Journal of Gerontology 1979, Vol. 34, No. 2, 201-208
Participation of Senior Citizens in the Swine Flu Inoculation Program: An Analysis of Health Belief Model Variables in Preventive Health Behavior1 Evidence is presented of statistically significant and strong relationships between Swine Flu Inoculation status and nine variables in the reformulated Health Belief Model with 122 randomly selected subjects, primarily Black and Portuguese-American, who are active members of two Providence, Rhode Island senior centers. No statistically significant relationship was discovered between inoculation status and previously having had the flu. The variables which were found related are: Efficacy, Safety, Knowledge of Side Effects, Prior Flu Shot Status, Proportion of Friends and Relatives Who Got the Shot, Sex, Race, Future Plans for Flu Shots, and Future Plans for Other Inoculations. The data were obtained through personal interviews in the Spring of 1977. It is suggested that the results provide some basis for optimism for successful intervention designed to change the future preventive health behaviors of nonparticipants in the Swine Flu Inoculation Program. Many nonparticipants had fears and doubts about the effectiveness and safety of the shot and are amenable to suggestions from physicians about future inoculation participation. Full information should be provided to high-risk groups such as senior citizens about the relative risks of suffering serious side effects, the effectiveness and safety of the procedure for persons their age with the typical health problems of senior citizens, and the relative risks and dangers to them of contracting the illness against which the shot is designed to protect them.
N the May, 1977 Supplement to Medical Care on "Issues in Promoting Health" IBecker and associates present a review of re-
search on individual health-related behavior, using the Health Belief Model (Fig. 1) as an organizing rubric. The authors propose a revised model for explaining and predicting health-related behaviors (Fig. 2), call for less emphasis in future research on variables previously shown to be unrelated to health behaviors and for work with varying population groups and in different settings (Becker et al., 1977). The research reported on in this article represents a contribution to the further refinement of the Health Belief Model, providing evidence about the role of several variables in the re'When this article was written the author was with the Program in Gerontology of the Univ. of Rhode Island. Kingston. RI. The author gratefully acknowledges the support of the Univ. of R.I. University Research Committee and the College of Home Economics, the cooperation of the staff and members of the two cooperating senior centers and the student and senior citizen interviewers. The University's Academic Computer Center handled the data processing. 2 Assoc. Prof, of Sociology, Dept. of Sociology and Social Welfare. Rhode Island College, Providence 02908.
formulated model in the 1976 Swine Flu Inoculation Program. The respondents are in the high-risk senior citizen category, active members of two senior centers in Providence, Rhode Island providing services primarily for Black and Portuguese-American senior citizens. BACKGROUND
The existence of the Becker et al. (1977) article, recent and readily available to readers of this journal, precludes the necessity of a detailed and lengthy discussion of the Health Belief Model. The model contains variables which are felt to be intervenable and modifiable; has core dimensions based on a wellestablished body of psychological and behavioral theory, particularly that of Lewin (1944), and focuses on the valence of a goal, the subjective probability of its attainment, and motivation. Originally used to account for preventive health actions, the model has also been used to explain illness and sick-role 201
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
William R. Aho, PhD2
202
AHO
INDIVIDUAL PERCEPTIONS
MODIFYING FACTORS
Demographic variables (age, sex, race. ethnicity, etc.) Sociopsychological variables (personality, social class, peer and reference group pressure, etc.)
LIKELIHOOD OF ACTION
Perceived benefits of preventive action ^minus Perceived barriers to preventive action
r
Perceived Susceptibility to Disease " X " Perceived Seriousness (Severity) of Disease " X "
\
\ i
f
Perceived Threat Disease " X "
Likelihood of Taking Health Action
ii
Cues to Action Mass media campaigns Advice from others Reminder postcard from physician or dentist Illness of family member or friend Newspaper or magazine article
Fig. 1. Variables and relationships in the original Health Belief Model (Becker & Maiman, 1975).
READINESS TO UNDERTAKE RECOMMENDED COMPLIANCE BEHAVIOR
MODIFYING AND ENABLING FACTORS
COMPLIANT BEHAVIORS
Motivations Concern about (salience of) health matters in general Willingness to seek and accept medical direction Intention to comply Positive health activities Demographic (very young or old) Value of Illness Threat Reduction Subjective estimates of: Susceptibility or resusceptibility (incl. belief in diagnosis) Vulnerability to illness in general Extent of possible bodily harm* Extent of possible interference with social roles" Presence of (or past experience with) symptoms
Probability That Compliant Behavior Will Reduce the Threat
Structural (cost, duration, complexity, side effects, accessibility of regimen; need for new panerns of behavior) Attitudes (satisfaction with visit, physician, other staff, clinic procedures and facilities) Interaction (length, depth, continuity, mutuality of expectation, quality, and type of doctorpatient relationship; physician agreement with patient; feedback to patient) Enabling (prior experience with action, illness or regimen; source of advice and referral (incl. social pressure)
Likelihood of: Compliance with preventive health recommendations and prescribed regimens: e.g., screening, immunizations, prophylactic exams, drugs, diet, exercise, personal and work habits, follow-up tests, referrals and follow-up appointments, entering or continuing a treatment program.
Subjective estimates of: The proposed regimen's safety The proposed regimen's efficacy to prevent, delay, or cure (incl. "faith in doctors and medical care" and "chance of recovery")
'At motivating, but not inhibiting, levels.
Fig. 2. Summary hypothesized model for explaining and predicting individual health-related behaviors (Becker & Maiman, 1975).
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
\
SENIOR CITIZENS AND SWINE FLU
behaviors. The Becker et al. article points out that numerous other models contain similar variables.
203
(4) Demographic (race and sex); (5) Structural (knowledge of side effects); (6) Enabling (prior experience with the shot; proportion of friends and relatives who got the shot).
