Patient Acceptance of Influenza Vaccination* NICHOLAS H. FIEBACH, M.D., CATHERINE M. VISCOLI, M.Phil., NewHaven. Connecticut
PURPOSE: To determine whether patients’ demographic, medical, and personal characteristics, including attitudes and beliefs about vaccination, health, and medical providers, are associated with acceptance of influenza vaccine. PATIENTS AND METHODS: Nine hundred sixtyfive patients attending a university hospitalbased general medicine clinic during the fall influenza vaccination period, including 624 patients for whom influenza vaccine was indicated, were observed in a prospective cohort study. In addition, 58 patients who refused influenza vaccine and an equal number who accepted it were interviewed over the telephone to examine their beliefs and behaviors in greater detail. IWSULTS: Seventy-five percent of patients for whom influenza vaccine was indicated received it. prospectively 8888888d patient characteristics that were significantly associated with nonvaccination included not believing vaccine prevents “flu” (relative risk [RR] 5.3), never received pneumococcal vaccine (RR 36), not vaccinated against influenza the previous year (RR 3.5), never vaccinated against influenza (RR 2.3), and felt sick after previous influenza vaccination (RR 2.3). Demographic characteristics and medical diagnoses were not significantly rehxted to vaccination. Ahnost one half of 58 interviewed subjects who refused influenza vaccine cited fear of a reaction. Among retrospectively determined attitudes and beliefs significantly associated with refusal of influenza vaccine were not believing the vaccine works well (odds ratio [OR] 11.6), concern about a reaction (OR 9.3), and perception that the medical provider had not recommended it (OR 5.8). CONCLUSION:
&mographiC
CharaCbri~tia
Of
patients and their medical diagnoses were not associated with acceptance of influenza vaccination. Among patients who were not vaccinated, doubts about the efficacy of influenza vaccine and fear of its side effects were common, and their perceptions of the medical provider’s recommendation of vaccine appeared to be an im*This is a rapid publication manuscript. From the Section of General Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut. Requests for reprints should be addressed to Nicholas H. Fiebach. M.D., Yale-New Haven Hospital, Primary Care Center, 20 York Street, New Haven, Connecticut 06504. Manuscript submitted February 26, 1991, and accepted in revised form August 19. 1991.
portant it.
factor in the decision whether
to accept
I
nfluenza vaccination reduces morbidity and mortality from respiratory infections in targeted populations [l-3] and is therefore recommended for patients with chronic medical conditions and those 65 years or older [4,5]. Nevertheless, during the last decade in the United States, only 32% of the elderly and only 10% of younger patients with chronic diseases received influenza vaccination each year [ 1,6]. Vaccination rates in some university hospital outpatient clinics were reported to be similarly low [7], and in the Primary Care Center at Yale-New Haven Hospital, only one third of patients for whom the vaccine was indicated had received it in previous years (J. Ginetti, unpublished data). The low rates of vaccination have been blamed on administrative obstacles, since physician awareness about the effectiveness of influenza vaccine and its indications has been documented [1,8]. Organized programs aimed at high-risk patients resulted in influenza vaccination rates of 50% to 70% [l], but still fell short of the goal set by the Immunization Practices Advisory Committee of the Centers for Disease Control of vaccination in 80% of patients for whom it is indicated [5]. The possibility that specific patient factors also contribute to nonvaccination has been investigated, and previous studies have shown that demographic characteristics, health beliefs regarding “flu” and the influenza vaccine, previous experience with the vaccine, and relationship to health care providers and the health care system influence vaccination [g-16]. Unfortunately, previous studies of acceptance of influenza vaccination were difficult to apply in our university hospital-based clinic. Some of the studies were based on community surveys [9,10,15] or were conducted during the intense publicity surrounding the swine influenza vaccination campaign in 1976 [9,10,12], and are not applicable to patients attending our clinic. In studies done in clinical settings, patient populations were not described fully enough to allow accurate comparison with our patients [ll-141. Some studies of factors influencing influenza vaccination included outreach efforts by mail or telephone, but patients’ attitudes regarding vaccination were not always assessed prospectively prior to their decision to receive or decline vaccination [11-141. Also, other patient factors, including
October
1991
The American
Journal
of Medicine
Volume
91
393
INFLUENZA VACCINATION / FIEBACH AND VISCOLI
previous compliance with physicians’ recommendations, other preventive health behaviors, and sense of well-being and ability to function, were not considered in previous analyses. We undertook to improve influenza vaccination rates in our clinic by using medical provider education and written reminders. In addition, in order to understand if patient characteristics were important determinants of vaccination in our clinic population, we sought to identify those patients who remained unvaccinated despite this intervention. Specifically, we assessed prospectively the demographic, medical, and personal characteristics of patients, including previous experience with influenza vaccine and beliefs about its efficacy, which may be associated with nonvaccination. We also designed a case-control study nested within the larger cohort study to document patients’ reasons for refusing influenza vaccine and to explore the relation between refusal and patients’ reports of well-being, function, previous compliance with doctors’ recommendations, and beliefs about influenza vaccine.
PATIENTS AND METHODS The study was conducted in the Primary Care Center of Yale-New Haven Hospital. Adult patients in this outpatient facility are cared for by faculty physicians, internal medicine residents, nurse practitioners, and a few physician volunteers. For 2 weeks prior to the study period, one of the authors reviewed the indications, use, and efficacy of influenza vaccine at the daily clinic conference and distributed a vaccine information sheet to the providers. Vaccination was recommended for all patients 65 years of age or older; patients with chronic diseases including congestive heart failure, ischemic cardiac disease, obstructive pulmonary disease, asthma, diabetes mellitus, renal insufficiency, severe anemia, or immunosuppression (including drug-induced, malignancy, and human immunodeficiency virus); household contacts of patients with these high-risk conditions; residents of chronic-care facilities; and health care workers. Providers were also asked about their own histories of influenza vaccination. The first part of the study was a prospective cohort analysis that included all adult patients seen in the Primary Care Center from October 17 through December 15, 1987. Data collection forms were stamped with patients’ hospital identification cards, which include demographic and insurance information, and attached to patients’ charts. As patients arrived in the clinic, nurses recorded information on their previous receipt of and reaction to influenza vaccine and whether they believed the influenza vaccine prevents the “flu.” Providers 394
October 1991 The American Journal of Medicine
