Search

Abstract Despite the promising capacity of large language model (LLM)-powered chatbots to diagnose diseases, they have not been tested for obsessive-compulsive disorder (OCD). We assessed the diagnostic accuracy of LLMs in OCD using vignettes and found that LLMs outperformed medical and mental health professionals. This highlights the potential benefit of LLMs in assisting in the timely and accurate diagnosis of OCD, which usually entails a long delay in diagnosis and treatment.

Abstract The development of diagnostic tools for skin cancer based on artificial intelligence (AI) is increasing rapidly and will likely soon be widely implemented in clinical use. Even though the performance of these algorithms is promising in theory, there is limited evidence on the impact of AI assistance on human diagnostic decisions. Therefore, the aim of this systematic review and meta-analysis was to study the effect of AI assistance on the accuracy of skin cancer diagnosis. We searched PubMed, Embase, IEE Xplore, Scopus and conference proceedings for articles from 1/1/2017 to 11/8/2022. We included studies comparing the performance of clinicians diagnosing at least one skin cancer with and without deep learning-based AI assistance. Summary estimates of sensitivity and specificity of diagnostic accuracy with versus without AI assistance were computed using a bivariate random effects model. We identified 2983 studies, of which ten were eligible for meta-analysis. For clinicians without AI assistance, pooled sensitivity was 74.8% (95% CI 68.6–80.1) and specificity was 81.5% (95% CI 73.9–87.3). For AI-assisted clinicians, the overall sensitivity was 81.1% (95% CI 74.4–86.5) and specificity was 86.1% (95% CI 79.2–90.9). AI benefitted medical professionals of all experience levels in subgroup analyses, with the largest improvement among non-dermatologists. No publication bias was detected, and sensitivity analysis revealed that the findings were robust. AI in the hands of clinicians has the potential to improve diagnostic accuracy in skin cancer diagnosis. Given that most studies were conducted in experimental settings, we encourage future studies to further investigate these potential benefits in real-life settings.

BackgroundThe COVID-19 pandemic rapidly changed the landscape of clinical practice in the United States; telehealth became an essential mode of health care delivery, yet many components of telehealth use remain unknown years after the disease’s emergence. ObjectiveWe aim to comprehensively assess telehealth use and its associated factors in the United States. MethodsThis cross-sectional study used a nationally representative survey (Health Information National Trends Survey) administered to US adults (≥18 years) from March 2022 through November 2022. To assess telehealth adoption, perceptions of telehealth, satisfaction with telehealth, and the telehealth care purpose, we conducted weighted descriptive analyses. To identify the subpopulations with low adoption of telehealth, we developed a weighted multivariable logistic regression model. ResultsAmong a total of 6252 survey participants, 39.3% (2517/6252) reported telehealth use in the past 12 months (video: 1110/6252, 17.8%; audio: 876/6252, 11.6%). The most prominent reason for not using telehealth was due to telehealth providers failing to offer this option (2200/3529, 63%). The most common reason for respondents not using offered telehealth services was a preference for in-person care (527/578, 84.4%). Primary motivations to use telehealth were providers’ recommendations (1716/2517, 72.7%) and convenience (1516/2517, 65.6%), mainly for acute minor illness (600/2397, 29.7%) and chronic condition management (583/2397, 21.4%), yet care purposes differed by age, race/ethnicity, and income. The satisfaction rate was predominately high, with no technical problems (1829/2517, 80.5%), comparable care quality to that of in-person care (1779/2517, 75%), and no privacy concerns (1958/2517, 83.7%). Younger individuals (odd ratios [ORs] 1.48-2.23; 18-64 years vs ≥75 years), women (OR 1.33, 95% CI 1.09-1.61), Hispanic individuals (OR 1.37, 95% CI 1.05-1.80; vs non-Hispanic White), those with more education (OR 1.72, 95% CI 1.03-2.87; at least a college graduate vs less than high school), unemployed individuals (OR 1.25, 95% CI 1.02-1.54), insured individuals (OR 1.83, 95% CI 1.25-2.69), or those with poor general health status (OR 1.66, 95% CI 1.30-2.13) had higher odds of using telehealth. ConclusionsTo our best knowledge, this is among the first studies to examine patient factors around telehealth use, including motivations to use, perceptions of, satisfaction with, and care purpose of telehealth, as well as sociodemographic factors associated with telehealth adoption using a nationally representative survey. The wide array of descriptive findings and identified associations will help providers and health systems understand the factors that drive patients toward or away from telehealth visits as the technology becomes more routinely available across the United States, providing future directions for telehealth use and telehealth research.

