Background/Purpose: Testimonials have been shown to have a strong influence on patient decision-making. Patients are increasingly accessing the Internet as a source of medical information. In this study, we sought to examine whether the number of positive or negative reviews accessed influence willingness to take a medication.

Methods: We administered a survey to Mechanical Turk workers (≥ 50 years old) who were not currently taking any medications for osteoporosis or osteopenia. Participants’ received information describing outcomes using one of three randomly distributed formats: icon array, icon array followed by narrative reviews, or narrative reviews followed by icon arrays. Subjects in the latter two groups were included for these analyses. The outcomes in the reviews were purposely composed to match the actual distribution of outcomes associated with a commonly used osteoporosis treatment. Each review was rated by actual patients on a 5-point scale (illustrated by stars). Willingness to take medication was measured on a 10-point scale before and after reading the description of the medication and reviews (in the two groups including testimonials). Reviews could be read by clicking on links. We examined whether ratings and number of online testimonials accessed influenced change in willingness to take the medication using linear mixed models adjusting for baseline willingness, age, sex, race, education, numeracy, osteoporosis/osteopenia and previous history of osteoporosis/fracture. Reviews rated above ‘3” were defined as positive; the rest were defined negative. The influence of positive and negative reviews was examined in separate models due to their strong collinearity.

Results: 276 participants were randomized to one of the two groups including testimonials. The mean (SD) age was 59 (7), and the majority were female (61%), white (73%) and college graduates (67%). We found significantly positive associations between the number of accessed positive testimonials and willingness to take the medication (p=0.015). However, no significant association was found between the number of accessed negative testimonials and willingness. Education level and numeracy did not modify the association between reviews accessed and willingness to take the medication (data not shown).

Conclusion: In this study, we found that willingness to take a medication for osteoporosis was positively associated with the number of positive online testimonials accessed. These results support the need to examine the impact of evidenced-based narratives as decision support tools.