Background: Artificial intelligence (AI) technologies are increasingly integrated into medical practice, with AI-assisted diagnosis showing promise. However, patient acceptance of AI-assisted diagnosis, compared with human-only procedures, remains understudied, especially in the wake of generative AI advancements such as ChatGPT. Objective: This research examines patient preferences for doctors using AI assistance versus those relying solely on human expertise. It also studies demographic, social, and experiential factors influencing these preferences. Methods: We conducted a preregistered 4-group randomized survey experiment among a national sample representative of the US population on several demographic benchmarks (n=1762). Participants viewed identical doctor profiles, with varying AI usage descriptions: no AI mention (control, n=421), explicit nonuse (No AI, n=435), moderate use (Moderate AI, n=481), and extensive use (Extensive AI, n=425). Respondents reported their tendency to seek help, trust in the doctor as a person and a professional, knowledge of AI, frequency of using AI in their daily lives, demographics, and partisan identification. We analyzed the results with ordinary least squares regression (controlling for sociodemographic factors), mediation analysis, and moderation analysis. We also explored the moderating effect of past AI experiences on the tendency to seek help and trust in the doctor. Results: Mentioning that the doctor uses AI to assist in diagnosis consistently decreased trust and intention to seek help. Trust and intention to seek help (measured with a 5-point Likert scale and coded as 0‐1 with equal intervals in between) were highest when AI was explicitly absent (control group: mean 0.50; No AI group: mean 0.63) and lowest when AI was extensively used (Extensive AI group: mean 0.30; Moderate AI group: mean 0.34). A linear regression controlling for demographics suggested that the negative effect of AI assistance was significant with a large effect size (β=−.45, 95% CI −0.49 to −0.40, t₁₇₄₀=−20.81; P<.001). This pattern was consistent for trust in the doctor as a person (β=−.33, 95% CI −0.37 to −0.28, t₁₇₃₃=−14.41; P<.001) and as a professional (β=−.40, 95% CI −0.45 to −0.36, t₁₇₃₅=−18.54; P<.001). Results were consistent across age, gender, education, and partisanship, indicating a broad aversion to AI-assisted diagnosis. Moderation analyses suggested that the “AI trust gap” shrank as AI use frequency increased (interaction term: β=.09, 95% CI 0.04-0.13, t₁₇₃₅=4.06; P<.001) but expanded as self-reported knowledge increased (interaction term: β=−.04, 95% CI −0.08 to 0.00, t₁₇₃₆=−1.75; P=.08). Conclusions: Despite AI’s growing role in medicine, patients still prefer human-only expertise, regardless of partisanship and demographics, underscoring the need for strategies to build trust in AI technologies in health care.

Human papillomavirus (HPV) is one of the most common sexually transmitted infections (STIs). However, despite widespread under-vaccination amongst men and the importance of vaccinating both sexes to curb the spread of HPV, research has focused on promoting HPV vaccination predominantly amongst women. Therefore, the current study examines the effectiveness of different informational interventions in promoting vaccination intentions amongst heterosexual men. In a preregistered study of 583 unvaccinated adult men, we randomly assigned participants to one of four informational interventions aimed at promoting awareness of HPV risks and vaccine uptake: (1) risks to oneself (n = 145), (2) risks to their female partner (n = 144), (3) risks to oneself and their female partner (n = 153), and (4) general vaccine information (n = 153). Amongst participants reporting a sexual history (67%), intentions to get vaccinated significantly increased by 10.75 points on a 100-point scale (p < 0.01) after they received information about the risks of HPV for both themselves and their female partner, compared to receiving information about only their own HPV risk. These findings provide valuable guidance for public health officials and policymakers into the effectiveness of different messaging strategies in promoting HPV vaccination amongst adult male populations to increase vaccination rates.

The COVID-19 vaccination rate among children ages 5–11 is low in the U.S., with parental vaccine hesitancy being the primary cause. Current work suggests that safety and side effect concerns are the primary reasons for such vaccine hesitancy. This study explores whether this hesitancy can be mitigated with information interventions. Based on theories of health decision making and persuasion, we designed four information interventions with varying contents and lengths. We wrote two messages on vaccine safety (a detailed safety-long message and a succinct safety-short message), explaining the vaccine’s lower dosage, low rate of side effects, and the rigorous approval process. We also had two messages on protection effects (protect-family, protect-child). We combined these four messages with a vaccine-irrelevant control message and compared their effects on parental vaccine intention. We measured the parental vaccination intention using a 0–6 Likert scale question. Among the four intervention groups, we found that the short version of the safety message increased the average vaccination intention by over 1 point compared to the control arm, while the other three interventions failed to show significance. Specifically, these effects are particularly pronounced (around 2 points) for Republican parents who had a much lower initial intention to vaccinate their children. Our study highlights the importance of concise and to-the-point information rendering in promoting public health activities and therefore has important policy implications for raising vaccination intentions among parents, especially those leaning towards more conservative political affiliation.

The US is amongst the worst-performing countries at combating COVID-19. And within the US, red (Republican) states have significantly higher cases per capita than blue (Democratic) states. We use cross-country, state, and county-level data to provide a comprehensive analysis of economic, political, and psychological factors contributing to these differences. An inferior social safety net and American conservatism systematically correlate with the realization and effectiveness of non-pharmaceutical interventions such as social distancing and mask wearing from April to September. Economic inequality and weak social safety nets drive the economically vulnerable to work outside their homes, increasing mobility and reducing social distancing during early stages of the pandemic. Conservative ideology, anti-intellectualism, and evangelicalism drive people to politicize social distancing and mask wearing. Both factors predict a premature reopening in many states, and have a strong correlation with the drifting of COVID-19 epicenters to red states over the course of 2020. These factors have more explanatory power than partisanship in the first half year of the COVID-19 outbreak in the United States. However, from October on, closer to the presidential elections, partisanship is a better predictor of anti-COVID measures and explains well the regional variances of confirmed cases across states and counties. This indicates that partisanship is not the solely important factor in determining COVID-19 response and outcome, but its impact is likely to have been magnified as time goes by.