The Health Information National Trends Survey (HINTS) is a nationally representative survey supported by the National Cancer Institute that aims to describe cancer-related knowledge, beliefs, attitudes, and behaviors of adults in the US. HINTS uses a probability sample of U.S. telephone numbers to reach a cross-section of the adult, non-institutionalized, and civilian U.S. population. Data were collected using a computer-assisted telephone interview system administered in English or Spanish. Response rates were calculated using the Response Rate 2 (RR2) formula of the American Association of Public Opinion Research (Nelson et al. 2004). We utilized data from the 2012 iteration of the HINTS 4 Cycle 2. There was a total of 3,630 participants in the 2012 iteration of the HINTS with an overall response rate of 40%. Racial and ethnic minorities were oversampled to increase the precision of estimates for minority sub-populations. Additional details on the HINTS methodology as well as access to the complete HINTS datasets can be found on hints.cancer.gov as well as previous literature and reports (Blake et al. 2016; Westat 2015).
For this study, we used the questions “How much do you worry that indoor air pollution will harm your health?” and “How much do you worry that outdoor air pollution will harm your health?”. Participants were able to respond: “not at all”, “a little”, “somewhat”, or “a lot” to each question. For indoor and outdoor air pollution, responses to “a little” and “somewhat” were combined to create three categories of responses: “a lot”, “some or a little”, and “not at all”.
Sociodemographic variables included sex (male and female), age (18–34, 35–49, 50–64, and ≥ 65), race/ethnicity (Hispanic, non-Hispanic white, non-Hispanic black, and Asian, Pacific Islander, other), whether someone was born in the United States, education level (high school diploma or less, some college or post-high school vocational training, bachelor’s degree or more), annual household income (≤$20,000, $20,000–34,999, $35,000–49,000, $50,000–74,000, and ≥$75,000), smoking status (never smoker, former smoker, and current smoker), and whether respondents resided in a rural or urban county. For our analysis, rural and urban residence was determined based on Rural Urban Continuum Codes in which categories 1 through 3 (counties with 250,000 residents or more) were considered “urban” and categories 4 through 9 (counties with less than 250,000 residents) were considered “rural” (Zahnd et al. 2010). In addition to sociodemographic variables, we also examined cancer-related variables including family history of cancer (No, Not Sure, Yes), previous diagnosis of lung disease, which included chronic lung disease, asthma, emphysema, and/or chronic bronchitis (No, Yes), and perceived likelihood of getting cancer (Very Unlikely or Unlikely, Neither Likely or Likely, and Very Likely or Likely).
Bivariate analyses of sociodemographic variables and cancer-related characteristic variables by worry of harm from IAP and OAP were conducted using the Wald chi-squared test. A p-value of less than or equal to 0.05 was used to determine statistical significance. Multinomial logistic regression was used to calculate odds ratios (OR) and 95 % confidence intervals (CI) for the association between sociodemographic and cancer-related variables and worry of harm from IAP and OAP (separately). Participants with any missing values were removed from the logistic regression analysis. All models were adjusted for demographic variables including gender, age, race/ethnicity, education, and rural-urban residence. SAS (version 9.4; Cary, NC) was used to conduct all statistical analyses.