Abstract Poor mental health has been found to be more prevalent among those with cancer and is considered a public health crisis since COVID-19. This study assessed the impact of COVID-19 and cancer survivorship on mental health and investigated factors, including online patient-provider communications (OPPC; email/internet/tablet/smartphone), associated with poor mental health prior to and during the early COVID-19. Nationally representative Health Information National Trends Survey data during 2017–2020 (n = 15,871) was used. While the prevalence of poor mental health was high (40–42%), Difference-In-Difference analyses revealed that cancer survivorship and COVID-19 were not associated with poor mental health. However, individuals that used OPPC had 40% higher odds of poor mental health. Low socioeconomic status (low education/income), younger age (18–64 years), and female birth gender were also associated with poor mental health. Findings highlight the persistence of long-standing mental health inequities and identify that OPPC users might be those who need mental health support.

BackgroundHealth care providers and health-related researchers face significant challenges when applying sentiment analysis tools to health-related free-text survey data. Most state-of-the-art applications were developed in domains such as social media, and their performance in the health care context remains relatively unknown. Moreover, existing studies indicate that these tools often lack accuracy and produce inconsistent results. ObjectiveThis study aims to address the lack of comparative analysis on sentiment analysis tools applied to health-related free-text survey data in the context of COVID-19. The objective was to automatically predict sentence sentiment for 2 independent COVID-19 survey data sets from the National Institutes of Health and Stanford University. MethodsGold standard labels were created for a subset of each data set using a panel of human raters. We compared 8 state-of-the-art sentiment analysis tools on both data sets to evaluate variability and disagreement across tools. In addition, few-shot learning was explored by fine-tuning Open Pre-Trained Transformers (OPT; a large language model [LLM] with publicly available weights) using a small annotated subset and zero-shot learning using ChatGPT (an LLM without available weights). ResultsThe comparison of sentiment analysis tools revealed high variability and disagreement across the evaluated tools when applied to health-related survey data. OPT and ChatGPT demonstrated superior performance, outperforming all other sentiment analysis tools. Moreover, ChatGPT outperformed OPT, exhibited higher accuracy by 6% and higher F-measure by 4% to 7%. ConclusionsThis study demonstrates the effectiveness of LLMs, particularly the few-shot learning and zero-shot learning approaches, in the sentiment analysis of health-related survey data. These results have implications for saving human labor and improving efficiency in sentiment analysis tasks, contributing to advancements in the field of automated sentiment analysis.

Introduction: Indoor tanning beds cause more than 450,000 new skin cancers each year, yet their use remains common, with a global indoor tanning prevalence of 10.4%. Social media provides an opportunity for cost-effective, targeted public health messaging. We sought to direct Instagram users at high risk of indoor tanning to accurate health information about the risks of indoor tanning and to reduce indoor tanning bed use. Methods: We disseminated a public health campaign on Instagram on April 6–27, 2022 with 34 video and still-image advertisements. We had 2 target audiences at high risk of indoor tanning: women aged 18–30 years in Kentucky, Nebraska, Ohio, or Tennessee interested in indoor tanning and men aged 18–45 years in California interested in indoor tanning. To evaluate the impact of the campaign, we tracked online metrics, including website visits, and conducted an interrupted time-series analysis of foot traffic data in our target states for all tanning salons documented on SafeGraph from January 1, 2018 to 3 months after the campaign. Results: Our indoor tanning health information advertisements appeared on Instagram feeds 9.1 million times, reaching 1.06 million individuals. We received 7,004 views of our indoor tanning health information landing page (Average Time on Page of 56 seconds). We did not identify a significant impact on foot traffic data on tanning salons. Conclusions: We show the successful use of social media advertising to direct high-risk groups to online health information about indoor tanning. Future research quantifying tanning visits before and after indoor tanning interventions is needed to guide future public health efforts.