completed a checklist of chronic disease indications for influenza vaccine and recorded information on the frequency of clinic visits during the previous year, prior receipt of pneumococcal vaccine, recent receipt of influenza vaccine elsewhere, contraindications to influenza vaccine, and patients’ reasons for refusing influenza vaccine. Finally, clinic nurses who administered vaccine recorded which patients received it. For each patient visit, information was also available on the type of provider and level of training (i.e., faculty, housestaff [postgraduate year 1 to postgraduate year 31, nurse practitioner), whether the provider had attended a vaccine review conference or received a vaccine information sheet, and whether the provider had personally received the influenza vaccine or planned to obtain it. In order to determine whether our patient sample was representative of patients who receive care at the Primary Care Center at other times during the year, we reviewed a convenience sample of patient charts. Every eighth patient was selected from an alphabetical list of Primary Care Center patients who had at least one clinic visit after October 1, 1986, approximately 1 year prior to the start of the study period, and the proportions of patients for whom influenza vaccine was indicated and who attended the clinic during the study period were calculated. In the second part of the study, patients for whom influenza vaccine was indicated but who had refused it were identified. Each refuser was matched to another patient seen by the same provider on the same day who accepted the vaccine. When more than one vaccine recipient was available for matching, one was chosen randomly; if no patients received the vaccine on the same day, then we selected one who came to the clinic on the nearest date. When the study was conducted, residents came to the clinic every 2 weeks or less frequently, so some vaccine refusers could not be matched to a vaccine recipient of the same resident. These vaccine refusers were matched to a vaccine recipient of another resident at the same level of training who came to the clinic on the same or nearest date. Trained research assistants (including one fluent in Spanish) made several attempts to contact matched patients by telephone. These patients were asked to respond to a 30-item questionnaire consisting of open-ended questions about reasons for accepting or refusing influenza vaccination, fear of adverse reactions, and noncompliance with physicians’ recommendations, and 25 scaled items assessing attitudes and beliefs about health care and vaccination (see Appendix). The data were analyzed using SAS statistical software [17]. For prospectively assessed patient
Volume 91
INFLUENZA
characteristics, relative risks and 95% confidence intervals for nonvaccination were determined for patients in whom influenza vaccine was indicated, excluding those who did not receive the vaccine because of egg allergy, acute illness, or temporary shortage of vaccine. The power of these analyses to detect significant associations was calculated [18]. For the case-control analysis of vaccine refusers and recipients, personal and medical characteristics of patients were tabulated and differences between groups were compared with the chi-square test. Median times between clinic visits and telephone interviews were calculated for each group and compared with the rank sum test. Responses to the scaled telephone questionnaire items were dichotomized, and both matched and unmatched analyses were performed to obtain odds ratios and 95% confidence intervals for nonvaccination [19]. RESULTS Prospective Cohort Analyses Data forms were completed for 1,175 visits by 965 patients, representing virtually all of the patients seen in the clinic during the study period. The majority of patients were women and single, and many were nonwhite and elderly (Table I). Most patients had either Medicaid or Medicare coverage and had at least two visits to the clinic during the previous year. About one half were the patients of nine faculty physicians, about one third were the patients of 59 residents, and the remainder were patients of three nurse practitioners and five volunteer physicians. Seventy-seven percent of the patients were seen by providers who had attended a conference or received an information sheet on influenza vaccine before the study period. Vaccination was indicated for 48% of patients because of chronic disease. Diabetes mellitus was the most prevalent chronic disease indication, and ischemit heart disease, congestive heart failure, chronic obstructive pulmonary disease, and asthma were also common. Fourteen percent of patients had more than one chronic disease indication; among diabetic patients, 36% had another chronic disease for which vaccine was indicated. Thirty-six percent of patients were aged 65 years or older, although vaccine was indicated only because of older age in 15%. Altogether, vaccination was indicated for 65% of all patients. Half of the patients for whom influenza vaccine was indicated had received it previously, including 33% who had received it the year before. Fewer patients had received pneumococcal vaccine. One in 10 patients who had previously received influenza vaccine reported feeling sick afterwards. Only 30% believed that influenza vaccine prevents the “flu.” Of the 624 patients for whom vaccine was indicat-
VACCINATION
/ FIEBACH
AND VISCOLI
TABLE I ProspectiveCohort Analysis: Patient Characteristics Patients Influenza Vaccine All Indicated (n = 965)* (n = 624)t Demographic factors Age t 65 Female Nonwhite Married
348 (36%) 348 (56%) 654 (68%) 431 (69%) 479 i56%j 305 i56%j 218 (26%)
150 (28%)
Health insurance+ None Commercial Medicaid Medicare ClinF;isits
during past year
211 (22%) 86 (17%) 345 (36%) 239 (48%) 211 (22%) 176 (35%)
2-4 >5 Medical provider Faculty Housestaff Nurse practitioner Other physicians Exposed to influenza vaccine review Received or planned influenza vaccine
450 (47%) 295 (47%) 357 (37%) 229 (37%)
292 (30%) 185 (30%)
Indications for influenza vaccine Age 2 65 only Medical diagnoses Congestive heart failure lschemic heart disease Chronic obstructive pulmonary disease Asthma Diabetes mellitus Chronic renal failure Significant anemia lmmunosuppressed Nursing home resident Health care provider Any indication Vaccination history and beliefs Ever received influenza vaccine Received influenza vaccine previous year Felt sick after previous influenza vaccinationr’ Ever received pneumococcal vaccine Believed influenza vaccine prevents “flu”
144 (15%)
144 (23%)
396 (41%)
314 (50%)
234 (24%) 205 (33%) 41 (10%) 29 (5%) 115 (12%)
99 (16%)
246 (26%) 184 (30%)
)ata missing for: gender, 5 (1%); race, 117 (12%); marital status, 129 (13%); clinic visits, 19 1%). lata missing for: race.! 74 (12%); marital status, 80 (13%); clinic visits, 123 (20%). Btegorles not mutually excIusIve. ,mongpatients who had previously received influenzavaccine (396 of all patients, 314 for whom
ed, 440 were vaccinated in the Primary Care Center during the study period and 26 reported receiving it elsewhere, resulting in a vaccination rate of 75%. Vaccination rates were similar among patients whose only indication was age 65 years or older (76%), those of older age who also had a chronic disease indication (76%), and younger patients with a chronic disease (73%). Patients with two or more chronic disease indications had a significantly higher vaccination rate than did patients with only one chronic disease (82% versus 71%, x2 = 6.84, p 5
AND VISCOLI
Number*
Patients Not Interviewed* (rl = 80)
65% 72% 58% 30% Health insurance None Commercial Medicaid Medicare
/ FIEBACH
TABLE IV Case-ControlAnalysis: Patients’ Principal Reasonsfor Accepting or Refusing Influenza Vaccination
TABLE Ill Case-ControlAnalysis: Patient Characteristics Interviewed Refused Vaccine (n = 58)
VACCINATION
26 (46%) ;; I;;;; 10 (18;)
4 (7%) 4 (7%) 18 (32%)
,
*Categories not mutually exclusrve. Among recipients, 16 gave two reasons and two gave three reasons; amongrefusers, 15 gave two reasons and five gavethree reasons.
TABLE V Case-ControlAnalysis: Questionnaire ResponsesSignificantly Associatec with Refusing Influenza Vaccination
Question/Response
Odds Ratio (95% Confidence Interval)
Do you think the flu vaccine works well to prevent people from getting sick with the flu? Not well at all or not sure versus very well or fairly well
(6.3,41.0)
Were you very concerned about a bad reaction from the flu vaccine? Very concerned or a little concerned versus not at all concerned
9.3 (3.9,22.3)
Did your doctor recommend that you get the flu vaccine? Did not recommend it versus strongly urged it or recommended it
5.8 (2.5, 13.2)
Do you think it’s possible that the flu vaccine could cause a person to have a severe reaction or a serious illness? Possible or not sure versus rare or never
(1.2,14.8)
In general, are you able to do the things in your home that you would like to do? Always or sometimes versus not usually or never
3.9 (1.5, 10.3)
74%
8
were not Intt?rVleWea: Jb vaccine WplentS and Zb vaccine retUSerS COUM not be ached on the telephone. Nine subjects who had refused vaccine were unable to communrcate equately over the telephoneor were unwrllingto respond to the questronnarre. Two patrents who received vaccme dred before they could be IntervIewed, five patients who refused the vaccine had received it by the trme they were mtervrewed, and two pabents who were coded as havrng recerved vaccine denied It. ‘Comparisons between all mtervrewed pabents and those not IntervIewed: p < 0 05. :eaSOnS patIm
ed that their medical provider had not recommended the vaccine, and more often described themselves as able to function in their homes. Scores for other questions, including those that assessed perceived risk of influenza, importance of vaccine cost, previous medical compliance, and health status, were similar between the two groups. The most common reasons for accepting or refusing vaccine, and the responses to the questionnaire items described above, did not differ between patients aged 65 years or older and younger patients.
