Medication Adherence: Definition, Measurement, and Assessment Methods

This topic covers how medication adherence is defined and quantified — from self-report questionnaires and pill counts to pharmacy refill metrics and electronic monitoring. Because no single method is a gold standard, the choice of measure shapes the estimate, which is why standardized definitions and transparent reporting are central to adherence research.

Definition

Adherence measurement is the set of direct and indirect methods used to estimate the extent to which a person's medicine-taking matches an agreed regimen, typically expressed for the implementation phase as the proportion of prescribed doses taken or days covered over a defined period.

Scope

The entry surveys the conceptual definition of adherence and the main assessment methods, their strengths and biases, and the metrics derived from refill data. It is methodological reference material and does not prescribe how to monitor any individual's medication.

Core questions

• What is being measured — initiation, implementation, or persistence — and over what window?
• Which method (self-report, pill count, refill data, electronic monitoring, biological assay) best fits the question, and what biases does it carry?
• How are refill-based metrics such as the medication possession ratio and proportion of days covered calculated and thresholded?
• How comparable are adherence estimates across studies using different measures?

Key concepts

• Direct methods (observed dosing, drug/metabolite assays)
• Indirect methods (self-report, pill counts, refill records, electronic monitoring)
• Medication possession ratio (MPR)
• Proportion of days covered (PDC)
• Adherence thresholds (e.g., commonly 80%)
• Self-report instruments and their ceiling/social-desirability bias
• Initiation, implementation, and persistence windows

Key theories

ABC taxonomy of adherence phases

Distinguishes initiation, implementation, and discontinuation, clarifying that different measurement methods capture different phases and that estimates are not interchangeable without specifying which phase and window apply.

Mechanisms

Measurement methods trade off accuracy, cost, and feasibility. Direct methods (directly observed therapy, drug or metabolite levels) confirm ingestion but are intrusive and can be confounded by metabolism. Indirect methods are more practical: self-report questionnaires are inexpensive but prone to overestimation; pill counts assume removed doses were taken; pharmacy refill metrics such as the medication possession ratio and proportion of days covered infer implementation and persistence from dispensing records; electronic monitoring timestamps container openings. Refill metrics require decisions about the observation window, handling of overlapping supplies, and an adherence cut-point — commonly 80% — above which a patient is classified as adherent, a threshold whose appropriateness varies by drug class.

Clinical relevance

Choosing and reporting an adherence measure correctly is part of appraising adherence evidence and of pharmacy services research. This entry explains how measurement works as a methodological matter; it does not direct how any clinician should monitor or act on an individual patient's adherence.

Epidemiology

Reported adherence rates vary widely partly because methods disagree: self-report tends to overestimate relative to refill or electronic data, and refill-based classification depends on the chosen threshold and window. This measurement dependence is a recurring caution in syntheses of adherence research.

Evidence & guidelines

Methodological reviews recommend specifying the adherence phase, the measurement method, and the metric definition (including window and threshold) so that estimates are interpretable and comparable; proposals for standardized refill-based definitions and empirically derived cut-points support this aim. No single instrument is endorsed as a universal gold standard.

History

Early work relied on pill counts and clinician judgement; the 1986 Morisky self-report scale became a widely used questionnaire, and the spread of pharmacy claims databases popularized refill metrics such as MPR and PDC in the 1990s and 2000s. The 2012 ABC taxonomy later clarified which phase each method captures, sharpening reporting standards.

Debates

Is an 80% adherence threshold appropriate across drug classes?

The widely used 80% cut-point for 'adherent' was derived in particular contexts; empirical analyses show that the clinically meaningful threshold can differ by therapeutic area, so a single cut-point may misclassify patients.

How much do self-report measures overestimate adherence?

Self-report is cheap and scalable but subject to recall and social-desirability bias, generally yielding higher estimates than objective methods, which complicates comparisons across studies.

Key figures

• Donald Morisky
• Bernard Vrijens
• John Urquhart
• Lisa Hess
• Bradley Martin

Seminal works

• vrijens-2012
• morisky-1986
• hess-2006
• karve-2009

Frequently asked questions

What is the difference between MPR and PDC?

Both estimate adherence from pharmacy refill records. The medication possession ratio sums days' supply dispensed over a period and can exceed 100% with overlapping fills, whereas the proportion of days covered counts the unique days a patient had medication available, capping at 100% and generally giving a more conservative estimate.

Why is there no single best way to measure adherence?

Each method captures a different phase and carries different biases — direct methods confirm ingestion but are intrusive, self-report is scalable but overestimates, and refill data infer behavior from dispensing rather than observing it — so the right choice depends on the research or service question.