Respondent Characteristics
Table 2 shows respondent characteristics of the National Survey of Health Attitudes. We observed a fairly even age distribution, with most respondents between 18 and 69. Approximately 13% of respondents were 70 or older. Respondents were 52% female, 61% were married or living with a partner, and household income was fairly evenly distributed between respondents who reported a household income of less than $30,000 per year and those who reported an income of $100,000 per year or higher. Forty percent of respondents had up to a high school education; 28% completed some college, and 32% had completed college or higher. Approximately 70% of respondents had between two and five people living in the household.
Respondents were from the Northeast, Midwest, South and West, with the highest percentage of respondents (37%) from the South. Sixty percent of respondents were employed and 38% of respondents had spent 20 or more years living in their community. Fifty-four percent of respondents lived in a large city (500,000+ residents) and 84% rated their health as good, very good, or excellent. About 38% of respondents reported that poor health of another affected their life and 36% reported that they suffered from a chronic health condition. Approximately 25% reported a financial problem due to health. Nearly 50% of respondents reported that they cared for others at least once per week who were ailing.
Health Civic Engagement, Sense of Community, Valued Investment in Community Health, and Barriers to Action
Our first aim was to describe key patterns in health civic engagement. Table 3 shows the breakdown of respondent answers specific to each item that comprised health civic engagement, sense of community, valued investment in community health and participants report of barriers to acting to improve community health as listed in table 1. Within health civic engagement, it is notable that 51% of participants reported that they voted for or against a candidate for public office because of his/her position on issues such as education, public safety or community funding, which are upstream drivers of health but not directly health specific. On the other hand, 19% reported contributing time or money to an organization working to pass a health law or policy, and about 21% reported lobbying or advocating for a health-specific cause in their community. About 22% reported engagement through attendance at a civic meeting or working with neighbors to fix community problems.
Items within sense of community were broken into three subscales: membership, emotional connection and health. Within the membership subscale, 45% of respondents said they felt they could trust people in their community, 29% reported being able to recognize most of the members of their community, and about 27% reported that being a member of their community was part of their identity. Respondents’ emotional connection varied: 38% reported that it was very important to be part of their community and 52% said they expected to be part of the community for a long time. Nearly 50% felt hopeful about the future of their community and 41% said members of their community cared about each other. Health subscale questions asked about the community working together and having resources to improve its health. Thirty-eight percent reported that their community can work together to improve its health and only 19% said that they knew their neighbors would help them stay healthy. This is particularly notable given that Culture of Health action Framework is partially premised on advancing community action to improve health.
Finally, as far as valued investment in community health, 45% of respondents reported their top priorities were to make sure that the disadvantaged had equal opportunity to be healthy, decent housing was available for all who needed it, and healthy foods were available at affordable prices in communities. Just 23% of respondents thought there should be bike lanes, sidewalks for walking and public transportation so that people did not need to rely on cars. Respondents thought that barriers to taking action to invest in community health included people not thinking their involvement would really make a difference in changing the health of the community (23%) and that there were other issues (not specified) that people cared more about (19%).
Predictors of Health Civic Engagement
Our second aim was to explore the relationship between health civic engagement and individuals’ reports on sense of community, valued investment in community health, and perceived barriers to engagement, net of other factors known to be related to health-specific civic engagement. Results of this analysis are shown in the regression analyses in table 4. Model 1 shows the full set of items related to sense of community, and models 2, 3 and 4 show the sense of community subscales of membership, emotional connection and health. Each of the models show qualitatively similar results, independent of which sense of community subscale (or full scale) was used. We observed that even after adjusting for all other covariates, sense of community, regardless of whether it is conceptualized as the full set of questions, or as one of the three subscales (membership, emotional connection and health), has a strong statistically significant (p<.0001) positive association with health civic engagement. As sense of community increased, so did health civic engagement. Specifically, respondents who endorsed all items on the full sense of community scale endorsed an average of 22.8% (95%CI: 19.8 - 25.7%) more of the health civic engagement items than those who did not endorse any sense of community items. Those who endorsed all the items in the sense of community membership subscale endorsed an average of 21.1% (95% CI: 18.3, 23.8%) more of the health civic engagement items than those not endorsing any of the membership items. The analogous difference in health civic engagement endorsement for the emotional connection subscale was 17.5% (95% CI: 15.1, 20.0%) and 15.8% (95% CI: 13.3, 18.4%) for the health subscale.
Similarly, respondents who valued investment in community health were also more likely to participate civically. Those who endorsed all items around valuing investment in community health endorsed 14% (95% CI: 11.2, 16.8%) more of the health civic engagement items than those who did not endorse valuing investment in community health. This too was a strong, statistically significant (p<.0001) positive relationship. Endorsement of items related to barriers to action to improve community health, including whether respondents thought that external groups (community members, businesses, government) could influence community health, or whether respondents believed that there were barriers to taking action to invest in community health, did not show a statistically significant relationship with health civic engagement. We note that these results were qualitatively similar, in both direction and statistical significance, to those we generated using exploratory factor analyses to derive the construct scales (data not shown).