COMMENTS Our study included more patients for whom influenza vaccine was indicated than can be found in previous studies of acceptance of influenza vaccination [g-16]. We did not find that demographic factors such as age, race, gender, and marital status were associated with influenza vaccination despite excellent statistical power. Previously reported associations between influenza vaccination and age [ll-131 or race [12,16] may have been due to chance
16.1
4.2
variations among small subgroups, or perhaps reflected a relationship between access to medical care and vaccination rather than an effect on acceptance of the influenza vaccine once it was available. Although a previous study found that patients with pulmonary diseases received influenza vaccine less often than other patients for whom it was indicated [12], we found no association between vaccination and chronic disease diagnoses. Contraindications to the vaccine such as acute febrile illness or egg allergy were rare, and the major reason for nonvaccination was patient refusal. Responses to questions in both the prospective and
October 1991
The American
Journal
of Medicine
Volume
91
397
INFLUENZA VACCINATION / FIEBACH AND VISCOLI
case-control analyses in our study, and results of previous studies, showed that the factor most strongly associated with nonvaccination was not believing that the vaccine prevents the “flu” [10,11,14,15]. However, the majority of patients who accepted vaccination also thought it would not prevent the “flu” when asked prospectively. Since vaccine recipients who were interviewed retrospectively often listed prevention of illness as their major reason for accepting the vaccine, this may reflect persuasion by their providers or rationalization for their action. In contrast, only 7% of interviewed vaccine refusers cited lack of effectiveness as their main reason for refusing the vaccine. These results suggest that while many patients have doubts about the efficacy of the influenza vaccine, this may not be the decisive factor in refusal of vaccination. Although a previous reaction to influenza vaccine was significantly associated with nonvaccination in our study and others [13-151, only 10% of all previously vaccinated patients reported a prior reaction in the prospective analysis. Other studies [20,21] and public health directives [4] also emphasize that serious reactions to influenza vaccine are infrequent. Nevertheless, in retrospective interviews, fear of a reaction was cited as the main reason for refusal by almost half of the vaccine refusers, and responses to specific questionnaire items confirmed that vaccine refusers were significantly more likely to worry about adverse reactions than vaccine recipients. The disparity between the actual and perceived risk from influenza vaccine seems to be a major reason why vaccination is refused by some patients [12]. Our study was not specifically designed to examine the effect of medical providers’ attitudes and behavior on patients’ acceptance of influenza vaccine, but their impact is suggested by the variation in vaccination rates among providers. Differences in type of provider, level of training, and exposure to recent information on influenza vaccine did not account for this variation. Although it may reflect random effects among small patient subgroups or real differences among the patients cared for by individual providers, the variation in vaccination rates was probably also the result of differences among providers in compulsiveness, enthusiasm, and persuasiveness in recommending vaccine. Although 40% of interviewed vaccine recipients cited their provider’s recommendation as the major reason for accepting the vaccine, vaccine refusers were six times more likely to report that their provider had not recommended it. We were not able to determine whether providers really did not recommend vaccine to these refusers, whether this represents rationalization by patients for not accepting vaccination, or whether patients and providers differed 398
October 1991 The American Journal of Medicine
in their perception of what was said in their interaction. Nevertheless, it appears that providers’ attitudes and recommendations are an important factor in patients’ vaccination decisions [10,15,16]. We hypothesized that patients’ previous compliance with treatment, sense of well-being and ability to function, and other preventive health behavior would influence acceptance of influenza vaccine, but this was not confirmed. Although vaccine refusers tended to score higher on questions about sense of well-being and function, and lower on questions about other preventive health behaviors, the difference was statistically significant only for the question that assessed function at home. Unlike previous studies that specifically sought to validate a “health belief model” [22] for influenza vaccination [g-11,13], we did not find a significant difference between vaccine recipients and refusers in their perceptions of susceptibility to influenza or its potential severity, or the importance of the cost of the vaccine. Two studies that found that vaccination was significantly associated with perceived susceptibility to influenza and its severity were conducted during the swine influenza campaign [9,10], and it is possible that widespread publicity about the seriousness of the anticipated swine influenza epidemic temporarily raised awareness of influenza and concern about its consequences. However, Larson and colleagues [ll] used a retrospective questionnaire to assess the vaccination decisions of patients attending a university medical clinic the year before the swine influenza campaign, and found that concerns about susceptibility to influenza, its severity, and cost of the vaccine were significantly associated with vaccination. Although the patients in that study resembled our sample in being predominantly older and female, other demographic comparisons are not possible. It is conceivable that differences in socioeconomic factors or secular trends in attitudes toward influenza and vaccination contributed to the conflicting results. The contrast between the results of the Larson study and ours also illustrates a crucial distinction in studies of vaccination outcomes. The “health belief model” emphasizes preventive health careseeking behavior by patients [ll], and studies of influenza vaccination that have tested its hypotheses have focused on whether patients came to the clinic to receive vaccine and the factors that influenced this action. Our study, on the other hand, analyzed the commonly encountered situation of a high-risk patient coming to a clinic but not being vaccinated, a so-called “missed opportunity” for vaccination [5]. Although our results apply only to patients who came to the clinic during the fall season, when vaccination is recommended, a convenience sample of clinic charts showed that approxi-
Volume 91
INFLUENZA VACCINATION / FIEBACH AND VISCOLI
mately three quarters of our patients for whom influenza vaccine was indicated had attended the clinic during the study period. We recognize that the “missing” patients may have had different attitudes toward influenza vaccine, that other patient populations with fewer indications for vaccine or less frequent visits to a clinic may have different determinants of vaccination, and that getting eligible patients to the clinic during the fall vaccination period will be necessary to achieve vaccination goals. Several other limitations of our study deserve mention. We had prospective information on some demographic characteristics, but lacked data on education, employment, and financial status other than insurance coverage. We were able to measure patients’ beliefs about the efficacy of influenza vaccine and prior experiences with it before they decided about vaccination, but we relied on a retrospective case-control analysis to assess patients’ decisions in more detail. Although the telephone questionnaire we used was straightforward and possessed face validity, it was not formally tested for reliability and accuracy. A substantial number of patients who could not be reached on the telephone were excluded from the case-control analysis. Although several characteristics differed significantly between interviewed and noninterviewed patients, only one of these, number of clinic visits, was predictive of vaccination in the prospective analysis. Perhaps more importantly, the reasons for exclusion differed between vaccine recipients and refusers who could not be interviewed. Among noninterviewed subjects, more vaccine recipients did not have a telephone, did not answer it, or were unable or unwilling to be interviewed after they were contacted. We do not know whether these circumstances were associated with factors that influence vaccination attitudes and decisions. Despite these limitations, our results show that some patients remain unconvinced of the benefit of influenza vaccine and wary of its side effects. Adverse publicity about the swine influenza vaccination program in 1976 may still contribute to this, although influenza vaccines in subsequent years have been studied extensively and no association with neurologic sequelae has been found [ZO]. Also, confusion between lay and medical uses of the term “flu” has obscured the important point that the main benefit of influenza vaccine is not to forestall “flu” symptoms, but rather to prevent serious complications and deaths during influenza epidemics [l-5]. Since patients’ perceptions of recommendations for influenza vaccine appear to be important in vaccination decisions, health care providers should not assume that there is a substantial group
of recalcitrant patients who will not accept vaccination if it is offered. Clearer and more widespread messages about the safety and efficacy of influenza vaccine in conjunction with unambiguous recommendations to patients for whom it is indicated may help achieve desired levels of vaccination.