METHODOLOGY
FINDINGS
The respondents were asked questions designed to elicit attitudes, beliefs, and facts relevant to ten variables in six categories of the expanded Health Belief Model portrayed in Fig. 2 and first presented in an earlier article by Becker and Maiman (1975). The six categories, with a parenthetical statement about how they were operationalized, are: (1) Motivation (plans for future flu and other shots); (2) Value of Threat Reduction (prior experience with the flu); (3) Probability that Behavior Will Reduce the Threat (perceived efficacy and safety);
While no analysis of age differences is presented here, the model suggests that those who are very young or very old are more likely than others to participate in preventive health behavior. Our findings show that for our aged 60 and over respondents 54% received the Swine Flu shot. Nationally, for persons aged 50 and over, the figure is 44% (Center for Disease Control, 1977). Motivation
Future inoculation plans. — The Health Belief Model includes intention to comply as a motivational variable affecting the readiness to undertake compliance behavior. Respondents were asked if they planned to get other shots in the future to protect them against the flu. An examination of Table 1-A indicates that there is a statistically significant and strong relationship between inoculation status and future flu shot plans — nearly three-fourths of those who got the Swine Flu shot plan future flu shots, compared to just over 10% of those who did not get the shot. There is reason for some optimism about these findings because nearly one-fifth of those who didn't get the shot may get one in the future depending on advice from their physician, and 12% said outright that they do plan to get a future flu shot. Here, then, are nearly one-third of the nonparticipants comprising a "swing" group which may move
Table 1-A. Future Flu Shot Plans, by Inoculation Status (%).
Future Flu Shot Plans Yes No Depends on Drs. advice Depends (other) or Don't know Total % Total N
Inoculation Status Did Not Got Shot Get Shot 73 08 09
12 60 19
Total 46 31 14
10
10
10
100 (64)
101 (52)
101 (116)
Chi-Square = 52.26; 3 df; p = < .001 Cramer's V = .67.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Personal interviews of an average one-half hour duration were conducted during the months of March and April, 1977, with 122 senior citizens aged 60 and over who were active members of the two senior centers. An active member was operationally defined as one who had visited the meal site at the senior center at least two times within the two months preceding the sample selection. Over one-half of the interviews were conducted by other senior citizens, themselves active senior center members. While most of the interviews were conducted at the centers, over one-third were conducted in the respondents' homes to shorten and facilitate completion of the study. A 45-item pretested interview schedule was used for interviewing. Random selection was made by prenumbering the registration cards of the members and using a table of random numbers. Sixty-three percent of the sample chosen responded; this represents 28% of the total active population. One of the senior centers serves primarily a Black population; the other a PortugueseAmerican population. The numbers involved are too small to permit a refined and valid separate analysis within centers of those who did and did not receive a Swine Flu shot.
204
AHO
Table 2. Previously Had Flu, By Swine Flu Inoculation Status (%).
Table 1-B. Plans for Future Shots Other Than Flu Shots, by Inoculation Status (%)
Plans to get Other Future Shots — not Flu Yes No Depends on Drs. advice Depends (other) or Don't know
56 19 08
11 56 13
Total 35 36 10
17
_2_1_
18
100 (64)
101 (55)
99 (119)
Chi-Square = 28.06; 3 df; p = < .001; Cramer's V = .49.
over into the "Got Shot" category in some future campaign. Adding in even one-half of the category, "Depends, Other or Don't Know" increases the size of this swing group to a sizable 41%. These results are even more promising when it is remembered that they were obtained from interviews held after the mass media messages about the dangerous side effects of the Swine Flu shot. The bad news, of course, is that fewer than three-fourths of those who got the shot plan to get one in the future. They may be convinced otherwise if and when the opportunity and need for a future shot arise; or their loss might be offset by convincing nonparticipants to get a shot. Table 1-B presents data on future plans for shots for conditions other than the flu and indicates that the relationship between flu shot status and plans for future non-flu shots is also statistically significant and strong, with a pattern quite similar to that of the data in Table 1-A, with the main exception that there are more "Don't Know's" among the non-flu-shot group. Again, over one-fourth of the latter may get future shots, continuing the note of optimism from the previous Table. The fact that fewer of those who got the flu shot would get shots for other conditions (56%) may say something about their preference for a flu shot over other shots. Perhaps the fact of having had the shot and not suffered either any ill effects or the flu itself is responsible for this. Again, the size of the "Depends, Other or Don't Know" category suggests there is room for intervention and modification of behavior in future shot campaigns.
Previously Had Flu Yes No Total % Total N
Inoculation Status Did Not Got Shot Get Shot 49 52 101 (66)
34 66 100 (56)
Total 42 58 100 (122)
Chi-Square = 2.07; I df; p = < .15; Phi = .15.
Value of Illness Threat Reduction Previous flu status. — This category in the model includes the variable of past experience with symptoms of the condition in question. The respondents were asked whether or not they had ever had the flu. The results are reported in Table 2. There is no statistically significant relationship between prior flu status and Swine Flu inoculation status, although the percentages reveal some directional support for the implied direct relationship between getting the shot and having had the flu. Probability that Compliant Behavior Will Reduce the Threat Efficacy. — The research reviewed by Becker and associates (1977) reflects general support for the belief that the perceived efficacy and perceived safety of a preventive health behavior are positively related to participation. In a 1959 review of polio vaccinations Rosenstock et al. found support for these variables, although Leventhal et al. (1960) did not measure these benefits and barriers to action in their 1960 study of the Asian Flu epidemic. Perceptions about the effectiveness of treatment for heart disease has been found to be significantly related to the preventive health orientations of wives being interviewed about their roles in preventing heart disease in their husbands (Aho, 1977). Table 3 presents data on perceived efficacy and inoculation status. The respondents were asked how much the Swine Flu shot helped protect them from getting the Swine Flu. For those who did not get the shot, the question was phrased, "How much would the Swine Flu shot have helped protect you?" There is a very significant and strong relationship shown between perceived
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Total % Total N
Inoculation Status Did Not Got Shot Get Shot
SENIOR CITIZENS AND SWINE FLU
Table 3. Belief in Degree of Protection, by Inoculation Status (%).
Belief About Degree of Protection Completely protected Almost completely Somewhat A little, or not at all Don't know
49 32 II
Total
02 08
00 02 07 50 41
102
100
100
(66)
(54)
(120)
27 18 09 23 23
Chi-Square = 83.25; 4 cIf; p = < .001; V = .83.
Table 4. Belief About Likelihood of Getting the Swine Flu, by Inoculation Status (%).
