Use of best practice strategies to improve longitudinal study participation: Analysis of telephone contact data from the northwest regional burn model system

Abstract

Successful acquisition of survey data in any longitudinal study is key to reducing bias and determining reliable, generalizable findings. As part of a multicenter longitudinal study of patient-reported outcomes, a minimum benchmark for follow-up was set at 80%. In review of our center's participation, we found this benchmark was not consistently met. Thus, we identified a set of actions for study personnel concerning data acquisition for subject surveys. These actions, termed Best Practices, were implemented in January 2015. The purpose of this review is to determine whether the Best Practices were associated with improved follow-up. A comparison of success rates for the 6-months prior to and following program implementation was made. In addition, cell phone records for the 6-month period following implementation were reviewed for call characteristics (e.g. minutes used, date, and time of call), demographic representation (call area code), call timing (day and time of week), and call success. A "successful call" was defined as a call time longer than 5 minutes with acquisition of survey data. For the cell phone review, we attempted to reach a total of 98 subjects. Table 1 outlines information gleaned from the review of records following implementation of the Best Practices. Table 2 outlines the success rates for the two time periods compared (significance set at < 0.05; two-tailed z-test). Use of the Best Practices by research personnel resulted in improved follow-up rates for 2 of 3 study time points. Of interest is the effort required to reach study participants by telephone with a success rate of only 12% of all calls made. These data provide information that can be used by investigators as they plan future research and require evidence to substantiate personnel effort and/or Full-Time-Equivalent (FTE) requirements.