APPENDIX Telephone Interview Questionnaire 1. What werevour reasons for acceotineinot acceotine the flu vaccine? 2. What was the most important reaso? ’ 3. When you think of “the flu,” what do you think of? - a cold _ a bad cold - worse than a bad cold - other 4. How would you rate your care at the Primary Care Center? - verveood - eood - fair Door 5. Do you u&ally follow;our doctors recommendations? 6. What doctors recommendations have you not followed? 7. Does your doctor prescribe medication for you? - to take every day - only when you are sick none 8. Do you take your prescribed medication? - almost always - sometimes - never 9. How often do you miss appointments with the doctor? - frequently __ sometimes _ rarely 10. Is it likely that you would get the flu this winter? probably would not sure - probably would not 11. If YOUwere to net the flu, is it likelv that YOUwould be verv ill? not very ifi - somewhat iii i very seriously iI1 not sure 12. Do you think the flu vaccine works well to prevent people from getting sick with the flu? - very well - farrly well - not well at all not sure 13. Was there an extra charge to you to get the flu vaccine? - yes _ not sure - no 14. Was the cost an important factor in your decision whether to get the flu vaccine? - very important - one of several considerations - not important 15. Were you very concerned about a bad reaction from the flu vaccine? - very concerned - a little concerned not concerned at all 16. What bad reactions were you worried about? 17. Do you think it’s possible that the flu vaccine could cause a person to have a severe reaction or a serious illness? possible _ rare possibility _ no, never not sure 18. f&e many of your friends and relatives received the flu vaccine? many - some - few or none 19. %ve any of your friends and relatives had a bad reaction from the flu vaccine? - yes _ not sure - no 20. Have you heard on TV or radio or read in a newspaper or magazine that flu vaccine is recommended for a person like you? - yes _ not sure no 21. Did your doctor recommend that you get the flu vaccine? strongly urged you to get it recommended it - did not recommend it 22. Are you nervous when you see your doctor? - very nervous - a little nervous not nervous at all 23. Are you scared of needles? - very scared - a little scared not scared 24. Do vou smoke ciearettes? -‘current smoker - former smoker non-smoker 25. Do you wear a seat belt in the car? --always sometimes never 26. Do vou eat a healthv diet? -‘always _ sometimes _ rarely 27. Do you have any serious illnesses? - yes _ no - notsure 28. In general, would you say that your health is good? - very good L good i fair poor 29. In general, are you able to do the things in your home that you would like to do? - always - sometimes not able to - never or usually not able to 30. In general, do you get out of the house a lot? - frequently - sometimes - rarely, never
October 1991 The American Journal of Medicine
Volume 91
399
INFLUENZA VACCINATION / FIEBACH AND VSCOLI
ACKNOWLEDGMENT We thank Ruth Gemmell for assistance in preparing the manuscript, Pat Bennett for help in obtaining patient charts, Ralph Horwitz, William Beckett. and Maryann Zitt for advice, and the nurses, clerical staff, and housestaff in the Primary Care Center for their cooperation in collecting the data.
REFERENCES 1. Fedson DS. Influenza prevention and cure. Past practices and future prospects. Am J Med 1987; 82: 42-7. 2. Barker WH. Mullooly JP. Influenza vaccination of elderly persons. Reduction of pneumonia and influenza hospitalizations and deaths. JAMA 1980; 244: 2547-9. 3. Gross PA, Quinnan GV, Rodstein M. et a/. Association of influenza immunization with reduction in mortality in an elderly population. A prospective study. Arch Intern Med 1988; 148: 562-5. 4. Prevention and control of influenza: Part 1. Vaccines. MMWR 1989; 38: 297-311. 5. Williams WW. Hickson MA, Kane MA, eta/. Immunization policies and vaccine coverage among adults. The risk for missed opportunities. Ann Intern Med 1988; 108: 61625. 6. Influenza vaccination levels in selected states-behavioral risk factor surveillance system, 1987. MMWR 1989; 38: 124-33. 7. Kosecoff J, Fink A, Brook RH. General medical care and the education of internists in university hospitals. An evaluation of the teaching hospital general medicine group practice program. Ann Intern Med 1985; 102: 250-7. 8. Cate TR, Ruben FL. Prevention, management, and control of influenza in internal medicine. Am J Med 1987; 82: 52-4. 9. Rundall TG, Wheeler JRC. Factors associated with utilization of the swine flu vaccination program among senior citizens in Tompkins County. Med Care
400
October 1991 The American Journal of Medicine
1979; 17: 191-200. 10. Cummings KM, Jette AM, Brock BM. Haefner DP. Psychosocial determinants of immunization behavior in a swine influenza campaign. Med Care 1979; 17: 639-49. 11. Larson EB. Olsen E. Cole W. Shortell S. The relationship of health benefits and a postcard reminder to influenza vaccination. J Fam Pratt 1979; 8: 1207-11. 12. Anderson C. Martin H. Effectiveness of patient recall system on immunization rates for influenza. J Fam Pratt 1979; 9: 727-30. 13. Larson EB. Bergman J, Heidrich F. Alvin BL, Schneeweiss R. Do postcard reminders improve influenza vaccination compliance? A prospective trial of different postcard “cues.” Med Care 1982; 20: 639-48. 14. Frank JW, Henderson M. McMurray L. Influenza vaccination in the elderly: 1. Determinants of acceptance. Can Med Assoc J 1985; 132: 371-5. 15. Carter WB. Beach LR. lnui TS. Kirscht JP, Prodzinski JC. Developing and testing a decision model for predicting influenza vaccination compliance. Health Serv Res 1986; 20: 897-932. 16. Adult immunization: knowledge,
attitudes,
and practices-DeKalb
and Ful-
ton counties, Georgia, 1988. MMWR 1988; 37: 657-62. 17. SAS Institute Inc. SAS/STAT user’s guide, release 6.03 ed. Cary, North Carolina: SAS Institute Inc.. 1988. 18. O’Connor GT. Beck JR. Epi-talc. Hanover: True Basic, Inc., 1987. 19. Rothman KJ. Modern epidemiology. 1st ed. Boston: Little Brown, 1986: 28&l. 20. LaMontagne JR, Quinnan GV, Curlin GT, et a/. Summary of clinical trials of inactivated influenza vaccine-1978. Rev Infect Dis 1983; 5: 723-36. 21. Margolis KL. Nichol KL. Poland GA, Pluhar RE. Frequency of adverse reactions to influenza vaccine in the elderly: a randomized, placebo-controlled trial. JAMA 1990; 264: 1139-41. 22. Becker MH. Maiman LA. Sociobehavioral determinants of compliance with health and medical care recommendations. Med Care 1975; 13: 10-24.
