Distinct Profiles on Subjective and Objective Adherence Measures in Patients Prescribed Antidepressants

Hans Wouters, Didi Rhebergen, Marcia Vervloet, Antoine Egberts, Katja Taxis, Liset van Dijk, Helga Gardarsdottir

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Objective: A recurrent observation is that associations between self-reported and objective medication adherence measures are often weak to moderate. Our aim was therefore to identify patients with different profiles on self-reported and objective adherence measures. Study Design and Setting: This was an observational study of 221 community pharmacy patients who were dispensed antidepressants. Adherence profiles were estimated with Latent Profile Analysis (LPA) using data on self-reported adherence (Medication Adherence Rating Scale) complemented with data on medication beliefs (perceived necessity and concerns measured with the Beliefs about Medicines Questionnaire) and data from objective adherence measures (electronic monitoring of medication taking and the Medication Possession Ratio calculated from pharmacy dispensing data). Results: ‘Goodness-of-fit’ statistics indicated the presence of three classes: “concordantly high adherent” (83%, high adherence on all measures), “concordantly suboptimal adherent” (11%, low adherence on all measures), and “discordant” (6%, high self-reported adherence but lower adherence on objective measures). Conclusion: Most patients had concordant outcomes on self-reported and objective measures of adherence. A small discordant class had high self-reported but low objective adherence. LPA will enable sensitivity analyses in future studies, for example excluding patients from the discordant class.
Original languageEnglish
JournalDrugs
DOIs
Publication statusPublished - 2019

Cite this

Wouters, Hans ; Rhebergen, Didi ; Vervloet, Marcia ; Egberts, Antoine ; Taxis, Katja ; van Dijk, Liset ; Gardarsdottir, Helga. / Distinct Profiles on Subjective and Objective Adherence Measures in Patients Prescribed Antidepressants. In: Drugs. 2019.
@article{2416ce4fcde744cba0e2027a97687751,
title = "Distinct Profiles on Subjective and Objective Adherence Measures in Patients Prescribed Antidepressants",
abstract = "Objective: A recurrent observation is that associations between self-reported and objective medication adherence measures are often weak to moderate. Our aim was therefore to identify patients with different profiles on self-reported and objective adherence measures. Study Design and Setting: This was an observational study of 221 community pharmacy patients who were dispensed antidepressants. Adherence profiles were estimated with Latent Profile Analysis (LPA) using data on self-reported adherence (Medication Adherence Rating Scale) complemented with data on medication beliefs (perceived necessity and concerns measured with the Beliefs about Medicines Questionnaire) and data from objective adherence measures (electronic monitoring of medication taking and the Medication Possession Ratio calculated from pharmacy dispensing data). Results: ‘Goodness-of-fit’ statistics indicated the presence of three classes: “concordantly high adherent” (83{\%}, high adherence on all measures), “concordantly suboptimal adherent” (11{\%}, low adherence on all measures), and “discordant” (6{\%}, high self-reported adherence but lower adherence on objective measures). Conclusion: Most patients had concordant outcomes on self-reported and objective measures of adherence. A small discordant class had high self-reported but low objective adherence. LPA will enable sensitivity analyses in future studies, for example excluding patients from the discordant class.",
author = "Hans Wouters and Didi Rhebergen and Marcia Vervloet and Antoine Egberts and Katja Taxis and {van Dijk}, Liset and Helga Gardarsdottir",
year = "2019",
doi = "10.1007/s40265-019-01107-y",
language = "English",
journal = "Drugs",
issn = "0012-6667",
publisher = "Adis International Ltd",

}

Distinct Profiles on Subjective and Objective Adherence Measures in Patients Prescribed Antidepressants. / Wouters, Hans; Rhebergen, Didi; Vervloet, Marcia; Egberts, Antoine; Taxis, Katja; van Dijk, Liset; Gardarsdottir, Helga.

In: Drugs, 2019.

Research output: Contribution to journalArticleAcademicpeer-review

TY - JOUR

T1 - Distinct Profiles on Subjective and Objective Adherence Measures in Patients Prescribed Antidepressants

AU - Wouters, Hans

AU - Rhebergen, Didi

AU - Vervloet, Marcia

AU - Egberts, Antoine

AU - Taxis, Katja

AU - van Dijk, Liset

AU - Gardarsdottir, Helga

PY - 2019

Y1 - 2019

N2 - Objective: A recurrent observation is that associations between self-reported and objective medication adherence measures are often weak to moderate. Our aim was therefore to identify patients with different profiles on self-reported and objective adherence measures. Study Design and Setting: This was an observational study of 221 community pharmacy patients who were dispensed antidepressants. Adherence profiles were estimated with Latent Profile Analysis (LPA) using data on self-reported adherence (Medication Adherence Rating Scale) complemented with data on medication beliefs (perceived necessity and concerns measured with the Beliefs about Medicines Questionnaire) and data from objective adherence measures (electronic monitoring of medication taking and the Medication Possession Ratio calculated from pharmacy dispensing data). Results: ‘Goodness-of-fit’ statistics indicated the presence of three classes: “concordantly high adherent” (83%, high adherence on all measures), “concordantly suboptimal adherent” (11%, low adherence on all measures), and “discordant” (6%, high self-reported adherence but lower adherence on objective measures). Conclusion: Most patients had concordant outcomes on self-reported and objective measures of adherence. A small discordant class had high self-reported but low objective adherence. LPA will enable sensitivity analyses in future studies, for example excluding patients from the discordant class.

AB - Objective: A recurrent observation is that associations between self-reported and objective medication adherence measures are often weak to moderate. Our aim was therefore to identify patients with different profiles on self-reported and objective adherence measures. Study Design and Setting: This was an observational study of 221 community pharmacy patients who were dispensed antidepressants. Adherence profiles were estimated with Latent Profile Analysis (LPA) using data on self-reported adherence (Medication Adherence Rating Scale) complemented with data on medication beliefs (perceived necessity and concerns measured with the Beliefs about Medicines Questionnaire) and data from objective adherence measures (electronic monitoring of medication taking and the Medication Possession Ratio calculated from pharmacy dispensing data). Results: ‘Goodness-of-fit’ statistics indicated the presence of three classes: “concordantly high adherent” (83%, high adherence on all measures), “concordantly suboptimal adherent” (11%, low adherence on all measures), and “discordant” (6%, high self-reported adherence but lower adherence on objective measures). Conclusion: Most patients had concordant outcomes on self-reported and objective measures of adherence. A small discordant class had high self-reported but low objective adherence. LPA will enable sensitivity analyses in future studies, for example excluding patients from the discordant class.

UR - https://www.scopus.com/inward/record.uri?partnerID=HzOxMe3b&scp=85064356765&origin=inward

UR - https://www.ncbi.nlm.nih.gov/pubmed/30941607

U2 - 10.1007/s40265-019-01107-y

DO - 10.1007/s40265-019-01107-y

M3 - Article

JO - Drugs

JF - Drugs

SN - 0012-6667

ER -