This study analysed for socioeconomic differences in acceptance of preventive home visits and for modifying effects of municipality invitational procedure for the preventive home visits. In summary, this study showed that high SES was associated with higher acceptance rate of preventive home visits. The findings further suggested that the more proactive invitational procedure chosen by municipalities might reduce the negative effect of low SES on the acceptance rate.
A major finding was that persons in low SES were less likely to accept the preventive home visits compared with persons in high SES. This is in line with a recent British review, which concluded that in public tax-financed health systems like Denmark there is a general tendency that preventive services are more often used by people in high SES [17
]. Likewise studies on cancer screenings have shown a SES disparity in participating in such preventive strategies in countries with universal insurance coverage [3
]. It has also been shown that a larger proportion of receivers of the lowest social benefits have refrained from buying medicine and visiting dentists than people who had resources from other sources [18
The findings of social inequality in accepting preventive home visits may have several explanations. Maybe the older person’s earlier experiences with the social and health care system had an influence: e.g. earlier contact with the municipalities, expectations on possibilities for help and knowledge about what the municipality can offer [19
]. Here the user's education, social status and communication skills may play a critical role. It is also possible that the preventive home visits are organized in a way, which is more focused towards the middle class. This supports findings from other health promotion studies, where preventive education was more easily understood by the middle class than by more socially disadvantaged groups [20
]. Maybe the organization of the preventive home visits should be more targeted to different groups in order to be attractive to all social groups. A first step could be to educate the preventive home visitors about social inequality in health and functional ability. A second step could be to interview older people in different social groups about how to make the preventive home visits the best way. This might give new ideas on how to be better at aiming the preventive home visits to older people in different social groups [20
A second major finding was that the association between SES and acceptance of preventive home visits was attenuated by invitational procedures to the preventive home visits. This means that a larger proportion of older people in low SES would accept the preventive home visits, if the invitation were a letter with a proposed date. This is in line with the careful attention on invitational procedures as an important strategy to increase participation rates among disadvantaged populations in community health promotion and community-based research [22
Even though this study was set up in a specific setting with a government-funded home visit program, our findings were, thus, much in line with studies on other preventive interventions. Since increasing participation rates has been shown to be essential in any preventive interventions, we do think that our findings that more proactive invitational procedure might decrease a social inequality can be translated to other preventive programs, and thus be useful in more general terms with regard to prevention.
We did consider a range of potential confounders of the association between SES and acceptance of preventive home visits, especially mobility disability, which is a good measure of health status of this age group, and psychosocial factors. These variables might be situated in the causal sequence between SES and acceptance of preventive home visits and might therefore be mediators. Adjusting for mediators could lead to an underestimation of the studied effect. Therefore, we chose not to include these factors in our final analyses [27
]. We are aware that these factors might explain some of the combined effect of SES and invitational procedure on acceptance of preventive home visits. Yet, it was not an aim of this study to identify explanatory mechanisms.
One limitation of our study was the non-validated question of the outcome measure: “Have you accepted the invitation from the municipality to receive a preventive home visit?”. There was no means to test reliability of this answer; therefore it is possible that some older people may have misunderstood the invitation as an invitation to other forms of home visits such as home help services, thus giving a risk for misclassification.
A possible impact on study results by participants excluded due to missing values (n
161) is also a limitation. They were almost identical with the included 1,023 regarding distributions of financial assets, and acceptance of preventive home visits, but were significantly different as regards invitational procedures. Only 47% of the excluded received ‘letter with date’, compared to 63% of the included (chi square p
0.034). This was because more people of those excluded due to missing values on acceptance of home visits had ‘telephone call or letter without date’ as the invitational procedures. It is reasonable to assume that older people who received those invitational procedures, where they did not need to do further action to decline, had a greater probability of forgetting whether they had the offer. If they had declined more often than the included people, which is more likely, the odds ratios of invitational procedures on acceptance of preventive home visits might be larger than estimated in this study.
Strengths of this study were both the robustness of the study population and the possibility to use comprehensive registry data of financial assets as a measure of SES. Self-reported income is most often used in studies investigating social differences in health and health behaviours [28
], but it is also pointed out that self-reported individual-level SES had poor agreement with aggregate-level from census data [31
]. In contrast, under the civil registration system in Denmark, all aspects of private finances connected with financial organizations are captured and furnish vital statistic to be available for research purposes.