Volume 91
PURPOSE: To determine whether patients’ demographic, medical, and personal characteristics, including attitudes and beliefs about vaccination, health, and medical providers, are associated with acceptance of influenza vaccine. PATIENTS AND METHODS: Nine hundred sixtyfive patients attending a university hospitalbased general medicine clinic during the fall influenza vaccination period, including 624 patients for whom influenza vaccine was indicated, were observed in a prospective cohort study. In addition, 58 patients who refused influenza vaccine and an equal number who accepted it were interviewed over the telephone to examine their beliefs and behaviors in greater detail. IWSULTS: Seventy-five percent of patients for whom influenza vaccine was indicated received it. prospectively 8888888d patient characteristics that were significantly associated with nonvaccination included not believing vaccine prevents “flu” (relative risk [RR] 5.3), never received pneumococcal vaccine (RR 36), not vaccinated against influenza the previous year (RR 3.5), never vaccinated against influenza (RR 2.3), and felt sick after previous influenza vaccination (RR 2.3). Demographic characteristics and medical diagnoses were not significantly rehxted to vaccination. Ahnost one half of 58 interviewed subjects who refused influenza vaccine cited fear of a reaction. Among retrospectively determined attitudes and beliefs significantly associated with refusal of influenza vaccine were not believing the vaccine works well (odds ratio [OR] 11.6), concern about a reaction (OR 9.3), and perception that the medical provider had not recommended it (OR 5.8). CONCLUSION:
&mographiC
CharaCbri~tia
Of
patients and their medical diagnoses were not associated with acceptance of influenza vaccination. Among patients who were not vaccinated, doubts about the efficacy of influenza vaccine and fear of its side effects were common, and their perceptions of the medical provider’s recommendation of vaccine appeared to be an im*This is a rapid publication manuscript. From the Section of General Medicine, Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut. Requests for reprints should be addressed to Nicholas H. Fiebach. M.D., Yale-New Haven Hospital, Primary Care Center, 20 York Street, New Haven, Connecticut 06504. Manuscript submitted February 26, 1991, and accepted in revised form August 19. 1991.
portant it.
factor in the decision whether
to accept
I
nfluenza vaccination reduces morbidity and mortality from respiratory infections in targeted populations [l-3] and is therefore recommended for patients with chronic medical conditions and those 65 years or older [4,5]. Nevertheless, during the last decade in the United States, only 32% of the elderly and only 10% of younger patients with chronic diseases received influenza vaccination each year [ 1,6]. Vaccination rates in some university hospital outpatient clinics were reported to be similarly low [7], and in the Primary Care Center at Yale-New Haven Hospital, only one third of patients for whom the vaccine was indicated had received it in previous years (J. Ginetti, unpublished data). The low rates of vaccination have been blamed on administrative obstacles, since physician awareness about the effectiveness of influenza vaccine and its indications has been documented [1,8]. Organized programs aimed at high-risk patients resulted in influenza vaccination rates of 50% to 70% [l], but still fell short of the goal set by the Immunization Practices Advisory Committee of the Centers for Disease Control of vaccination in 80% of patients for whom it is indicated [5]. The possibility that specific patient factors also contribute to nonvaccination has been investigated, and previous studies have shown that demographic characteristics, health beliefs regarding “flu” and the influenza vaccine, previous experience with the vaccine, and relationship to health care providers and the health care system influence vaccination [g-16]. Unfortunately, previous studies of acceptance of influenza vaccination were difficult to apply in our university hospital-based clinic. Some of the studies were based on community surveys [9,10,15] or were conducted during the intense publicity surrounding the swine influenza vaccination campaign in 1976 [9,10,12], and are not applicable to patients attending our clinic. In studies done in clinical settings, patient populations were not described fully enough to allow accurate comparison with our patients [ll-141. Some studies of factors influencing influenza vaccination included outreach efforts by mail or telephone, but patients’ attitudes regarding vaccination were not always assessed prospectively prior to their decision to receive or decline vaccination [11-141. Also, other patient factors, including
October
1991
The American
Journal
of Medicine
Volume
91
393
INFLUENZA VACCINATION / FIEBACH AND VISCOLI
previous compliance with physicians’ recommendations, other preventive health behaviors, and sense of well-being and ability to function, were not considered in previous analyses. We undertook to improve influenza vaccination rates in our clinic by using medical provider education and written reminders. In addition, in order to understand if patient characteristics were important determinants of vaccination in our clinic population, we sought to identify those patients who remained unvaccinated despite this intervention. Specifically, we assessed prospectively the demographic, medical, and personal characteristics of patients, including previous experience with influenza vaccine and beliefs about its efficacy, which may be associated with nonvaccination. We also designed a case-control study nested within the larger cohort study to document patients’ reasons for refusing influenza vaccine and to explore the relation between refusal and patients’ reports of well-being, function, previous compliance with doctors’ recommendations, and beliefs about influenza vaccine.
PATIENTS AND METHODS The study was conducted in the Primary Care Center of Yale-New Haven Hospital. Adult patients in this outpatient facility are cared for by faculty physicians, internal medicine residents, nurse practitioners, and a few physician volunteers. For 2 weeks prior to the study period, one of the authors reviewed the indications, use, and efficacy of influenza vaccine at the daily clinic conference and distributed a vaccine information sheet to the providers. Vaccination was recommended for all patients 65 years of age or older; patients with chronic diseases including congestive heart failure, ischemic cardiac disease, obstructive pulmonary disease, asthma, diabetes mellitus, renal insufficiency, severe anemia, or immunosuppression (including drug-induced, malignancy, and human immunodeficiency virus); household contacts of patients with these high-risk conditions; residents of chronic-care facilities; and health care workers. Providers were also asked about their own histories of influenza vaccination. The first part of the study was a prospective cohort analysis that included all adult patients seen in the Primary Care Center from October 17 through December 15, 1987. Data collection forms were stamped with patients’ hospital identification cards, which include demographic and insurance information, and attached to patients’ charts. As patients arrived in the clinic, nurses recorded information on their previous receipt of and reaction to influenza vaccine and whether they believed the influenza vaccine prevents the “flu.” Providers 394
October 1991 The American Journal of Medicine