Belief About Getting Swine Flu Very unlikely Unlikely Very likely, or likely Don't know Total % Total N
Inoculation Status Did Not Get Shot Got Shot
Total 40 34 11 15
38 45 09 08
43 20 14 23
100
100
100
(66)
(56)
(122)
Chi-Square = 11.84; 3 df; p = < .01; Cramer's V = .31.
efficacy and inoculation status, but among those who didn't get the shot there are many "Don't Knows" (41%). Another way of getting at beliefs about efficacy for those who got the shot, and discovering how well protected those who didn't get the shot feel, is to ask about the likelihood of getting the Swine Flu. As indicated from analysis of the data reported in Table 4, there is a significant relationship between inoculation status and how likely respondents feel they are to get the Swine Flu. The relationship is far from clear-cut or simple, however. It raises questions about why more of those who didn't get the shot say it is very unlikely that they will get the flu. This may, of course, be a reflection of a factual situation — it was already April and they had not had the flu. But that does not explain why fewer of those who actually got the shot felt it very unlikely they would get the flu. It may be that the latter are generally more cautious or less confident about their health status, and that
is in fact why they got the shot in the first place. It may also be a matter of cognitive dissonance for those who did not get the shot — they are displaying the conviction that all is going to be well with them, that they made the right decision in not getting the shot and choose not to think otherwise. The large percentage of "Don't Knows" for the nonparticipants (23%) coupled with the 14% who felt it very likely or likely they would get the Swine Flu, reveals a situation of doubt, uncertainty and possibly anxiety for quite a large proportion of the nonparticipants. This may help explain the previously reported finding that many of the nonparticipants may plan to get future shots. A companion item asked about the likelihood of other people in Rhode Island getting the Swine Flu if they did not have the shot. The findings here were consistent with those reported in Table 4, but even more statistically significant and stronger. Another interview item bearing on perceived efficacy dealt with beliefs about how serious the Swine Flu would be for the respondent if she or he got it this year. There was no statistical significance for this relationship, although more of those who got the shot felt it would not be at all serious, and nearly onehalf of those who did not reported they didn't know (see Table 5). Safety. — There is a significant and strong relationship between inoculation status and belief about the safety of the Swine Flu shot (see Table 6-A). Over one-fifth of those who didn't get the shot, however, believe it to be Table 5. Beliefs About How Serious Swine Flu Would be for Respondent, if He/She got Swine Flu This Year, by Inoculation Status (%). Belief About How Serious Very serious Somewhat serious A little serious Not at all serious Don't know Total % Total N
Inoculation Status Did Not - Got Shot Get Shot 15 09 14 38 23
11 07 14 21 46
Total 13 08 14 31 _34
99
99
100
(65)
(56)
(121)
Chi-Square = 8.18; 4df; p = < .10; Cramer's V = .26.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Total % Total N
Inoculation Status Did Not Get Shot Got Shot
205
206
AHO
Table 7. Sex and Race of Respondent, by Inoculation Status (%).
Table 6-A. Belief About Safety of Swine Flu Shot, by Inoculation Status (%).
Belief About Safety of Shot Safe Not safe Don't know Total % Total N
Inoculation Status Did Not Got Shot Get Shot
Total
85 06 09
22 51 27
56 26 17
100 (66)
100 (55)
99 (121)
Sex and Race Male Female White Black Other
Inoculation Status Did Not Got Shot Get Shot 38 62 44 42 14
21 79 47 51 02
Total 30 70 45 46 09
Table 6-B. Safety of Shot Last Fall, by Inoculation Status (%).
Did You Believe Shot Was Safe Last Fall? Yes No Don't know Total % Total N
Inoculation Status Did Not Got Shot Get Shot 92 06 02 100 (63)
Table 8. Heard of Someone Getting III From the Shot, by Inoculation Status (%). Total
26 52 22
62 27 II
100 (54)
100 (117)
Chi-Square = 53.82; 2 df; p = < .001; Cramer's V = .68.
Heard of Someone Getting III from Shot Yes No Total % Total N
Inoculation Status Did Not Got Shot Get Shot
Total
49 51
68 32
58 42
100 (65)
100 (56)
(121)
100
Chi-Square = 3.55; I df; p = < .06; Phi = .19.
safe, with another 27% saying they don't know. This suggests that there were other reasons for not participating in the flu shot program. Other data from this study reveal that the most frequently cited reasons given for not getting the shot were "doctor's advice," "don't think it's effective," and "just don't get shots." Respondents were also asked if they believed last Fall that the shot was safe. The results, reported in Table 6-B, are very similar to those reported in Table 6-A — a significant and strong relationship exists between shot status and believing last Fall that it was safe. There is some downward shift between Fall and Spring for both categories of respondents and this may help to explain why some who got the shot would not get it again. Demographic Variables Sex and race. — Among those who got the Swine Flu shot, males were overrepresented and females underrepresented, relative to their proportions in the sample (see Table 7). This is the reverse of the usual pattern found in studies of the utilization of health
services (Aday, 1972; McKinlay, 1972). Blacks and whites were both underrepresented, with the overrepresentation going to those of other races, in this case people from the Portuguese-American senior center who are of Cape Verdean background. While the customary level of statistical significance is not found for these relationships, the arguments put forth by Skipper et al. (1967) and by Labovitz (1968) for accepting levels higher than .05 justify accepting these levels as significant. Structural Variables Side effects. — One important structural variable mentioned by Becker et al. (1977), is side effects of the preventive health behavior. This was a controversial and central factor in the Swine Flu Inoculation Program, eventuating in the early cancellation of the program in December. The respondents were asked if they had heard of anyone getting ill from the shot, and the results are reported in Table 8.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Note: For Race, there are six missing observations; Significance is < .07 for Sex, < .06 for Race. Phi. = .18 for Sex; Cramer's V = .22 for Race.
Chi-Square = 47.39; 2 df; p = < .001; Cramer's V = .63.
SENIOR CITIZENS AND SWINE FLU
Table 10. Proportion of Friends and Relatives Who Got Shot, by Inoculation Status.
