Social and experiential factors linked to concern about climate-related health harm in the United States

Abstract

Introduction

Public concern that climate change will harm personal health influences individual protective behaviors and shapes support for climate-related health policies. We analyzed a national survey to identify factors associated with expressed concern that climate change will harm one's health among U.S. adults.

Methods

We analyzed cross-sectional data from the Health Information National Trends Survey (HINTS 7; N = 6888 adults). Analyses examined correlates of perceptions that climate change will harm personal health (any concern vs no concern/unsure).

Results

Overall, 65% of U.S. adults expressed concern that climate change will harm their health. After adjustment, high trust in scientific information was associated with a 7-percentage point higher probability of concern. Compared with liberals, moderates and conservatives had 11- and 28-percentage point lower probabilities of expressing concern, respectively. Experiencing extreme weather in one's neighborhood (+21 points) and worry about outdoor air quality (+19 points) were also strongly associated with concern.

Conclusion

Concern about the health impacts of climate change is patterned primarily by trust in science, political viewpoint, and environmental attitudes and experiences rather than demographic characteristics. Public health communication should reinforce scientific credibility, engage audiences across political perspectives, and link climate health messages to locally relevant environmental conditions.

Key points.

• High trust in scientific information was associated with a 7-percentage point higher probability of concern about the health impacts of climate change.

• Compared with liberals, moderates and conservatives had 11- and 28-percentage point lower probabilities of expressing concern about the health impacts of climate change, respectively.

• Experiencing extreme weather in one's neighborhood (+21 points) and worry about outdoor air quality (+19 points) were also strongly associated with concern about the health impacts of climate change.

Introduction

Public concern about climate change is a critical determinant of both individual health behavior and public support for environmental and health policies.^1-4 Recent shifts in federal climate regulation have renewed attention to how the public interprets climate-related health risks, underscoring the need to understand the factors shaping these perceptions.^5-8 Although prior research shows that social and political attitudes shape climate change risk perception, little is known about how these factors specifically influence perceived health threats.^9-13

Environmental psychology and health behavior research offer robust theoretical frameworks that help explain how individuals perceive and respond to climate-related health risks. For example, the Climate Change Risk Perception Model (CCRPM) identifies cognitive (eg, trust in information), experiential (eg, personal exposure), and sociocultural (eg, political identity) factors as key determinants of perceived climate risks.^2,14 Cultural cognition theory further posits that individuals interpret risk information in ways that align with their group values (eg, political identity).¹⁵ In parallel, health behavior models such as the Health Belief Model (HBM) and Protection Motivation Theory (PMT) emphasize the role of perceived susceptibility, severity, and trust in authorities in shaping health behaviors.^16,17 These frameworks suggest that trust in science can amplify perceived risk, while personal experience with environmental hazards may increase emotional salience and perceived vulnerability.^18-22 By integrating these perspectives, our study examines how trust in science, political identity, and environmental experience interact to shape concern about climate-related health risks.

We analyzed nationally representative survey data from U.S. adults to examine how trust in science, political orientation, perceived impacts of extreme weather, and worry about outdoor air quality were associated with concern that climate change will harm one's health. Understanding whether these factors shape climate attitudes can help public health agencies and policymakers tailor communication strategies and maintain public engagement during periods of evolving climate policy.

Methods

We analyzed data from the Health Information National Trends Survey (HINTS) 7, a nationally representative survey of U.S. adults aged ≥18 years administered by the National Cancer Institute between March and September 2024.²³ HINTS employs a stratified probability-sampling design to capture the civilian, noninstitutionalized adult population. Of 7278 respondents, 390 (5.4%) were excluded due to missing outcome information, yielding an analytic sample of 6888 adults. Power calculations (α = 0.05; two-sided) indicated >80% power to detect an odds ratio (OR) of 1.30 (or its reciprocal, 0.77) for exposures with ≥20% prevalence, assuming a moderate outcome prevalence (10%-30%) and the covariate structure of the final regression models. As the dataset is publicly available and de-identified, this study did not constitute human subjects research.

The primary outcome was self-reported concern that climate change would harm respondents’ health (“Not at all,” “A little,” “Some,” “A lot”). For the primary analysis, this measure was dichotomized into No Concern/Unsure vs Any Concern, consistent with prior studies that distinguish between minimal and elevated concern levels.^24-26 This approach facilitates interpretability and aligns with common practice in climate risk perception research. Sensitivity analyses using ordinal logistic regression produced estimates of similar magnitude and direction (Appendix Table SA1), supporting the robustness of the dichotomized model. Covariates included age, sex, marital status, census region, metropolitan residence, race and ethnicity, education, employment, income adequacy, health insurance status, self-rated health, health information seeking behavior, daily social media use, trust in science information, political viewpoint, perceived extreme weather neighborhood impact, and worry about outdoor air quality.

