The Medication Adherence Report Scale (MARS-5) in a Swedish sample with bipolar disorder - a pilot study

Abstract

Objectives: To test the acceptability and reliability of the Swedish translation of the Medication Adherence Report Scale-5 (MARS-5) in a sample of patients who use mood stabilising medicines for bipolar disorder. A further aim was to compare the Swedish translation of the MARS-5 with the Swedish translation of the four-item scale Morisky Medication Adherence Scale (MMAS-4).Method: The study population (n=47, 70% women) was recruited through patient education sessions, at a Patient Association meeting, in Gothenburg, Sweden, as well as through advertisements on the home pages and newsletters of the Swedish patient associations. Participants received the Swedish translations of the MARS-5 and the MMAS-4, and questions on age, education, and country of birth. Reliability was examined for internal consistency (Cronbach’s α) and test-retest (intraclass correlation (ICC), MARS-5: Pearson’s correlation coefficient (r), MMAS-4: Spearman’s rho (ρ)). The acceptability of the MARS-5 was examined with a correlation analysis between the MARS-5 and the MMAS-4 and for face validity.Results: In the study population 53.3% were categorised as adherent with the MARS-5 and 82.6% using the MMAS-4. The value of Cronbach’s α was 0.66 for the MARS-5 and 0.37 for the MMAS-4. The test-retest of the MARS-5 resulted in r=0.90 and ICC=0.91. Corresponding values for the MMAS-4 were ρ=0.84 and ICC=0.85. The correlation between the MARS-5 and the MMAS-4 was 0.55. The face validity resulted in four comments regarding difficulties in answering the MARS-5.Conclusion: The Swedish translation of the MARS-5 ought to be used instead of the MMAS-4 to measure self-reported adherence in a Swedish sample with bipolar disorder. The MARS-5 showed good psychometric properties.