Evaluation of the Effect of Methodological Assumptions on Estimates of Adherence to Antipsychotics: A Real-World Data Study

Speaker(s)

Fuente-Moreno M1, Serrano-Blanco A1, Rubio-Valera M2, Dima AL1, Aznar-Lou I3
1Sant Joan de Déu Research Institute, Sant Boi de Llobregat, Spain, 2Parc Sanitari Sant Joan de Déu, CIBERESP, Barcelona, Spain, 3Institut de Recerca Sant Joan de Déu, Sant Boi de Llobregat, Barcelona, Spain

Presentation Documents

OBJECTIVES: The use of routinely collected data holds great potential for medication adherence research. However, data inconsistencies may challenge adherence assessment, particularly in complex clinical settings such as antipsychotic treatments. This study aimed to assess the frequency of dosing inconsistencies in prescription data and the effect of four dosing assumption strategies on adherence estimates for antipsychotic treatment.

METHODS: This was a retrospective cohort study, which linked prescription and dispensing data of patients (≥ 18 years) with at least one AP prescription between 2015-2016 and followed up until 2019, in Catalonia (Spain). Four strategies were proposed for selecting the recommended dosing in overlapping prescription periods for the same patient and antipsychotic drug: 1) the minimum dosing prescribed; 2) the dose corresponding to the latest prescription issued, 3) the highest dosing prescribed, and 4) all doses included in the overlapped period. For each strategy, one treatment episode per patient was selected and the Continuous Medication Availability measure was used to assess adherence. Descriptive statistics were used to describe results by strategy.

RESULTS: Of 277,324 prescriptions included, 76% overlapped with other prescriptions (40% with different recommended dosing instructions). The number and characteristics of patients and treatment episodes (18,292, 18,303, 18,339, and 18,536, respectively per strategy) were similar across strategies. Mean adherence was similar between strategies, ranging from 57-60%.

Whereas the proportion of poorly adherent patients (adherence ≤ 10%) was similar between strategies, the proportion of highly adherent patients (adherence ≥ 90%) was lower when selecting all doses (28%) compared to the other strategies (35%).

CONCLUSIONS: Despite the high prevalence of overlapping prescriptions, the strategies proposed did not show a major effect on the adherence estimates for antipsychotic treatment. Taking into consideration the particularities of antipsychotic prescription practices, selecting the highest dose in the overlapped period provided a more accurate adherence estimate.

Code

RWD151

Topic

Patient-Centered Research, Study Approaches

Topic Subcategory

Adherence, Persistence, & Compliance, Electronic Medical & Health Records

Disease

Mental Health (including addition), No Additional Disease & Conditions/Specialized Treatment Areas