We handled missing covariate data with multiple imputation by chained equations (MICE), generating 50 imputed datasets with a fixed seed and confirming adequate model fit through standard diagnostic checks. Sensitivity analyses using listwise deletion produced estimates similar in magnitude and statistical significance (Appendix Table SA2). We produced survey-weighted descriptive statistics for the overall sample and stratified by the binary climate concern outcome. To estimate adjusted associations, we fit a survey-weighted logistic regression model of the dichotomized outcome and calculated average marginal effects (AMEs) to express results as percentage-point differences in predicted probability. All analyses were derived using survey design variables and pooling 50 multiply imputed datasets using Rubin's rules.

Results

Among the 6888 adults in the analytic sample, 65.2% reported concern that climate change will harm their health, while 34.8% reported no concern or were unsure (Table 1). Substantial attitudinal and experiential differences were evident between these groups. For example, 88.7% of concerned adults reported high trust in scientific information, compared with 78.8% of those with no concern or uncertainty. Liberal political orientation was more common among those expressing concern (38.8%) than among those not concerned (14.1%), whereas conservative orientation showed the opposite pattern (21.8% vs 51.6%). Experiential factors showed similarly large differences: 81.4% of concerned adults reported that their neighborhood had been affected by extreme weather compared with 52.7% among those not concerned, and 61.3% of concerned adults reported worry about outdoor air quality vs 30.0% among those not concerned.

Table 1.

Weighted descriptive characteristics of U.S. adults overall and by perceived climate-related health harm, multiple imputation and survey-weighted estimates: HINTS 7 (2024), N = 6888.

		How much do you think climate change will harm your health?
Variable	Overall %	None/Unsure %	Concern %
Outcome		34.8	65.2
Trust in science information about cancer
Low	14.7	21.2	11.3
High	85.3	78.8	88.7
Political viewpoint
Liberal	30.2	14.1	38.8
Moderate	37.6	34.4	39.4
Conservative	32.2	51.6	21.8
Extreme weather affected neighborhood
Not Affected	28.6	47.3	18.6
Affected	71.4	52.7	81.4
Worry outdoor air quality harm health
None/Little	49.6	70.0	38.7
Some/A lot	50.4	30.0	61.3
Age group
18-34	25.0	19.9	27.8
35-49	26.0	23.9	27.1
50-64	26.4	28.7	25.2
65+	22.5	27.5	19.9
Sex
Male	49.2	47.1	50.3
Female	50.8	52.9	49.7
Marital status
Married/Cohabiting	56.4	59.3	54.9
Formerly married	13.0	15.7	11.6
Never married	30.5	25.0	33.5
Census region
Northeast	17.4	16.8	17.7
Midwest	20.9	20.4	21.2
South	38.7	43.2	36.3
West	23.0	19.6	24.9
Metropolitan status
Nonmetro	13.1	15.4	11.8
Metro	86.9	84.6	88.2
Race and ethnicity
Non-Hispanic White	60.2	66.2	57.0
Non-Hispanic Black	11.2	10.4	11.7
Hispanic	17.8	14.9	19.3
Non-Hispanic Asian	5.5	4.1	6.3
Non-Hispanic Other	5.2	4.5	5.6
Education
Not college graduate	65.7	67.7	64.6
College graduate or higher	34.3	32.3	35.4
Employment status
Not employed	34.7	38.5	32.6
Employed	65.3	61.5	67.4
Perceived income adequacy
Finding it very/difficult on present income	27.0	24.6	28.3
Getting by on present income	37.7	37.7	37.8
Living comfortably on present income	35.3	37.8	34.0
Health insurance
No	8.5	8.1	8.7
Yes	91.5	91.9	91.3
General health
Fair/Poor	17.5	15.0	18.8
Excellent/Very Good/Good	82.5	85.0	81.2
Ever looked for cancer information
No	48.8	53.9	46.1
Yes	51.2	46.1	53.9
Daily social media use
< Daily	34.9	38.8	32.8
Daily	65.1	61.2	67.2

Values represent weighted percentages derived using survey design variables and pooling 50 multiply imputed datasets using Rubin's rules. The outcome variable “How much do you think climate change will harm your health?” was dichotomized as None/Unsure vs Concern.

After multivariable adjustment, several attitudinal and experiential variables were strongly associated with climate-related health concern (Table 2; Appendix Table SA3). High trust in scientific information was associated with a 7-percentage-point higher probability of concern (AME =+7.2%; 95% CI, 2.3%-12.0%). Compared with liberals, adults identifying as moderate had an 11-percentage-point lower probability of concern (AME = −10.7%; 95% CI, −15.1% to −6.4%), and conservatives had a 28-percentage-point lower probability (AME = −27.5%; 95% CI, −32.0% to −23.0%). Environmental experiences were also strongly associated with concern. Adults who reported that their neighborhood had been affected by extreme weather had a 21-percentage-point higher probability of concern (AME = +21.0%; 95% CI, 17.0%-24.9%). Worry about outdoor air quality was associated with a 19-percentage-point higher probability (AME = +18.7%; 95% CI, 15.2%-22.3%).

Table 2.

Multivariable adjusted average marginal effects (AMEs) from survey-weighted logistic regression predicting concern about climate-related health harm: HINTS 7 (2024), N = 6888.