Proportion of Friends and Relatives All or Nearly all Several Just a few None Don't know Total % Total N
Inoculation Status Did Not Get Shot Got Shot
Total
29 41 17 05 09
04 25 38 25 09
17 34 26 14 09
101 (66)
101 (56)
100 (122)
Chi-Square = 27.58; 4 df; p = < .001; Cramer's V = .48.
shot. The results, reported in Table 10, indicate that there is a statistically significant and strong relationship between these variables. At the extremes, one-fourth of those who didn't get the shot, compared to just 5% of those who did, reported that none of their friends or relatives had received a shot, Enabling Factors Prior flu shots. — A final category of var- and nearly the reverse percentages reported iables contains prior experience with the pre- that all or nearly all had received the shot. ventive health action, and social pressure. The respondents were asked whether or not they had previously had the flu. The results, DISCUSSION reported in Table 9, are statistically signifiWhile this study was retrospective, the cant. Since 79% of the nonparticipants had interviews were conducted within four months not previously received any kind of flu shot, after the height of the Swine Flu Inoculation they appear to constitute a category of "hard- Program and some retrospective questions core" nonparticipants, given their age and the were asked to discover the time order of fact that they have lived through several events, attitudes or behaviors. In addition, epidemics of flu and presumably had oppor- several items relevant to a key model varitunities and cause to obtain shots in the past. able, Efficacy, were used to provide a check on the reliability of the items. The consisSocial pressure. — An indirect measure tency of the findings, arising as they do in a of social pressure to obtain the shot was ob- previously unexplored setting with different tained by asking about the proportion of and unique subjects and with a health probfriends and relatives who had received the lem of recent and wide-spread interest, enhances the model as a useful framework for continuing research and refinement and efTable 9. Previous Flu Shot Status, by forts at application by health care providers. Swine Flu Inoculation Status (%). The findings provide some basis for optimism that behaviors can be changed, because Inoculation Status of the rather sizable percentages of nonparPreviously Got Did Not ticipants who responded "Don't Know" or Flu Shots Got Shot Get Shot Total "It depends on my doctor's advice" to items Yes dealing with future preventive health be62 21 43 No 39 79 57 haviors. This suggests that efforts with phyTotal % 101 100 100 sicians as well as potential recipients of inocuTotal N (56) (121) (65) lations might prove useful. To allay some of Chi-Square = 18.14; I df; p = < .001; Cramer's V = .40.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Using the rationale of Labovitz and Skipper and associates again, a significance level of .06 is viewed as acceptable, and the data offered as support for the contention that knowledge of bad side effects differentiates those who got the shot from those who did not. Unfortunately, we do not have information about when the respondents learned of the side effects. It may well have been after they had decided not to get a shot. If so, this may be evidence of selective perception on their part to validate their prior decision not to participate. Furthermore, nearly onehalf of those who got the shot had heard of someone who became ill from the shot, a finding which fits with the assumption that this knowledge came after the shots were given. The respondents were also asked to report the number of persons they heard became ill from the shot, but the results were not statistically significant (p = < .44).
207
208
AHO
the fears of senior citizens about future inoculations, accurate and full information should be provided to them about the safety of the procedure for people of their age with health problems they are likely to have, the relative risks of suffering serious side effects, and the relative risks and dangers to them of contracting the illness the shot is designed to protect them against.
This paper reports the results of a 1977 survey involving personal interviews with 122 randomly selected senior citizens in Providence, Rhode Island. The interviewees, primarily Black and Portuguese-American, were asked about their participation in the Swine Flu Inoculation Program and their plans for getting flu shots in the future. All of those interviewed were active members of two different senior centers in Providence. The results of the survey were analyzed by comparing answers of those who got the shot to those who did not. A number of factors felt to be related to having received a shot were used in the analysis to see if they were in fact related, and if they differentiated between those who did, and those who didn't, get the shot. These included: the effectiveness of the shot; its safety, its side effects, how many friends or relatives got the shot, sex, race, plans for future shots for flu or other conditions, and previously having had the flu. All except the last factor were found to be related in a statistically significant way with having had a swine flu shot. Specifically, those who got the shot, compared to those who didn't, (a) (b) (c) (d)
Plan to get a flu shot in the future Feel the shot is effective Feel the shot is safe Feel the shot does not have bad side effects (e) Had previously had a flu shot (f) Reported nearly all their friends or relatives had received the swine flu shot.
REFERENCES
Aday, L., & Eichorn, R. The utilization of health services: indices and correlates. A research bibliography. U.S. DHEW, Washington, DC, 1972. Aho, W. R. Relationship of wives' preventive health orientation to their beliefs about heart disease in husbands. Public Health Reports, 1977, 92, 65-71. Becker, M., Haefner,^)., Kasl, S., Kirscht, J., Maiman, L., & Rosenstock, I. Selected psychosocial models and correlates of individual health-related behaviors. Medical Care, 1977, 15, 27-46. Becker, M., & Maiman L. Sociobehavioral determinants of compliance with health and medical care recommendations. Medical Care, 1975, 13, 10-24. Center for Disease Control. National survey of public attitudes toward A/New Jersey/76 influenza vaccination: Report No. 8, Table 3, March 31, 1977, Atlanta, GA (Mimeo). Labovitz, S. Criteria for selecting a significance level: A note on the sacredness of .05. American Sociologist, 1968, 3, 220-222. Leventhal, H., Hochbaum, G., & Rosenstock, I. Epidemic impact on the general population in two cities. In The Impact of Asian Influenza on community life: A study in five cities, USPHS, Washington, DC 1960. Lewin, K., Dembo, T., Festinger, L., & Sears, P. S. Level of aspiration. In J. McV. Hunt,(Ed.), Ronald Press, NY, 1944. McKinlay, J. Some approaches and problems in the study of the use of services — an overview. Journal of Health and Social Behavior, 1972,13, 115-152. Rosenstock, I., Derryberry, M., & Carriger, B. Why people fail to seek poliomyelitis vaccination. PublicHealth Reports, 1959, 74, 98-103. Skipper, J. K. Jr., Guenther, A., & Nass, G. The sacredness of .05: A note concerning the uses of statistical levels of significance in social science. American Sociologist, 1967,2, 16-18.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
SUMMARY
The results provide some basis for optimism that the behavior of senior citizens relative to getting preventive shots can be changed, based on the proportions answering "Don't Know", or "It depends on my doctor's advice" when asked about whether they would get shots in the future. It is suggested that the fears of senior citizens should be allayed by providing them with full and accurate information concerning the safety of shots for people of their age and with health problems they are likely to face. They should be informed of the relative risks of side effects and the illness the shot is designed to protect them against.
