2023 National Survey of Health Attitudes: Description and Top-Line Summary Data

Since 2013, the Robert Wood Johnson Foundation (RWJF) has led a groundbreaking effort to advance a Culture of Health that is rooted in equity, with a “future where everyone has a fair and just opportunity to achieve their fullest health and wellbeing.” Together with RAND, RWJF initially developed an approach to operationalizing and measuring that vision (Chandra et al., 2016; Plough et al., 2015). Today, RWJF's efforts have built on that work, but now with a stronger focus on approaches to transforming institutions and systems by addressing structural racism and promoting the narrative that health is a right, not a privilege.

RAND researchers have been working with RWJF to capture public perspective on these topics, designing and fielding the National Survey of Health Attitudes (NSCH) to provide insight and perspective on how people in the United States think about, value, and prioritize health and consider issues of health equity. The survey was first fielded in 2015, and an updated version was fielded in 2018. This study provides a brief overview of the 2023 survey development and content, and then a top-line summary of descriptive statistics. The study complements the overview of the 2015 survey described in the RAND report Development of the Robert Wood Johnson Foundation National Survey of Health Attitudes (Carman et al., 2016), and its subsequent topline 2018 Survey of National Health Attitudes: Description and Top-Line Summary (Carman et al., 2019) and is organized similarly for consistency. A companion set of longitudinal surveys during the COVID-19 pandemic was fielded between 2020 and 2021 and is further described in four top-line reports, COVID-19 and the Experiences of Populations at Greater Risk (Carman et al., 2020a; Carman et al., 2020b; Carman et al., 2021a; Carman et al., 2021b).

The questions in the 2023 survey uniquely capture aspects of American mindset about health, health equity, structural racism, and wellbeing in ways that are not present in other surveys. This version of the NSCH can be viewed in three main sections: (1) individual health experiences, perspectives, and knowledge (making health a shared value); (2) health equity perspectives; and (3) community wellbeing, including climate views and barriers to community engagement. Insights from the surveys referenced above, including this one, have established a baseline and set of cross-sectional pulse checks on where the American public is regarding their recognition of social determinants of health, their understanding of health inequities including structural racism, their willingness to address those inequities and their indication of who in society should be responsible for solving health inequities.

As with prior NSCH instruments, we designed this survey to measure constructs that could not be assessed using existing data. In some cases, existing data were out of date or collected only in small samples that were not nationally representative. Where possible, we used questions drawn from available survey instruments including prior versions of the NSCH or the aforementioned COVID-19 surveys. However, in some cases, we had to modify existing questions or develop new questions because we needed to delve further into issues of race and health equity as well as community wellbeing, where survey questions did not exist.

Using methods comparable to those used in the 2018 survey, we collected data via the RAND American Life Panel (ALP) and the KnowledgePanel (administered by Ipsos). Both panels are nationally representative internet panels whose members are recruited via probability-based sampling methods. Both provide computers and internet connections for respondents who do not have them at the time of panel recruitment. Both compensate respondents for their participation. Both panels collect demographic information about respondents separately and provide this information with each data set. We fielded the same survey instrument in both panels. The resulting survey contains 37 questions, some with subquestions or multiple parts. The median length for survey completion was between 19 and 20 minutes.

The two survey efforts combined resulted in a final total sample of 5,620 completed surveys: 1,570 from the ALP and 4,050 from the KnowledgePanel. Each panel brings distinct benefits. With the ALP, we can link responses to a very rich set of background variables collected through other surveys, as well as to responses collected in the 2015 and 2018 National Survey of Health Attitudes. On the other hand, the KnowledgePanel provides a significantly larger sample size. We used a raking algorithm to create weights to match the distribution of characteristics in our sample as closely as possible to the distribution of characteristics of the population from the 2022 Current Population Survey (U.S. Census Bureau, 2022). We weighted each sample to the CPS 2022 first. Then, we combined the two samples and matched the distribution of characteristics of the pooled sample to the distribution of the CPS. In other words, our weighting procedure treated observations from the two panels as equivalent. We combined the results from the two panels and calculated weights to make the combined panel representative of the population. Because the overall sample size of our survey is large, 5,620 respondents, the margin of error (MOE) for the full sample ranges from 0.32 percent to 3.33 percent, for proportions near 0 percent and 50 percent, respectively. For the ALP subsample, the MOE ranges from 0.51 percent to 5.01 percent. For the KnowledgePanel subsample, the MOE ranges from 0.38 percent to 4.07 percent. As were done with prior NSCH, we have not adjusted the MOEs to reflect design effects.