completed a checklist of chronic disease indications for influenza vaccine and recorded information on the frequency of clinic visits during the previous year, prior receipt of pneumococcal vaccine, recent receipt of influenza vaccine elsewhere, contraindications to influenza vaccine, and patients’ reasons for refusing influenza vaccine. Finally, clinic nurses who administered vaccine recorded which patients received it. For each patient visit, information was also available on the type of provider and level of training (i.e., faculty, housestaff [postgraduate year 1 to postgraduate year 31, nurse practitioner), whether the provider had attended a vaccine review conference or received a vaccine information sheet, and whether the provider had personally received the influenza vaccine or planned to obtain it. In order to determine whether our patient sample was representative of patients who receive care at the Primary Care Center at other times during the year, we reviewed a convenience sample of patient charts. Every eighth patient was selected from an alphabetical list of Primary Care Center patients who had at least one clinic visit after October 1, 1986, approximately 1 year prior to the start of the study period, and the proportions of patients for whom influenza vaccine was indicated and who attended the clinic during the study period were calculated. In the second part of the study, patients for whom influenza vaccine was indicated but who had refused it were identified. Each refuser was matched to another patient seen by the same provider on the same day who accepted the vaccine. When more than one vaccine recipient was available for matching, one was chosen randomly; if no patients received the vaccine on the same day, then we selected one who came to the clinic on the nearest date. When the study was conducted, residents came to the clinic every 2 weeks or less frequently, so some vaccine refusers could not be matched to a vaccine recipient of the same resident. These vaccine refusers were matched to a vaccine recipient of another resident at the same level of training who came to the clinic on the same or nearest date. Trained research assistants (including one fluent in Spanish) made several attempts to contact matched patients by telephone. These patients were asked to respond to a 30-item questionnaire consisting of open-ended questions about reasons for accepting or refusing influenza vaccination, fear of adverse reactions, and noncompliance with physicians’ recommendations, and 25 scaled items assessing attitudes and beliefs about health care and vaccination (see Appendix). The data were analyzed using SAS statistical software [17]. For prospectively assessed patient
Volume 91
INFLUENZA
characteristics, relative risks and 95% confidence intervals for nonvaccination were determined for patients in whom influenza vaccine was indicated, excluding those who did not receive the vaccine because of egg allergy, acute illness, or temporary shortage of vaccine. The power of these analyses to detect significant associations was calculated [18]. For the case-control analysis of vaccine refusers and recipients, personal and medical characteristics of patients were tabulated and differences between groups were compared with the chi-square test. Median times between clinic visits and telephone interviews were calculated for each group and compared with the rank sum test. Responses to the scaled telephone questionnaire items were dichotomized, and both matched and unmatched analyses were performed to obtain odds ratios and 95% confidence intervals for nonvaccination [19]. RESULTS Prospective Cohort Analyses Data forms were completed for 1,175 visits by 965 patients, representing virtually all of the patients seen in the clinic during the study period. The majority of patients were women and single, and many were nonwhite and elderly (Table I). Most patients had either Medicaid or Medicare coverage and had at least two visits to the clinic during the previous year. About one half were the patients of nine faculty physicians, about one third were the patients of 59 residents, and the remainder were patients of three nurse practitioners and five volunteer physicians. Seventy-seven percent of the patients were seen by providers who had attended a conference or received an information sheet on influenza vaccine before the study period. Vaccination was indicated for 48% of patients because of chronic disease. Diabetes mellitus was the most prevalent chronic disease indication, and ischemit heart disease, congestive heart failure, chronic obstructive pulmonary disease, and asthma were also common. Fourteen percent of patients had more than one chronic disease indication; among diabetic patients, 36% had another chronic disease for which vaccine was indicated. Thirty-six percent of patients were aged 65 years or older, although vaccine was indicated only because of older age in 15%. Altogether, vaccination was indicated for 65% of all patients. Half of the patients for whom influenza vaccine was indicated had received it previously, including 33% who had received it the year before. Fewer patients had received pneumococcal vaccine. One in 10 patients who had previously received influenza vaccine reported feeling sick afterwards. Only 30% believed that influenza vaccine prevents the “flu.” Of the 624 patients for whom vaccine was indicat-
VACCINATION
/ FIEBACH
AND VISCOLI
TABLE I ProspectiveCohort Analysis: Patient Characteristics Patients Influenza Vaccine All Indicated (n = 965)* (n = 624)t Demographic factors Age t 65 Female Nonwhite Married
348 (36%) 348 (56%) 654 (68%) 431 (69%) 479 i56%j 305 i56%j 218 (26%)
150 (28%)
Health insurance+ None Commercial Medicaid Medicare ClinF;isits
during past year
211 (22%) 86 (17%) 345 (36%) 239 (48%) 211 (22%) 176 (35%)
2-4 >5 Medical provider Faculty Housestaff Nurse practitioner Other physicians Exposed to influenza vaccine review Received or planned influenza vaccine
450 (47%) 295 (47%) 357 (37%) 229 (37%)
292 (30%) 185 (30%)
Indications for influenza vaccine Age 2 65 only Medical diagnoses Congestive heart failure lschemic heart disease Chronic obstructive pulmonary disease Asthma Diabetes mellitus Chronic renal failure Significant anemia lmmunosuppressed Nursing home resident Health care provider Any indication Vaccination history and beliefs Ever received influenza vaccine Received influenza vaccine previous year Felt sick after previous influenza vaccinationr’ Ever received pneumococcal vaccine Believed influenza vaccine prevents “flu”
144 (15%)
144 (23%)
396 (41%)
314 (50%)
234 (24%) 205 (33%) 41 (10%) 29 (5%) 115 (12%)
99 (16%)
246 (26%) 184 (30%)
)ata missing for: gender, 5 (1%); race, 117 (12%); marital status, 129 (13%); clinic visits, 19 1%). lata missing for: race.! 74 (12%); marital status, 80 (13%); clinic visits, 123 (20%). Btegorles not mutually excIusIve. ,mongpatients who had previously received influenzavaccine (396 of all patients, 314 for whom
ed, 440 were vaccinated in the Primary Care Center during the study period and 26 reported receiving it elsewhere, resulting in a vaccination rate of 75%. Vaccination rates were similar among patients whose only indication was age 65 years or older (76%), those of older age who also had a chronic disease indication (76%), and younger patients with a chronic disease (73%). Patients with two or more chronic disease indications had a significantly higher vaccination rate than did patients with only one chronic disease (82% versus 71%, x2 = 6.84, p 5
AND VISCOLI
Number*
Patients Not Interviewed* (rl = 80)
65% 72% 58% 30% Health insurance None Commercial Medicaid Medicare
/ FIEBACH
TABLE IV Case-ControlAnalysis: Patients’ Principal Reasonsfor Accepting or Refusing Influenza Vaccination
TABLE Ill Case-ControlAnalysis: Patient Characteristics Interviewed Refused Vaccine (n = 58)
VACCINATION
26 (46%) ;; I;;;; 10 (18;)
4 (7%) 4 (7%) 18 (32%)
,
*Categories not mutually exclusrve. Among recipients, 16 gave two reasons and two gave three reasons; amongrefusers, 15 gave two reasons and five gavethree reasons.
TABLE V Case-ControlAnalysis: Questionnaire ResponsesSignificantly Associatec with Refusing Influenza Vaccination
Question/Response
Odds Ratio (95% Confidence Interval)
Do you think the flu vaccine works well to prevent people from getting sick with the flu? Not well at all or not sure versus very well or fairly well
(6.3,41.0)
Were you very concerned about a bad reaction from the flu vaccine? Very concerned or a little concerned versus not at all concerned
9.3 (3.9,22.3)
Did your doctor recommend that you get the flu vaccine? Did not recommend it versus strongly urged it or recommended it
5.8 (2.5, 13.2)
Do you think it’s possible that the flu vaccine could cause a person to have a severe reaction or a serious illness? Possible or not sure versus rare or never
(1.2,14.8)
In general, are you able to do the things in your home that you would like to do? Always or sometimes versus not usually or never
3.9 (1.5, 10.3)
74%
8
were not Intt?rVleWea: Jb vaccine WplentS and Zb vaccine retUSerS COUM not be ached on the telephone. Nine subjects who had refused vaccine were unable to communrcate equately over the telephoneor were unwrllingto respond to the questronnarre. Two patrents who received vaccme dred before they could be IntervIewed, five patients who refused the vaccine had received it by the trme they were mtervrewed, and two pabents who were coded as havrng recerved vaccine denied It. ‘Comparisons between all mtervrewed pabents and those not IntervIewed: p < 0 05. :eaSOnS patIm
ed that their medical provider had not recommended the vaccine, and more often described themselves as able to function in their homes. Scores for other questions, including those that assessed perceived risk of influenza, importance of vaccine cost, previous medical compliance, and health status, were similar between the two groups. The most common reasons for accepting or refusing vaccine, and the responses to the questionnaire items described above, did not differ between patients aged 65 years or older and younger patients.