Participation of Senior Citizens in the Swine Flu Inoculation Program: An Analysis of Health Belief Model Variables in Preventive Health Behavior1 Evidence is presented of statistically significant and strong relationships between Swine Flu Inoculation status and nine variables in the reformulated Health Belief Model with 122 randomly selected subjects, primarily Black and Portuguese-American, who are active members of two Providence, Rhode Island senior centers. No statistically significant relationship was discovered between inoculation status and previously having had the flu. The variables which were found related are: Efficacy, Safety, Knowledge of Side Effects, Prior Flu Shot Status, Proportion of Friends and Relatives Who Got the Shot, Sex, Race, Future Plans for Flu Shots, and Future Plans for Other Inoculations. The data were obtained through personal interviews in the Spring of 1977. It is suggested that the results provide some basis for optimism for successful intervention designed to change the future preventive health behaviors of nonparticipants in the Swine Flu Inoculation Program. Many nonparticipants had fears and doubts about the effectiveness and safety of the shot and are amenable to suggestions from physicians about future inoculation participation. Full information should be provided to high-risk groups such as senior citizens about the relative risks of suffering serious side effects, the effectiveness and safety of the procedure for persons their age with the typical health problems of senior citizens, and the relative risks and dangers to them of contracting the illness against which the shot is designed to protect them.
N the May, 1977 Supplement to Medical Care on "Issues in Promoting Health" IBecker and associates present a review of re-
search on individual health-related behavior, using the Health Belief Model (Fig. 1) as an organizing rubric. The authors propose a revised model for explaining and predicting health-related behaviors (Fig. 2), call for less emphasis in future research on variables previously shown to be unrelated to health behaviors and for work with varying population groups and in different settings (Becker et al., 1977). The research reported on in this article represents a contribution to the further refinement of the Health Belief Model, providing evidence about the role of several variables in the re'When this article was written the author was with the Program in Gerontology of the Univ. of Rhode Island. Kingston. RI. The author gratefully acknowledges the support of the Univ. of R.I. University Research Committee and the College of Home Economics, the cooperation of the staff and members of the two cooperating senior centers and the student and senior citizen interviewers. The University's Academic Computer Center handled the data processing. 2 Assoc. Prof, of Sociology, Dept. of Sociology and Social Welfare. Rhode Island College, Providence 02908.
formulated model in the 1976 Swine Flu Inoculation Program. The respondents are in the high-risk senior citizen category, active members of two senior centers in Providence, Rhode Island providing services primarily for Black and Portuguese-American senior citizens. BACKGROUND
The existence of the Becker et al. (1977) article, recent and readily available to readers of this journal, precludes the necessity of a detailed and lengthy discussion of the Health Belief Model. The model contains variables which are felt to be intervenable and modifiable; has core dimensions based on a wellestablished body of psychological and behavioral theory, particularly that of Lewin (1944), and focuses on the valence of a goal, the subjective probability of its attainment, and motivation. Originally used to account for preventive health actions, the model has also been used to explain illness and sick-role 201
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
William R. Aho, PhD2
202
AHO
INDIVIDUAL PERCEPTIONS
MODIFYING FACTORS
Demographic variables (age, sex, race. ethnicity, etc.) Sociopsychological variables (personality, social class, peer and reference group pressure, etc.)
LIKELIHOOD OF ACTION
Perceived benefits of preventive action ^minus Perceived barriers to preventive action
r
Perceived Susceptibility to Disease " X " Perceived Seriousness (Severity) of Disease " X "
\
\ i
f
Perceived Threat Disease " X "
Likelihood of Taking Health Action
ii
Cues to Action Mass media campaigns Advice from others Reminder postcard from physician or dentist Illness of family member or friend Newspaper or magazine article
Fig. 1. Variables and relationships in the original Health Belief Model (Becker & Maiman, 1975).
READINESS TO UNDERTAKE RECOMMENDED COMPLIANCE BEHAVIOR
MODIFYING AND ENABLING FACTORS
COMPLIANT BEHAVIORS
Motivations Concern about (salience of) health matters in general Willingness to seek and accept medical direction Intention to comply Positive health activities Demographic (very young or old) Value of Illness Threat Reduction Subjective estimates of: Susceptibility or resusceptibility (incl. belief in diagnosis) Vulnerability to illness in general Extent of possible bodily harm* Extent of possible interference with social roles" Presence of (or past experience with) symptoms
Probability That Compliant Behavior Will Reduce the Threat
Structural (cost, duration, complexity, side effects, accessibility of regimen; need for new panerns of behavior) Attitudes (satisfaction with visit, physician, other staff, clinic procedures and facilities) Interaction (length, depth, continuity, mutuality of expectation, quality, and type of doctorpatient relationship; physician agreement with patient; feedback to patient) Enabling (prior experience with action, illness or regimen; source of advice and referral (incl. social pressure)
Likelihood of: Compliance with preventive health recommendations and prescribed regimens: e.g., screening, immunizations, prophylactic exams, drugs, diet, exercise, personal and work habits, follow-up tests, referrals and follow-up appointments, entering or continuing a treatment program.
Subjective estimates of: The proposed regimen's safety The proposed regimen's efficacy to prevent, delay, or cure (incl. "faith in doctors and medical care" and "chance of recovery")
'At motivating, but not inhibiting, levels.
Fig. 2. Summary hypothesized model for explaining and predicting individual health-related behaviors (Becker & Maiman, 1975).
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
\
SENIOR CITIZENS AND SWINE FLU
behaviors. The Becker et al. article points out that numerous other models contain similar variables.
203
(4) Demographic (race and sex); (5) Structural (knowledge of side effects); (6) Enabling (prior experience with the shot; proportion of friends and relatives who got the shot).
