Objectives: The objectives of this study were twofold: 1) to evaluate the construct validity of the Treatment Satisfaction Questionnaire for Medication (TSQM v. II) using structural equation modeling (SEM); and 2) to assess its concurrent validity using medication adherence criteria.
Methods: Pharmacy patients filling a new medication prescription (n = 342) were recruited from 14 Michigan pharmacies to participate in a 4-week treatment satisfaction study. The TSQM v. II was tested for model fit against an established theoretical model (the Decisional Balance Model of Treatment Satisfaction) using hierarchical confirmatory factor analysis (HCFA). Regression and discriminant analytic models were used to examine the criterion-related validity of the measure.
Results: An exploratory factor analysis, used for TSQM v. II item reduction, revealed a strongly dimensional instrument (Effectiveness, Side Effects, and Convenience) and explained 88% of total pooled variance. Results of an HCFA using the final TSQM v. II items suggested a good model fit with the data (P > 0.54). In support of concurrent validity, the TSQM scales explained between 9% and 20% of the variance in dosing adherence and 60% of the variance in the likelihood of future use. Discriminant analysis demonstrated the superior classification power of the hierarchical model of treatment satisfaction over the discrete attribute model when predicting medication discontinuation.
Conclusions: The TSQM v. II has equivalent measurement characteristics as the TSQM v. I, yet uses four fewer items and more consistent wording. The value of the Decisional Balance Model for estimation of dosing adherence and medication persistence over time is discussed.