Covariates that were positively associated with health civic engagement include rating one’s health as excellent (compared with ‘good’), responding that the poor health of another has impacted one’s life, having financial problems due to health, and having helped (or helping) others who are ailing. We observed relationships with educational attainment, race/ethnicity, and city size of residence as well. As table 4 indicates, we found that non-Hispanic blacks have a significant positive association with health civic engagement, compared with non-Hispanic whites. Non-Hispanic Asians were less likely to engage civically. Compared with living in the South, living in the West was strongly positively associated with health civic engagement. Higher education is positively associated with higher levels of health civic engagement.
Figure 1 shows the fully adjusted associations of our main outcomes of health civic engagement with our main variables of interest: each of the sense of community subscales, value invested in community health, the impact of external groups on community health, and barriers to taking action to invest in community health. In sum, one’s sense of community to a large extent, and the value invested in community health to a lesser extent, are positively and significantly associated with health civic engagement. One’s perception of barriers to action to improve community health is not significantly associated with health civic engagement.
Discussion and public health implications
This analysis set out to explore patterns in health civic engagement, and especially the relationship with sense of community and mindset and expectations (identified as key drivers of making health a shared value in the Culture of Health Action Framework), net of other factors known to be related to health civic engagement.
As expected, respondents reporting a strong sense of community and those who prioritize community investments in health more frequently reported engaging civically around health, controlling for variations in measures of socioeconomic status, perceived barriers, and a host of other variables. This is consistent with the view, embodied in the Culture of Health action framework and grounded in the literature, that individuals who view health as an issue requiring collective or community investment and who identify strongly with their communities are more likely to engage civically specifically to address health issues (21, 44, 45).
Our findings also confirm that, on a population level, overall health-related sense of community and health civic engagement are generally limited. Given that building a Culture of Health in America requires some amount of community action, the fact that less than forty percent of respondents felt their community could work together to improve health suggests room for further work and action. Action could include efforts that bring communities together and/or initiatives and programs to promote residents to be more involved with decisions surrounding their communities’ futures. More institutional supports could be imagined with improved and increased civics included in primary education. In addition, with one exception (voting on an issue related to education, public safety or another community issue), the health civic engagement items did not exceed half of the respondents, and the case of advocating specifically for “health policy” was reported positively by just one-fifth of respondents. This may suggest that an entry point to activate health-specific action is through these health-related activities on upstream drivers such as education or public safety.
The findings also have potentially interesting implications for how we understand the barriers to health civic engagement. We did not find clear evidence that those perceiving barriers to act to improve community health (e.g., not knowing how to get involved, not believing that involvement makes a difference) were less likely to engage civically around health, suggesting that addressing these attitudinal barriers may not be enough to activate health civic engagement. The limited literature in this space to support or contradict our finding highlights the question of how to inspire and motivate health civic engagement.
We did find evidence that one’s own poor health, was associated with lower levels of civic engagement around health. This aligns with a number of earlier findings (18)from the U.S. and elsewhere suggesting that those in poor health or with disabilities are less likely to vote or engage civically in other ways. In one study, people with disabilities were 20 percentage points less likely to vote than people without disabilities of similar demographics (9). In another longitudinal study in Ireland and Britain, individuals with poor self-rated health were significantly less likely to report voting in past general elections (46). While the direction of the causality is not always clear, it might point to the need to broaden access to mail-in ballots and other measures to lower the burdens of political participation. It also suggests that activities to expand health civic engagement may need to address the unique challenges faced by those with chronic health conditions, including, for example, difficulty with mobility and regular participation in activities which require travel. Respondents who reported that poor health of a friend of family member impacts their life, those who reported financial problems due to health, and those who reported helping others who were ailing were all more likely to be civically engaged in health-related activity. This suggests that some degree of experiencing a health-related challenge or barrier may motivate individuals to take action. Without being able to assess the direction of causality, this does suggest the possibility that the burdens of caring for others may be a catalyst for health civic engagement, or perhaps that those who help others in need are also more inclined towards involvement in the community.
Finally, consistent with many other studies of civic engagement and political participation, we found that those with higher levels of education and higher income were more likely to engage civically around health. In light of previous research cited above linking civic engagement with individual health (16, 47, 48)benefits, such findings are an important reminder that the health and well-being benefits of civic engagement do not accrue equitably to all. First, as efforts are underway to advance health equity, these findings further underscore the challenges in engaging Americans in health civic engagement. Second, the health benefits of civic engagement are also not experienced equally. We did observe a notable finding worthy of further investigation: African American/black respondents were more likely to engage on health issues civically than white Americans, while Asian respondents were less likely than all other groups, adjusting for all other factors. There are many potential explanations for this disparity (data not shown), given that in the broader survey, we also observe the same racial/ethnic differences in willingness to invest in community health (only 17% of African Americans/blacks would not invest in any community health priorities vs. 33% of white Americans). It could be that African American/blacks believe there is a greater need to engage civically for health, or there may be a stronger view that this approach to health improvement will be more effective.
This study offers important new insights about health civic engagement and the role of community health attitudes and sense of community. Still, the findings are subject to a few limitations. First, the cross-sectional nature of the data used in this analysis means we cannot assume the observed relationships between civic engagement, sense of community and value placed in community health investments are causal. Perhaps more importantly, though, the study was unable to assess the extent to which individual civic engagement impacts the health of communities through fostering changes in socio-economic drivers of health, or the possibility that civic engagement may promote policies or other collective actions that compromise the health of some groups (e.g., lobby for weakening of environmental protections). This also reflects a broader tendency in the current research (e.g., 43, 44) to focus on health impacts on those who are engaging in civic activities (19).