COMMENTS Our study included more patients for whom influenza vaccine was indicated than can be found in previous studies of acceptance of influenza vaccination [g-16]. We did not find that demographic factors such as age, race, gender, and marital status were associated with influenza vaccination despite excellent statistical power. Previously reported associations between influenza vaccination and age [ll-131 or race [12,16] may have been due to chance
16.1
4.2
variations among small subgroups, or perhaps reflected a relationship between access to medical care and vaccination rather than an effect on acceptance of the influenza vaccine once it was available. Although a previous study found that patients with pulmonary diseases received influenza vaccine less often than other patients for whom it was indicated [12], we found no association between vaccination and chronic disease diagnoses. Contraindications to the vaccine such as acute febrile illness or egg allergy were rare, and the major reason for nonvaccination was patient refusal. Responses to questions in both the prospective and
October 1991
The American
Journal
of Medicine
Volume
91
397
INFLUENZA VACCINATION / FIEBACH AND VISCOLI
case-control analyses in our study, and results of previous studies, showed that the factor most strongly associated with nonvaccination was not believing that the vaccine prevents the “flu” [10,11,14,15]. However, the majority of patients who accepted vaccination also thought it would not prevent the “flu” when asked prospectively. Since vaccine recipients who were interviewed retrospectively often listed prevention of illness as their major reason for accepting the vaccine, this may reflect persuasion by their providers or rationalization for their action. In contrast, only 7% of interviewed vaccine refusers cited lack of effectiveness as their main reason for refusing the vaccine. These results suggest that while many patients have doubts about the efficacy of the influenza vaccine, this may not be the decisive factor in refusal of vaccination. Although a previous reaction to influenza vaccine was significantly associated with nonvaccination in our study and others [13-151, only 10% of all previously vaccinated patients reported a prior reaction in the prospective analysis. Other studies [20,21] and public health directives [4] also emphasize that serious reactions to influenza vaccine are infrequent. Nevertheless, in retrospective interviews, fear of a reaction was cited as the main reason for refusal by almost half of the vaccine refusers, and responses to specific questionnaire items confirmed that vaccine refusers were significantly more likely to worry about adverse reactions than vaccine recipients. The disparity between the actual and perceived risk from influenza vaccine seems to be a major reason why vaccination is refused by some patients [12]. Our study was not specifically designed to examine the effect of medical providers’ attitudes and behavior on patients’ acceptance of influenza vaccine, but their impact is suggested by the variation in vaccination rates among providers. Differences in type of provider, level of training, and exposure to recent information on influenza vaccine did not account for this variation. Although it may reflect random effects among small patient subgroups or real differences among the patients cared for by individual providers, the variation in vaccination rates was probably also the result of differences among providers in compulsiveness, enthusiasm, and persuasiveness in recommending vaccine. Although 40% of interviewed vaccine recipients cited their provider’s recommendation as the major reason for accepting the vaccine, vaccine refusers were six times more likely to report that their provider had not recommended it. We were not able to determine whether providers really did not recommend vaccine to these refusers, whether this represents rationalization by patients for not accepting vaccination, or whether patients and providers differed 398
October 1991 The American Journal of Medicine
in their perception of what was said in their interaction. Nevertheless, it appears that providers’ attitudes and recommendations are an important factor in patients’ vaccination decisions [10,15,16]. We hypothesized that patients’ previous compliance with treatment, sense of well-being and ability to function, and other preventive health behavior would influence acceptance of influenza vaccine, but this was not confirmed. Although vaccine refusers tended to score higher on questions about sense of well-being and function, and lower on questions about other preventive health behaviors, the difference was statistically significant only for the question that assessed function at home. Unlike previous studies that specifically sought to validate a “health belief model” [22] for influenza vaccination [g-11,13], we did not find a significant difference between vaccine recipients and refusers in their perceptions of susceptibility to influenza or its potential severity, or the importance of the cost of the vaccine. Two studies that found that vaccination was significantly associated with perceived susceptibility to influenza and its severity were conducted during the swine influenza campaign [9,10], and it is possible that widespread publicity about the seriousness of the anticipated swine influenza epidemic temporarily raised awareness of influenza and concern about its consequences. However, Larson and colleagues [ll] used a retrospective questionnaire to assess the vaccination decisions of patients attending a university medical clinic the year before the swine influenza campaign, and found that concerns about susceptibility to influenza, its severity, and cost of the vaccine were significantly associated with vaccination. Although the patients in that study resembled our sample in being predominantly older and female, other demographic comparisons are not possible. It is conceivable that differences in socioeconomic factors or secular trends in attitudes toward influenza and vaccination contributed to the conflicting results. The contrast between the results of the Larson study and ours also illustrates a crucial distinction in studies of vaccination outcomes. The “health belief model” emphasizes preventive health careseeking behavior by patients [ll], and studies of influenza vaccination that have tested its hypotheses have focused on whether patients came to the clinic to receive vaccine and the factors that influenced this action. Our study, on the other hand, analyzed the commonly encountered situation of a high-risk patient coming to a clinic but not being vaccinated, a so-called “missed opportunity” for vaccination [5]. Although our results apply only to patients who came to the clinic during the fall season, when vaccination is recommended, a convenience sample of clinic charts showed that approxi-
Volume 91
INFLUENZA VACCINATION / FIEBACH AND VISCOLI
mately three quarters of our patients for whom influenza vaccine was indicated had attended the clinic during the study period. We recognize that the “missing” patients may have had different attitudes toward influenza vaccine, that other patient populations with fewer indications for vaccine or less frequent visits to a clinic may have different determinants of vaccination, and that getting eligible patients to the clinic during the fall vaccination period will be necessary to achieve vaccination goals. Several other limitations of our study deserve mention. We had prospective information on some demographic characteristics, but lacked data on education, employment, and financial status other than insurance coverage. We were able to measure patients’ beliefs about the efficacy of influenza vaccine and prior experiences with it before they decided about vaccination, but we relied on a retrospective case-control analysis to assess patients’ decisions in more detail. Although the telephone questionnaire we used was straightforward and possessed face validity, it was not formally tested for reliability and accuracy. A substantial number of patients who could not be reached on the telephone were excluded from the case-control analysis. Although several characteristics differed significantly between interviewed and noninterviewed patients, only one of these, number of clinic visits, was predictive of vaccination in the prospective analysis. Perhaps more importantly, the reasons for exclusion differed between vaccine recipients and refusers who could not be interviewed. Among noninterviewed subjects, more vaccine recipients did not have a telephone, did not answer it, or were unable or unwilling to be interviewed after they were contacted. We do not know whether these circumstances were associated with factors that influence vaccination attitudes and decisions. Despite these limitations, our results show that some patients remain unconvinced of the benefit of influenza vaccine and wary of its side effects. Adverse publicity about the swine influenza vaccination program in 1976 may still contribute to this, although influenza vaccines in subsequent years have been studied extensively and no association with neurologic sequelae has been found [ZO]. Also, confusion between lay and medical uses of the term “flu” has obscured the important point that the main benefit of influenza vaccine is not to forestall “flu” symptoms, but rather to prevent serious complications and deaths during influenza epidemics [l-5]. Since patients’ perceptions of recommendations for influenza vaccine appear to be important in vaccination decisions, health care providers should not assume that there is a substantial group
of recalcitrant patients who will not accept vaccination if it is offered. Clearer and more widespread messages about the safety and efficacy of influenza vaccine in conjunction with unambiguous recommendations to patients for whom it is indicated may help achieve desired levels of vaccination.