METHODOLOGY
FINDINGS
The respondents were asked questions designed to elicit attitudes, beliefs, and facts relevant to ten variables in six categories of the expanded Health Belief Model portrayed in Fig. 2 and first presented in an earlier article by Becker and Maiman (1975). The six categories, with a parenthetical statement about how they were operationalized, are: (1) Motivation (plans for future flu and other shots); (2) Value of Threat Reduction (prior experience with the flu); (3) Probability that Behavior Will Reduce the Threat (perceived efficacy and safety);
While no analysis of age differences is presented here, the model suggests that those who are very young or very old are more likely than others to participate in preventive health behavior. Our findings show that for our aged 60 and over respondents 54% received the Swine Flu shot. Nationally, for persons aged 50 and over, the figure is 44% (Center for Disease Control, 1977). Motivation
Future inoculation plans. — The Health Belief Model includes intention to comply as a motivational variable affecting the readiness to undertake compliance behavior. Respondents were asked if they planned to get other shots in the future to protect them against the flu. An examination of Table 1-A indicates that there is a statistically significant and strong relationship between inoculation status and future flu shot plans — nearly three-fourths of those who got the Swine Flu shot plan future flu shots, compared to just over 10% of those who did not get the shot. There is reason for some optimism about these findings because nearly one-fifth of those who didn't get the shot may get one in the future depending on advice from their physician, and 12% said outright that they do plan to get a future flu shot. Here, then, are nearly one-third of the nonparticipants comprising a "swing" group which may move
Table 1-A. Future Flu Shot Plans, by Inoculation Status (%).
Future Flu Shot Plans Yes No Depends on Drs. advice Depends (other) or Don't know Total % Total N
Inoculation Status Did Not Got Shot Get Shot 73 08 09
12 60 19
Total 46 31 14
10
10
10
100 (64)
101 (52)
101 (116)
Chi-Square = 52.26; 3 df; p = < .001 Cramer's V = .67.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Personal interviews of an average one-half hour duration were conducted during the months of March and April, 1977, with 122 senior citizens aged 60 and over who were active members of the two senior centers. An active member was operationally defined as one who had visited the meal site at the senior center at least two times within the two months preceding the sample selection. Over one-half of the interviews were conducted by other senior citizens, themselves active senior center members. While most of the interviews were conducted at the centers, over one-third were conducted in the respondents' homes to shorten and facilitate completion of the study. A 45-item pretested interview schedule was used for interviewing. Random selection was made by prenumbering the registration cards of the members and using a table of random numbers. Sixty-three percent of the sample chosen responded; this represents 28% of the total active population. One of the senior centers serves primarily a Black population; the other a PortugueseAmerican population. The numbers involved are too small to permit a refined and valid separate analysis within centers of those who did and did not receive a Swine Flu shot.
204
AHO
Table 2. Previously Had Flu, By Swine Flu Inoculation Status (%).
Table 1-B. Plans for Future Shots Other Than Flu Shots, by Inoculation Status (%)
Plans to get Other Future Shots — not Flu Yes No Depends on Drs. advice Depends (other) or Don't know
56 19 08
11 56 13
Total 35 36 10
17
_2_1_
18
100 (64)
101 (55)
99 (119)
Chi-Square = 28.06; 3 df; p = < .001; Cramer's V = .49.
over into the "Got Shot" category in some future campaign. Adding in even one-half of the category, "Depends, Other or Don't Know" increases the size of this swing group to a sizable 41%. These results are even more promising when it is remembered that they were obtained from interviews held after the mass media messages about the dangerous side effects of the Swine Flu shot. The bad news, of course, is that fewer than three-fourths of those who got the shot plan to get one in the future. They may be convinced otherwise if and when the opportunity and need for a future shot arise; or their loss might be offset by convincing nonparticipants to get a shot. Table 1-B presents data on future plans for shots for conditions other than the flu and indicates that the relationship between flu shot status and plans for future non-flu shots is also statistically significant and strong, with a pattern quite similar to that of the data in Table 1-A, with the main exception that there are more "Don't Know's" among the non-flu-shot group. Again, over one-fourth of the latter may get future shots, continuing the note of optimism from the previous Table. The fact that fewer of those who got the flu shot would get shots for other conditions (56%) may say something about their preference for a flu shot over other shots. Perhaps the fact of having had the shot and not suffered either any ill effects or the flu itself is responsible for this. Again, the size of the "Depends, Other or Don't Know" category suggests there is room for intervention and modification of behavior in future shot campaigns.
Previously Had Flu Yes No Total % Total N
Inoculation Status Did Not Got Shot Get Shot 49 52 101 (66)
34 66 100 (56)
Total 42 58 100 (122)
Chi-Square = 2.07; I df; p = < .15; Phi = .15.
Value of Illness Threat Reduction Previous flu status. — This category in the model includes the variable of past experience with symptoms of the condition in question. The respondents were asked whether or not they had ever had the flu. The results are reported in Table 2. There is no statistically significant relationship between prior flu status and Swine Flu inoculation status, although the percentages reveal some directional support for the implied direct relationship between getting the shot and having had the flu. Probability that Compliant Behavior Will Reduce the Threat Efficacy. — The research reviewed by Becker and associates (1977) reflects general support for the belief that the perceived efficacy and perceived safety of a preventive health behavior are positively related to participation. In a 1959 review of polio vaccinations Rosenstock et al. found support for these variables, although Leventhal et al. (1960) did not measure these benefits and barriers to action in their 1960 study of the Asian Flu epidemic. Perceptions about the effectiveness of treatment for heart disease has been found to be significantly related to the preventive health orientations of wives being interviewed about their roles in preventing heart disease in their husbands (Aho, 1977). Table 3 presents data on perceived efficacy and inoculation status. The respondents were asked how much the Swine Flu shot helped protect them from getting the Swine Flu. For those who did not get the shot, the question was phrased, "How much would the Swine Flu shot have helped protect you?" There is a very significant and strong relationship shown between perceived
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Total % Total N
Inoculation Status Did Not Got Shot Get Shot
SENIOR CITIZENS AND SWINE FLU
Table 3. Belief in Degree of Protection, by Inoculation Status (%).
Belief About Degree of Protection Completely protected Almost completely Somewhat A little, or not at all Don't know
49 32 II
Total
02 08
00 02 07 50 41
102
100
100
(66)
(54)
(120)
27 18 09 23 23
Chi-Square = 83.25; 4 cIf; p = < .001; V = .83.
Table 4. Belief About Likelihood of Getting the Swine Flu, by Inoculation Status (%).