BackgroundOnline patient-provider communication (OPPC) is crucial in enhancing access to health information, self-care, and related health outcomes among cancer survivors. The necessity of OPPC increased during SARS-CoV-2/COVID-19, yet investigations in vulnerable subgroups have been limited. ObjectiveThis study aims to assess the prevalence of OPPC and sociodemographic and clinical characteristics associated with OPPC among cancer survivors and adults without a history of cancer during COVID-19 versus pre–COVID-19. MethodsNationally representative cross-sectional survey data (Health Information National Trends Survey 5, 2017-2020) were used among cancer survivors (N=1900) and adults without a history of cancer (N=13,292). COVID-19 data included data from February to June 2020. We calculated the prevalence of 3 types of OPPC, defined as using the email/internet, tablet/smartphone, or electronic health record (EHR) for patient-provider communication, in the past 12 months. To investigate the associations of sociodemographic and clinical factors with OPPC, multivariable-adjusted weighted logistic regression was performed to obtain odds ratios (ORs) and 95% CIs. ResultsThe average prevalence of OPPC increased from pre-COVID to COVID among cancer survivors (39.7% vs 49.7%, email/internet; 32.2% vs 37.9%, tablet/smartphone; 19.0% vs 30.0%, EHR). Cancer survivors (OR 1.32, 95% CI 1.06-1.63) were slightly more likely to use email/internet communications than adults without a history of cancer prior to COVID-19. Among cancer survivors, the email/internet (OR 1.61, 95% CI 1.08-2.40) and EHRs (OR 1.92, 95% CI 1.22-3.02) were more likely to be used during COVID-19 than pre–COVID-19. During COVID-19, subgroups of cancer survivors, including Hispanics (OR 0.26, 95% CI 0.09-0.71 vs non-Hispanic Whites) or those with the lowest income (US $50,000-

Objective Although black patients are more likely to have advanced melanomas at diagnosis, with a 5-year survival rate among black patients of 70% compared with 92% for white patients, black people are generally not the focus of melanoma public health campaigns. We sought to explore awareness and perspectives of melanoma among black people to inform the development of relevant and valued public health messages to promote early detection of melanoma.Design Inductive thematic analysis of in-depth semistructured interviews.Setting Interviews were conducted with participants via video software or telephone in the USA.Participants Participants were adults from the USA who self-identified as African American or black. Recruitment flyers were posted around the San Francisco Bay Area and shared on our team Facebook page, with further participants identified through snowball sampling.Results We interviewed 26 participants from 10 different states. Overall, 12 were men and 14 were women, with a mean age of 43 years (range 18–85). We identified five key themes regarding melanoma awareness in black people: (1) lack of understanding of term ‘melanoma’ and features of skin cancer; (2) do not feel at risk of melanoma skin cancer; (3) surprise that melanoma can occur on palms, soles and nails; (4) skin cancer awareness messages do not apply to or include black people; and (5) Importance of relationship with healthcare and habits of utilisation.Conclusions Analysis of these in-depth semistructured interviews illuminate the pressing need for health information on melanoma designed specifically for black people. We highlight two key points for focused public health messaging: (1) melanoma skin cancer does occur in black people and (2) high-risk sites for melanoma in black people include the palms, soles and nail beds. Therefore, public health messages for black people and their healthcare providers may involve productively checking these body surface areas.

Differences between computationally generated and human-generated themes in unstructured text are important to understand yet difficult to assess formally. In this study, we bridge these approaches through two contributions. First, we formally compare a primarily computational approach, topic modeling, to a primarily human-driven approach, qualitative thematic coding, in an impactful context: physician mothers’ experience of workplace discrimination. Second, we compare our chosen topic model to a principled alternative topic model to make explicit study design decisions meriting consideration in future research. By formally contrasting computationally generated (i.e. topic modeling) and human-generated (i.e. thematic coding) knowledge, we shed light on issues of interest to several audiences, notably computational social scientists who wish to understand study design tradeoffs, and qualitative researchers who may wish to leverage computational methods to improve the speed and reproducibility of labor-intensive coding. Although useful in other domains, we highlight the value of fast, reproducible methods to better understand experiences of workplace discrimination.

