indicators for comparing primary care organizations in Israel. 2002 From the final report to The Israel National Institute for Health Policy and Health Services Research and The Ben-Gurion University of The Negev.
7. Centers for Disease control. Inflenza. MMWR 1991:40:33-6
8. Habib S, Rishpon S, Rubin L. Influenza vaccination among healthcare workers. Isr Med Assoc J. 2000 Dec;2(12):899-901.
9. Petersen RL, Saag K, Wallace RB, Doebbeling BN. Influenza and pneumococcal vaccine receipt in older persons with chronic disease: a population-based study. Med Care. 1999 May;37(5):502-9.
10. Nichol KL, Mac Donald R, Hauge M. Factors associated with influenza and pneumococcal vaccination behavior among high-risk adults. J Gen Intern Med. 1996 Nov;11(11):673-7.
11. Centers for Disease control and Prevention. Adult immunization: knowledge,attitudes and practice-DeKalb and Fulton Counties, Georgia,1988. MMWR Moerb Mortal Wkly Rep.1988;37:657-661
12. Nichol KL, Zimmerman R. Generalist and subspecialist physicians' knowledge, attitudes, and practices regarding influenza and pneumococcal vaccinations for elderly and other high-risk patients: a nationwide survey. Arch Intern Med. 2001 Dec 10-24;161(22):2702-8.
13. Newcombe, Robert G. Two-Sided Confidence Intervals for the Single Proportion: Comparison of Seven Methods. Statistics in Medicine, 1998; 17, 857-872.
14. Shapiro Y,Shemer J, Heymann A,Shalev V, Maharshak N,Chodik G ,Green M, Kokia E. Influenza vaccination:reduction in hospitalizations and death rates among members of ' Maccabi Health care Services" during the 2000-2001 influenza season. IMAJ 2003 5;706-708.