Real-Time Data Collection Using Text Messaging in a Primary Care Clinic

Background:The use of text messaging is nearly ubiquitous and represents a promising method of collecting data from diverse populations.Introduction:The purpose of this study was to assess the feasibility and acceptability of text message surveys in a clinical setting and to describe key lessons to minimize attrition.Methods:We obtained a convenience sample of individuals who entered the waiting room of a low-income, primary care clinic. Participants were asked to answer between 17 and 30 survey questions on a variety of health-related topics, including both open- and closed-ended questions. Descriptive statistics were used to characterize the participants and determine the response rates. Bivariate analyses were used to identify predictors of incomplete surveys.Results:Our convenience sample consisted of 461 individuals. Of those who attempted the survey, 80% (370/461) completed it in full. The mean age of respondents was 35.4 years (standard deviation=12.4). Respondents were predominantly non-Hispanic black (42%) or non-Hispanic white (41%), female (75%), and with at least some college education (70%). Of those who completed the survey, 84% (312/370) reported willingness to do another text message survey. Those with incomplete surveys answered a median of nine questions before stopping. Smartphone users were less likely to leave the survey incomplete compared with non-smartphone users (p=0.004).Discussion:Text-message surveys are a feasible and acceptable method to collect real-time data among low-income, clinic-based populations. Offering participants a setting for immediate survey completion, minimizing survey length, simplifying questions, and allowing free text responses for all questions may optimize response rates.