Belief About Getting Swine Flu Very unlikely Unlikely Very likely, or likely Don't know Total % Total N
Inoculation Status Did Not Get Shot Got Shot
Total 40 34 11 15
38 45 09 08
43 20 14 23
100
100
100
(66)
(56)
(122)
Chi-Square = 11.84; 3 df; p = < .01; Cramer's V = .31.
efficacy and inoculation status, but among those who didn't get the shot there are many "Don't Knows" (41%). Another way of getting at beliefs about efficacy for those who got the shot, and discovering how well protected those who didn't get the shot feel, is to ask about the likelihood of getting the Swine Flu. As indicated from analysis of the data reported in Table 4, there is a significant relationship between inoculation status and how likely respondents feel they are to get the Swine Flu. The relationship is far from clear-cut or simple, however. It raises questions about why more of those who didn't get the shot say it is very unlikely that they will get the flu. This may, of course, be a reflection of a factual situation — it was already April and they had not had the flu. But that does not explain why fewer of those who actually got the shot felt it very unlikely they would get the flu. It may be that the latter are generally more cautious or less confident about their health status, and that
is in fact why they got the shot in the first place. It may also be a matter of cognitive dissonance for those who did not get the shot — they are displaying the conviction that all is going to be well with them, that they made the right decision in not getting the shot and choose not to think otherwise. The large percentage of "Don't Knows" for the nonparticipants (23%) coupled with the 14% who felt it very likely or likely they would get the Swine Flu, reveals a situation of doubt, uncertainty and possibly anxiety for quite a large proportion of the nonparticipants. This may help explain the previously reported finding that many of the nonparticipants may plan to get future shots. A companion item asked about the likelihood of other people in Rhode Island getting the Swine Flu if they did not have the shot. The findings here were consistent with those reported in Table 4, but even more statistically significant and stronger. Another interview item bearing on perceived efficacy dealt with beliefs about how serious the Swine Flu would be for the respondent if she or he got it this year. There was no statistical significance for this relationship, although more of those who got the shot felt it would not be at all serious, and nearly onehalf of those who did not reported they didn't know (see Table 5). Safety. — There is a significant and strong relationship between inoculation status and belief about the safety of the Swine Flu shot (see Table 6-A). Over one-fifth of those who didn't get the shot, however, believe it to be Table 5. Beliefs About How Serious Swine Flu Would be for Respondent, if He/She got Swine Flu This Year, by Inoculation Status (%). Belief About How Serious Very serious Somewhat serious A little serious Not at all serious Don't know Total % Total N
Inoculation Status Did Not - Got Shot Get Shot 15 09 14 38 23
11 07 14 21 46
Total 13 08 14 31 _34
99
99
100
(65)
(56)
(121)
Chi-Square = 8.18; 4df; p = < .10; Cramer's V = .26.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Total % Total N
Inoculation Status Did Not Get Shot Got Shot
205
206
AHO
Table 7. Sex and Race of Respondent, by Inoculation Status (%).
Table 6-A. Belief About Safety of Swine Flu Shot, by Inoculation Status (%).
Belief About Safety of Shot Safe Not safe Don't know Total % Total N
Inoculation Status Did Not Got Shot Get Shot
Total
85 06 09
22 51 27
56 26 17
100 (66)
100 (55)
99 (121)
Sex and Race Male Female White Black Other
Inoculation Status Did Not Got Shot Get Shot 38 62 44 42 14
21 79 47 51 02
Total 30 70 45 46 09
Table 6-B. Safety of Shot Last Fall, by Inoculation Status (%).
Did You Believe Shot Was Safe Last Fall? Yes No Don't know Total % Total N
Inoculation Status Did Not Got Shot Get Shot 92 06 02 100 (63)
Table 8. Heard of Someone Getting III From the Shot, by Inoculation Status (%). Total
26 52 22
62 27 II
100 (54)
100 (117)
Chi-Square = 53.82; 2 df; p = < .001; Cramer's V = .68.
Heard of Someone Getting III from Shot Yes No Total % Total N
Inoculation Status Did Not Got Shot Get Shot
Total
49 51
68 32
58 42
100 (65)
100 (56)
(121)
100
Chi-Square = 3.55; I df; p = < .06; Phi = .19.
safe, with another 27% saying they don't know. This suggests that there were other reasons for not participating in the flu shot program. Other data from this study reveal that the most frequently cited reasons given for not getting the shot were "doctor's advice," "don't think it's effective," and "just don't get shots." Respondents were also asked if they believed last Fall that the shot was safe. The results, reported in Table 6-B, are very similar to those reported in Table 6-A — a significant and strong relationship exists between shot status and believing last Fall that it was safe. There is some downward shift between Fall and Spring for both categories of respondents and this may help to explain why some who got the shot would not get it again. Demographic Variables Sex and race. — Among those who got the Swine Flu shot, males were overrepresented and females underrepresented, relative to their proportions in the sample (see Table 7). This is the reverse of the usual pattern found in studies of the utilization of health
services (Aday, 1972; McKinlay, 1972). Blacks and whites were both underrepresented, with the overrepresentation going to those of other races, in this case people from the Portuguese-American senior center who are of Cape Verdean background. While the customary level of statistical significance is not found for these relationships, the arguments put forth by Skipper et al. (1967) and by Labovitz (1968) for accepting levels higher than .05 justify accepting these levels as significant. Structural Variables Side effects. — One important structural variable mentioned by Becker et al. (1977), is side effects of the preventive health behavior. This was a controversial and central factor in the Swine Flu Inoculation Program, eventuating in the early cancellation of the program in December. The respondents were asked if they had heard of anyone getting ill from the shot, and the results are reported in Table 8.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Note: For Race, there are six missing observations; Significance is < .07 for Sex, < .06 for Race. Phi. = .18 for Sex; Cramer's V = .22 for Race.
Chi-Square = 47.39; 2 df; p = < .001; Cramer's V = .63.
SENIOR CITIZENS AND SWINE FLU
Table 10. Proportion of Friends and Relatives Who Got Shot, by Inoculation Status.