Variable	AME	P-value	Lower 95% CI	Upper 95% CI
Trust in science information about cancer
Low
High	7.2%	0.00	2.3%	12.0%
Political viewpoint
Liberal
Moderate	−10.7%	0.00	−15.1%	−6.4%
Conservative	−27.5%	0.00	−32.0%	−23.0%
Extreme weather affected neighborhood
Not affected
Affected	21.0%	0.00	17.0%	24.9%
Worry outdoor air quality harm health
None/Little
Some/A lot	18.7%	0.00	15.2%	22.3%

Estimates reflect average marginal effects (AMEs) on the probability scale from a multivariable adjusted, multiply imputed, survey-weighted logistic regression model of concern that climate change will harm one's health, measured as no concern/unsure vs any concern. All estimates incorporate sampling weights, strata, primary sampling units, and pooling 50 imputations using Rubin's rule.

Discussion

In this nationally representative sample of U.S. adults, concern that climate change will harm personal health was common but strongly patterned by attitudinal and experiential factors. Trust in scientific expertise, political viewpoint, perceived impact of recent weather events, and worry about outdoor air quality were each independently associated with substantially different probabilities of expressing concern, whereas most demographic characteristics showed no significant adjusted associations. These findings suggest that climate-related health concern in the United States is shaped less by demographic position and more by how individuals interpret scientific information, their political identities, and their lived environmental experiences.

Trust in science emerged as a key predictor: adults reporting high trust in scientific information had a notably higher probability of expressing concern about the health impacts of climate change.^25,26 In an environment in which scientific evidence is often contested, this finding highlights the importance of visible, credible, and consistent communication from scientific and public health institutions.^27,28 Strategies that reinforce transparency and clearly link scientific assessments to practical health guidance may help sustain public engagement with climate-related risks.²⁹

Political viewpoint was also strongly associated with health concern about climate change. Compared with liberal adults, those identifying as moderate or conservative were substantially less likely to express concern about climate-related health harms. This ideological gradient mirrors broader patterns in climate attitudes and suggests that uniform messaging may be insufficient.^9,30 Communication strategies framed around shared values such as community safety, economic stability, disaster preparedness, or health system resilience may resonate more effectively across political perspectives and broaden public engagement.³¹

Experiential factors played a similarly important role. Adults who reported that their neighborhood had been affected by extreme weather, or who expressed worry about outdoor air quality, had markedly higher probabilities of reporting concern. These findings suggest that tangible environmental exposures help translate abstract climate risks into personally relevant health threats. Public health messaging that connects local environmental conditions such as heat advisories, wildfire smoke, flooding, or air-quality alerts to broader climate trends may therefore be especially effective in motivating awareness and protective behavior.^32-34

The limited associations observed for most demographic characteristics indicate that demographic targeting alone may be insufficient to build climate-health awareness. Although structural inequities remain central to climate vulnerability, our findings underscore the value of strategies that address trust, ideology, and lived experience as more proximate drivers of concern.³

Our findings reinforce and extend several established theoretical frameworks in climate risk perception and health behavior. Consistent with CCRPM, we found that cognitive (trust in science), sociocultural (political identity), and experiential (extreme weather exposure) factors were significantly associated with concern about climate-related health risks.^2,14 The strong association between political identity and concern supports cultural cognition theory's assertion that individuals align their risk perceptions with their ideological group's values.¹⁵ Similarly, the positive relationship between trust in science and concern aligns with prior research showing that trust in scientific institutions enhances acceptance of climate risks.^20,25,26 Health behavior models such as the HBM and PMT emphasize that perceived susceptibility and severity, shaped by personal experience and trust in credible sources, are central to risk perception and action.^16,17 Our findings suggest that individuals who trust scientific sources and have experienced environmental hazards are more likely to perceive climate change as a serious health threat, consistent with these models.¹³ Together, these theoretical perspectives provide a coherent framework for understanding how trust, ideology, and experience shape climate health risk perceptions. By integrating insights across disciplines, our study contributes to a more nuanced understanding of the psychological and social mechanisms underlying public concern about climate-related health risks.

This study has several limitations. Its cross-sectional design precludes causal inference, and reliance on self-reported measures may introduce recall or social desirability bias. Although multiple imputation reduced bias from missing data, unmeasured confounding cannot be ruled out. Nonetheless, results were consistent across binary, ordinal, and listwise deletion models, supporting the robustness of our findings. Additionally, potential measurement error in self-reported attitudes and residual confounding may persist despite adjustments. The cross-sectional design also limits our ability to assess temporal changes in perceptions.

In summary, public concern about the health impacts of climate change is shaped primarily by trust in science, political viewpoint, and environmental experiences. Communication strategies that strengthen scientific credibility, engage audiences across political perspectives, and connect climate-related health guidance to locally salient environmental signals may be especially important as climate policies continue to evolve.

Supplementary material

Supplementary material is available at Health Affairs Scholar online.

Funding

Research reported in this publication was supported by the National Institute on Minority Health and Health Disparities under Award Numbers R61MD019939 and R01MD016426. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflicts of interest

Please see ICMJE form(s) for author conflicts of interest. These have been provided as supplementary materials.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data are publicly available from the National Cancer Institute https://hints.cancer.gov/.