A Web-Based Physical Activity Intervention for Spanish-Speaking Latinas: A Costs and Cost-Effectiveness Analysis

BACKGROUND: Latinas report particularly low levels of physical activity and suffer from greater rates of lifestyle-related conditions such as obesity and diabetes. Interventions are needed that can increase physical activity in this growing population in a large-scale, cost-effective manner. Web-based interventions may have potential given the increase in Internet use among Latinas and the scalability of Web-based programs. OBJECTIVE: To examine the costs and cost-effectiveness of a Web-based, Spanish-language physical activity intervention for Latinas compared to a wellness contact control. METHODS: Healthy adult Latina women (N=205) were recruited from the community and randomly assigned to receive a Spanish-language, Web-based, individually tailored physical activity intervention (intervention group) or were given access to a website with content on wellness topics other than physical activity (control group). Physical activity was measured using the 7-Day Physical Activity Recall interview and ActiGraph accelerometers at baseline, 6 months (ie, postintervention), and 12 months (ie, maintenance phase). Costs were estimated from a payer perspective and included all features necessary to implement the intervention in a community setting, including staff time (ie, wages, benefits, and overhead), materials, hardware, website hosting, and routine website maintenance. RESULTS: At 6 months, the costs of running the intervention and control groups were US $17 and US $8 per person per month, respectively. These costs fell to US $12 and US $6 per person per month at 12 months, respectively. Linear interpolation showed that intervention participants increased their physical activity by 1362 total minutes at 6 months (523 minutes by accelerometer) compared to 715 minutes for control participants (186 minutes by accelerometer). At 6 months, each minute increase in physical activity for the intervention group cost US $0.08 (US $0.20 by accelerometer) compared to US $0.07 for control participants (US $0.26 by accelerometer). Incremental cost-per-minute increases associated with the intervention were US $0.08 at 6 months and US $0.04 at 12 months (US $0.16 and US $0.08 by accelerometer, respectively). Sensitivity analyses showed variations in staffing costs or intervention effectiveness yielded only modest changes in incremental costs. CONCLUSIONS: While the Web-based physical activity intervention was more expensive than the wellness control, both were quite low cost compared to face-to-face or mail-delivered interventions. Cost-effectiveness ranged markedly based on physical activity measure and was similar between the two conditions. Overall, the Web-based intervention was effective and low cost, suggesting a promising channel for increasing physical activity on a large scale in this at-risk population. CLINICALTRIAL: Clinicaltrials.gov NCT01834287; https://clinicaltrials.gov/ct2/show/NCT01834287 (Archived by WebCite at http://www.webcitation.org/6nyjX9Jrh)