Proportion of Friends and Relatives All or Nearly all Several Just a few None Don't know Total % Total N
Inoculation Status Did Not Get Shot Got Shot
Total
29 41 17 05 09
04 25 38 25 09
17 34 26 14 09
101 (66)
101 (56)
100 (122)
Chi-Square = 27.58; 4 df; p = < .001; Cramer's V = .48.
shot. The results, reported in Table 10, indicate that there is a statistically significant and strong relationship between these variables. At the extremes, one-fourth of those who didn't get the shot, compared to just 5% of those who did, reported that none of their friends or relatives had received a shot, Enabling Factors Prior flu shots. — A final category of var- and nearly the reverse percentages reported iables contains prior experience with the pre- that all or nearly all had received the shot. ventive health action, and social pressure. The respondents were asked whether or not they had previously had the flu. The results, DISCUSSION reported in Table 9, are statistically signifiWhile this study was retrospective, the cant. Since 79% of the nonparticipants had interviews were conducted within four months not previously received any kind of flu shot, after the height of the Swine Flu Inoculation they appear to constitute a category of "hard- Program and some retrospective questions core" nonparticipants, given their age and the were asked to discover the time order of fact that they have lived through several events, attitudes or behaviors. In addition, epidemics of flu and presumably had oppor- several items relevant to a key model varitunities and cause to obtain shots in the past. able, Efficacy, were used to provide a check on the reliability of the items. The consisSocial pressure. — An indirect measure tency of the findings, arising as they do in a of social pressure to obtain the shot was ob- previously unexplored setting with different tained by asking about the proportion of and unique subjects and with a health probfriends and relatives who had received the lem of recent and wide-spread interest, enhances the model as a useful framework for continuing research and refinement and efTable 9. Previous Flu Shot Status, by forts at application by health care providers. Swine Flu Inoculation Status (%). The findings provide some basis for optimism that behaviors can be changed, because Inoculation Status of the rather sizable percentages of nonparPreviously Got Did Not ticipants who responded "Don't Know" or Flu Shots Got Shot Get Shot Total "It depends on my doctor's advice" to items Yes dealing with future preventive health be62 21 43 No 39 79 57 haviors. This suggests that efforts with phyTotal % 101 100 100 sicians as well as potential recipients of inocuTotal N (56) (121) (65) lations might prove useful. To allay some of Chi-Square = 18.14; I df; p = < .001; Cramer's V = .40.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
Using the rationale of Labovitz and Skipper and associates again, a significance level of .06 is viewed as acceptable, and the data offered as support for the contention that knowledge of bad side effects differentiates those who got the shot from those who did not. Unfortunately, we do not have information about when the respondents learned of the side effects. It may well have been after they had decided not to get a shot. If so, this may be evidence of selective perception on their part to validate their prior decision not to participate. Furthermore, nearly onehalf of those who got the shot had heard of someone who became ill from the shot, a finding which fits with the assumption that this knowledge came after the shots were given. The respondents were also asked to report the number of persons they heard became ill from the shot, but the results were not statistically significant (p = < .44).
207
208
AHO
the fears of senior citizens about future inoculations, accurate and full information should be provided to them about the safety of the procedure for people of their age with health problems they are likely to have, the relative risks of suffering serious side effects, and the relative risks and dangers to them of contracting the illness the shot is designed to protect them against.
This paper reports the results of a 1977 survey involving personal interviews with 122 randomly selected senior citizens in Providence, Rhode Island. The interviewees, primarily Black and Portuguese-American, were asked about their participation in the Swine Flu Inoculation Program and their plans for getting flu shots in the future. All of those interviewed were active members of two different senior centers in Providence. The results of the survey were analyzed by comparing answers of those who got the shot to those who did not. A number of factors felt to be related to having received a shot were used in the analysis to see if they were in fact related, and if they differentiated between those who did, and those who didn't, get the shot. These included: the effectiveness of the shot; its safety, its side effects, how many friends or relatives got the shot, sex, race, plans for future shots for flu or other conditions, and previously having had the flu. All except the last factor were found to be related in a statistically significant way with having had a swine flu shot. Specifically, those who got the shot, compared to those who didn't, (a) (b) (c) (d)
Plan to get a flu shot in the future Feel the shot is effective Feel the shot is safe Feel the shot does not have bad side effects (e) Had previously had a flu shot (f) Reported nearly all their friends or relatives had received the swine flu shot.
REFERENCES
Aday, L., & Eichorn, R. The utilization of health services: indices and correlates. A research bibliography. U.S. DHEW, Washington, DC, 1972. Aho, W. R. Relationship of wives' preventive health orientation to their beliefs about heart disease in husbands. Public Health Reports, 1977, 92, 65-71. Becker, M., Haefner,^)., Kasl, S., Kirscht, J., Maiman, L., & Rosenstock, I. Selected psychosocial models and correlates of individual health-related behaviors. Medical Care, 1977, 15, 27-46. Becker, M., & Maiman L. Sociobehavioral determinants of compliance with health and medical care recommendations. Medical Care, 1975, 13, 10-24. Center for Disease Control. National survey of public attitudes toward A/New Jersey/76 influenza vaccination: Report No. 8, Table 3, March 31, 1977, Atlanta, GA (Mimeo). Labovitz, S. Criteria for selecting a significance level: A note on the sacredness of .05. American Sociologist, 1968, 3, 220-222. Leventhal, H., Hochbaum, G., & Rosenstock, I. Epidemic impact on the general population in two cities. In The Impact of Asian Influenza on community life: A study in five cities, USPHS, Washington, DC 1960. Lewin, K., Dembo, T., Festinger, L., & Sears, P. S. Level of aspiration. In J. McV. Hunt,(Ed.), Ronald Press, NY, 1944. McKinlay, J. Some approaches and problems in the study of the use of services — an overview. Journal of Health and Social Behavior, 1972,13, 115-152. Rosenstock, I., Derryberry, M., & Carriger, B. Why people fail to seek poliomyelitis vaccination. PublicHealth Reports, 1959, 74, 98-103. Skipper, J. K. Jr., Guenther, A., & Nass, G. The sacredness of .05: A note concerning the uses of statistical levels of significance in social science. American Sociologist, 1967,2, 16-18.
Downloaded from http://geronj.oxfordjournals.org/ at University of Sussex on August 19, 2015
SUMMARY
The results provide some basis for optimism that the behavior of senior citizens relative to getting preventive shots can be changed, based on the proportions answering "Don't Know", or "It depends on my doctor's advice" when asked about whether they would get shots in the future. It is suggested that the fears of senior citizens should be allayed by providing them with full and accurate information concerning the safety of shots for people of their age and with health problems they are likely to face. They should be informed of the relative risks of side effects and the illness the shot is designed to protect them against.