APPENDIX Telephone Interview Questionnaire 1. What werevour reasons for acceotineinot acceotine the flu vaccine? 2. What was the most important reaso? ’ 3. When you think of “the flu,” what do you think of? - a cold _ a bad cold - worse than a bad cold - other 4. How would you rate your care at the Primary Care Center? - verveood - eood - fair Door 5. Do you u&ally follow;our doctors recommendations? 6. What doctors recommendations have you not followed? 7. Does your doctor prescribe medication for you? - to take every day - only when you are sick none 8. Do you take your prescribed medication? - almost always - sometimes - never 9. How often do you miss appointments with the doctor? - frequently __ sometimes _ rarely 10. Is it likely that you would get the flu this winter? probably would not sure - probably would not 11. If YOUwere to net the flu, is it likelv that YOUwould be verv ill? not very ifi - somewhat iii i very seriously iI1 not sure 12. Do you think the flu vaccine works well to prevent people from getting sick with the flu? - very well - farrly well - not well at all not sure 13. Was there an extra charge to you to get the flu vaccine? - yes _ not sure - no 14. Was the cost an important factor in your decision whether to get the flu vaccine? - very important - one of several considerations - not important 15. Were you very concerned about a bad reaction from the flu vaccine? - very concerned - a little concerned not concerned at all 16. What bad reactions were you worried about? 17. Do you think it’s possible that the flu vaccine could cause a person to have a severe reaction or a serious illness? possible _ rare possibility _ no, never not sure 18. f&e many of your friends and relatives received the flu vaccine? many - some - few or none 19. %ve any of your friends and relatives had a bad reaction from the flu vaccine? - yes _ not sure - no 20. Have you heard on TV or radio or read in a newspaper or magazine that flu vaccine is recommended for a person like you? - yes _ not sure no 21. Did your doctor recommend that you get the flu vaccine? strongly urged you to get it recommended it - did not recommend it 22. Are you nervous when you see your doctor? - very nervous - a little nervous not nervous at all 23. Are you scared of needles? - very scared - a little scared not scared 24. Do vou smoke ciearettes? -‘current smoker - former smoker non-smoker 25. Do you wear a seat belt in the car? --always sometimes never 26. Do vou eat a healthv diet? -‘always _ sometimes _ rarely 27. Do you have any serious illnesses? - yes _ no - notsure 28. In general, would you say that your health is good? - very good L good i fair poor 29. In general, are you able to do the things in your home that you would like to do? - always - sometimes not able to - never or usually not able to 30. In general, do you get out of the house a lot? - frequently - sometimes - rarely, never
October 1991 The American Journal of Medicine
Volume 91
399
INFLUENZA VACCINATION / FIEBACH AND VSCOLI
ACKNOWLEDGMENT We thank Ruth Gemmell for assistance in preparing the manuscript, Pat Bennett for help in obtaining patient charts, Ralph Horwitz, William Beckett. and Maryann Zitt for advice, and the nurses, clerical staff, and housestaff in the Primary Care Center for their cooperation in collecting the data.
REFERENCES 1. Fedson DS. Influenza prevention and cure. Past practices and future prospects. Am J Med 1987; 82: 42-7. 2. Barker WH. Mullooly JP. Influenza vaccination of elderly persons. Reduction of pneumonia and influenza hospitalizations and deaths. JAMA 1980; 244: 2547-9. 3. Gross PA, Quinnan GV, Rodstein M. et a/. Association of influenza immunization with reduction in mortality in an elderly population. A prospective study. Arch Intern Med 1988; 148: 562-5. 4. Prevention and control of influenza: Part 1. Vaccines. MMWR 1989; 38: 297-311. 5. Williams WW. Hickson MA, Kane MA, eta/. Immunization policies and vaccine coverage among adults. The risk for missed opportunities. Ann Intern Med 1988; 108: 61625. 6. Influenza vaccination levels in selected states-behavioral risk factor surveillance system, 1987. MMWR 1989; 38: 124-33. 7. Kosecoff J, Fink A, Brook RH. General medical care and the education of internists in university hospitals. An evaluation of the teaching hospital general medicine group practice program. Ann Intern Med 1985; 102: 250-7. 8. Cate TR, Ruben FL. Prevention, management, and control of influenza in internal medicine. Am J Med 1987; 82: 52-4. 9. Rundall TG, Wheeler JRC. Factors associated with utilization of the swine flu vaccination program among senior citizens in Tompkins County. Med Care
400
October 1991 The American Journal of Medicine
1979; 17: 191-200. 10. Cummings KM, Jette AM, Brock BM. Haefner DP. Psychosocial determinants of immunization behavior in a swine influenza campaign. Med Care 1979; 17: 639-49. 11. Larson EB. Olsen E. Cole W. Shortell S. The relationship of health benefits and a postcard reminder to influenza vaccination. J Fam Pratt 1979; 8: 1207-11. 12. Anderson C. Martin H. Effectiveness of patient recall system on immunization rates for influenza. J Fam Pratt 1979; 9: 727-30. 13. Larson EB. Bergman J, Heidrich F. Alvin BL, Schneeweiss R. Do postcard reminders improve influenza vaccination compliance? A prospective trial of different postcard “cues.” Med Care 1982; 20: 639-48. 14. Frank JW, Henderson M. McMurray L. Influenza vaccination in the elderly: 1. Determinants of acceptance. Can Med Assoc J 1985; 132: 371-5. 15. Carter WB. Beach LR. lnui TS. Kirscht JP, Prodzinski JC. Developing and testing a decision model for predicting influenza vaccination compliance. Health Serv Res 1986; 20: 897-932. 16. Adult immunization: knowledge,
attitudes,
and practices-DeKalb
and Ful-
ton counties, Georgia, 1988. MMWR 1988; 37: 657-62. 17. SAS Institute Inc. SAS/STAT user’s guide, release 6.03 ed. Cary, North Carolina: SAS Institute Inc.. 1988. 18. O’Connor GT. Beck JR. Epi-talc. Hanover: True Basic, Inc., 1987. 19. Rothman KJ. Modern epidemiology. 1st ed. Boston: Little Brown, 1986: 28&l. 20. LaMontagne JR, Quinnan GV, Curlin GT, et a/. Summary of clinical trials of inactivated influenza vaccine-1978. Rev Infect Dis 1983; 5: 723-36. 21. Margolis KL. Nichol KL. Poland GA, Pluhar RE. Frequency of adverse reactions to influenza vaccine in the elderly: a randomized, placebo-controlled trial. JAMA 1990; 264: 1139-41. 22. Becker MH. Maiman LA. Sociobehavioral determinants of compliance with health and medical care recommendations. Med Care 1975; 13: 10-24.
Volume 91