The pervasiveness of social media is irrefutable, with 72% of adults reporting using at least one social media platform and an average daily usage of 2 hours. Social media has been shown to influence health-related behaviors, and it offers a powerful tool through which we can rapidly reach large segments of the population with tailored health messaging. However, despite increasing interest in using social media for dissemination of public health messaging and research exploring the dangers of misinformation on social media, the specifics of how public health practitioners can effectively use social media for health promotion are not well described. In this viewpoint, we propose a novel framework with the following 5 key principles to guide the use of social media for public health campaigns: (1) tailoring messages and targeting them to specific populations—this may include targeting messages to specific populations based on age, sex, or language spoken; interests; or geotargeting messages at state, city, or zip code level; (2) including members of the target population in message development—messages should be designed with and approved by members of the community they are designed to reach, to ensure cultural sensitivity and trust-building; (3) identifying and addressing misinformation—public health practitioners can directly address misinformation through myth-busting messages, in which false claims are highlighted and explained and accurate information reiterated; (4) leveraging information sharing—when designing messages for social media, it is crucial to consider their “shareability,” and consider partnering with social media influencers who are trusted messengers among their online followers; and (5) evaluating impact by measuring real-world outcomes, for example measuring foot traffic data. Leveraging social media to deliver public health campaigns enables us to capitalize on sophisticated for-profit advertising techniques to disseminate tailored messaging directly to communities that need it most, with a precision far beyond the reaches of conventional mass media. We call for the Centers for Disease Control and Prevention as well as state and local public health agencies to continue to optimize and rigorously evaluate the use of social media for health promotion.

Abstract Background The COVID‐19 pandemic has accelerated the adoption of telemedicine, including teledermatology. Monitoring skin lesions using teledermatology may become increasingly important for several skin diseases, including low‐risk skin cancers. The purpose of this study was to describe the key factors that could serve as barriers or facilitators to skin disease monitoring using mobile health technology (mHealth) in older adults. Methods Older adult dermatology patients 65 years or older and their caregivers who have seen a dermatologist in the last 18 months were interviewed and surveyed between December 2019 and July 2020. The purpose of these interviews was to better understand attitudes, beliefs and behaviours that could serve as barriers and facilitators to the use of mHealth and active surveillance to monitor low‐risk skin cancers. Results A total of 33 interviews leading to 6022 unique excerpts yielded 8 factors, or themes, that could serve as barriers, facilitators or both to mHealth and active surveillance. We propose an integrated conceptual framework that highlights the interaction of these themes at both the patient and provider level, including care environment, support systems and personal values. Discussion and conclusions These preliminary findings reveal factors influencing patient acceptance of active surveillance in dermatology, such as changes to the patient‐provider interaction and alignment with personal values. These factors were also found to influence adoption of mHealth interventions. Given such overlap, it is essential to address barriers and facilitators from both domains when designing a new dermatology active surveillance approach with novel mHealth technology. Patient or public contribution The patients included in this study were participants during the data collection process. Members of the Stanford Healthcare and Denver Tech Dermatology health‐care teams aided in the recruitment phase of the data collection process.

Abstract Background Basal cell carcinoma (BCC) is a slow-growing, rarely lethal skin cancer that affects people 65 years or older. A range of treatment options exist for BCC, but there is little evidence available to guide patients and providers in selecting the best treatment options. Objectives This study outlines the development of a patient decision aid (PDA) for low-risk BCC that can be used by patients and providers to assist in shared decision-making. Methods In accordance with the International Patient Decision Aids Standards (IPDAS) Collaboration framework, feedback from focus groups and semi-structured interviews with patients and providers, an initial prototype of the PDA was developed. This was tested using cognitive interviews and iteratively updated. Results We created eighteen different iterations using feedback from 24 patients and 34 providers. The key issues identified included: 1) Addressing fear of cancer; 2) Communicating risk and uncertainty; 3) Values clarification; and 4) Time lag to benefit. Limitations The PDA does not include all possible treatment options and is currently paper based. Conclusions Our PDA has been specifically adapted and designed to support patients with a limited life expectancy in making decisions about their low risk BCC together with their doctors.

COVID-19 has presented an unprecedented challenge to human welfare. Indeed, we have witnessed people experiencing a rise of depression, acute stress disorder, and worsening levels of subclinical psychological distress. Finding ways to support individuals' mental health has been particularly difficult during this pandemic. An opportunity for intervention to protect individuals' health & well-being is to identify the existing sources of consolation and hope that have helped people persevere through the early days of the pandemic. In this paper, we identified positive aspects, or “silver linings,” that people experienced during the COVID-19 crisis using computational natural language processing methods and qualitative thematic content analysis. These silver linings revealed sources of strength that included finding a sense of community, closeness, gratitude, and a belief that the pandemic may spur positive social change. People's abilities to engage in benefit-finding and leverage protective factors can be bolstered and reinforced by public health policy to improve society's resilience to the distress of this pandemic and potential future health crises.

