Abstract
BACKGROUND: Technology is being increasingly used to communicate health information, but there is limited knowledge on whether these strategies are effective for vulnerable populations, including non-English speaking or low-income individuals. OBJECTIVE: This study assessed how language preferences (eg, English, Spanish, or Chinese), smartphone ownership, and the type of clinic for usual source of care (eg, no usual source of care, nonintegrated safety net, integrated safety net, private or community clinic, academic tertiary medical center, or integrated payer-provider) affect technology use for health-related communication. METHODS: From May to September 2017, we administered a nonrandom, targeted survey to 1027 English-, Spanish-, and Chinese-speaking San Francisco residents and used weighted multivariable logistic regression analyses to assess predictors of five technology use outcomes. The three primary predictors of interest-language preference, smartphone ownership, and type of clinic for usual care-were adjusted for age, gender, race or ethnicity, limited English proficiency, educational attainment, health literacy, and health status. Three outcomes focused on use of email, SMS text message, or phone apps to communicate with clinicians. The two other outcomes were use of Web-based health videos or online health support groups. RESULTS: Nearly one-third of participants watched Web-based health videos (367/1027, 35.74%) or used emails to communicate with their clinician (318/1027, 30.96%). In adjusted analyses, individuals without smartphones had significantly lower odds of texting their clinician (adjusted odds ratio [aOR] 0.27, 95% CI 0.13-0.56), using online health support groups (aOR 0.14, 95% CI 0.04-0.55), or watching Web-based health videos (aOR 0.31, 95% CI 0.15-0.64). Relative to English-speaking survey respondents, individuals who preferred Chinese had lower odds of texting their clinician (aOR 0.25, 95% CI 0.08-0.79), whereas Spanish-speaking survey respondents had lower odds of using apps to communicate with clinicians (aOR 0.34, 95% CI 0.16-0.75) or joining an online support group (aOR 0.30, 95% CI 0.10-0.92). Respondents who received care from a clinic affiliated with the integrated safety net, academic tertiary medical center, or integrated payer-provider systems had higher odds than individuals without a usual source of care at using emails, SMS text messages, or apps to communicate with clinicians. CONCLUSIONS: In vulnerable populations, smartphone ownership increases the use of many forms of technology for health purposes, but device ownership itself is not sufficient to increase the use of all technologies for communicating with clinicians. Language preference impacts the use of technology for health purposes even after considering English proficiency. Health system factors impact patients' use of technology-enabled approaches for communicating with clinicians. No single factor was associated with higher odds of using technology for all health purposes; therefore, existing disparities in the use of digital health tools among diverse and vulnerable populations can only be addressed using a multipronged approach.