BackgroundThe U.S. has experienced an unprecedented number of orders to shelter in place throughout the ongoing COVID-19 pandemic. We aimed to ascertain whether social distancing; difficulty with daily activities; and levels of concern regarding COVID-19 changed after the March 16, 2020 announcement of the nation's first shelter-in-place orders (SIPO) among individuals living in the seven affected counties in the San Francisco Bay Area.MethodsWe conducted an online, cross-sectional social media survey from March 14 -April 1, 2020. We measured changes in social distancing behavior; experienced difficulties with daily activities (i.e., access to healthcare, childcare, obtaining essential food and medications); and level of concern regarding COVID-19 after the March 16 shelter-in-place announcement in the San Francisco Bay Area versus elsewhere in the U.S.ResultsIn this non-representative sample, the percentage of respondents social distancing all of the time increased following the shelter-in-place announcement in the Bay Area (9.2%, 95% CI: 6.6, 11.9) and elsewhere in the U.S. (3.4%, 95% CI: 2.0, 5.0). Respondents also reported increased difficulty obtaining hand sanitizer, medications, and in particular respondents reported increased difficulty obtaining food in the Bay Area (13.3%, 95% CI: 10.4, 16.3) and elsewhere (8.2%, 95% CI: 6.6, 9.7). We found limited evidence that level of concern regarding the COVID-19 crisis changed following the announcement.ConclusionThis study characterizes early changes in attitudes, behaviors, and difficulties. As states and localities implement, rollback, and reinstate shelter-in-place orders, ongoing efforts to more fully examine the social, economic, and health impacts of COVID-19, especially among vulnerable populations, are urgently needed.

Background: Social media holds exciting promise for advancing mental health research recruitment, however, the extent and efficacy to which these platforms are currently in use are underexplored. Objective: A systematic review was conducted to characterize the current use and efficacy of social media in recruiting participants for mental health research. Method: A literature review was performed using MEDLINE, EMBASE, and PsychINFO. Only non-duplicative manuscripts written in the English language and published between 1/1/2004–3/31/2019 were selected for further screening. Data extracted included study type and design, participant inclusion criteria, social media platform, advertising strategy, final recruited sample size, recruitment location, year, monetary incentives, comparison to other recruitment methods if performed, and final cost per participant. Results: A total of 176 unique studies that used social media for mental health research recruitment were reviewed. The majority of studies were cross-sectional (62.5%) in design and recruited adults. Facebook was overwhelmingly the recruitment platform of choice (92.6%), with the use of paid advertisements being the predominant strategy (60.8%). Of the reviewed studies, substance abuse (43.8%) and mood disorders (15.3%) were the primary subjects of investigation. In 68.3% of studies, social media recruitment performed as well as or better than traditional recruitment methods in the number and cost of final enrolled participants. The majority of studies used Facebook for recruitment at a median cost per final recruited study participant of $19.47. In 55.6% of the studies, social media recruitment was the more cost-effective recruitment method when compared to traditional methods (e.g., referrals, mailing). Conclusion: Social media appears to be an effective and economical recruitment tool for mental health research. The platform raises methodological and privacy concerns not covered in current research regulations that warrant additional consideration.

The Institute of Medicine has identified serious deficiencies in the measurement of cancer care quality, including the effects on quality of life and patient experience. Moreover, comparisons of quality in Veterans Affairs Medical Centers (VA) and other sites are timely now that many Veterans can choose where to seek care. To compare quality of ambulatory surgical care for keratinocyte carcinoma (KC) between a VA and fee-for-service (FFS) practice, we used unique clinical and patient-reported data from a comparative effectiveness study. Patients were enrolled in 1999-2000 and followed for a median of 7.2 years. The practices differed in a few process measures (e.g., median time between biopsy and treatment was 7.5 days longer at VA) but there were no substantial or consistent differences in clinical outcomes or a broad range of patient-reported outcomes. For example, 5-year tumor recurrence rates were equally low (3.6% [2.3-5.5] at VA and 3.4% [2.3-5.1] at FFS), and similar proportions of patients reported overall satisfaction at one year (78% at VA and 80% at FFS, P = 0.69). These results suggest that the quality of care for KC can be compared comprehensively in different health care systems, and suggest that quality of care for KC was similar at a VA and FFS setting.