Methodological quality of the review: Low confidence

Author: Velentgas P, Sheffield R, Nordstrom BL, Johnson E, Do T, Mentor SM, Seeger JD

Geographical coverage: United States of America (USA), Canada and Europe

Sector: Glaucoma

Sub-sector: Persistence with medications

Equity focus: None specified

Review type: Other review

Quantitative synthesis method: Narrative/thematic synthesis

Qualitative synthesis method: Not applicable

Background: It is important for clinical decisions to understand the factors for patients’ adherence with medications. Therefore, comparing analysis of adherence and understanding these factors is crucial.

Objectives: To summarise studies of persistence with ocular hypotensives, and review the literature of adherence and comparing methods used for analysis of adherence for glaucoma, anti-hypertensives and anti-hyperlipidemics.

Main findings: In this review 45 articles were included, of which 25 were for hypertension, 12 for hyperlipidemia and eight for glaucoma. All studies describing the use of ocular hypotensives – and the majority of studies in the other treatment areas – identified inception cohorts of drug initiators. Survival analysis techniques were frequently used to analyse adherence to medication therapy in glaucoma than in the hypertension and hyperlipidemia treatment.

In the treatment of hypertension, use of angiotensin II receptor antagonists or angiotensin-converting enzyme inhibitors was associated with higher levels of adherence. Statins in the treatment of hyperlipidemia, and topical prostaglandins as ocular hypotensive medications, were also associated with higher levels of adherence. Based on findings of included studies, authors concluded that there is a relative persistence of therapy advantage for topical prostaglandins over other drugs used in the treatment of glaucoma.

Authors also noted that the use of survival analysis in future studies of persistence might improve comparability of results across studies. Studying factors that influence adherence to ocular hypotensive medication may identify patient or medication characteristics that impart differential adherence and represent an opportunity for clinical intervention.

Methodology: MEDLINE was searched for English articles from 1990 to 2004, using search terms (discontinu* OR persistenc* OR adheren* OR complian*) AND (pharmacy* OR database* OR claims OR records) AND (drug* OR treatment* OR therap* OR medication* OR prescription* OR refill*). Reference lists of the selected articles were also checked for additional reports.

The following inclusion criteria were used:

(a) Drug(s) used in treatment of glaucoma, hypertension or hyperlipidemia

(b) Data source for pharmacy information was an automated pharmacy database (eg, managed care health plan, Medicare or Medicaid, or similar)

(c) Quantitative results of adherence to medication use were provided

(d) Follow-up duration was 90 days or longer.

Abstracts identified were screened against the inclusion criteria. Details of methods used to estimate adherence or persistence and estimates of persistence with ocular hypotensive therapy, anti-hypertensives and lipid-lowering agents were extracted.

The characteristics of studies and their methodological approach to assessment of adherence, as well as results, were descriptively summarised.

Applicability/external validity: Authors did not assess applicability/external validity.

Geographic focus: This review focused on all countries. However, included studies were conducted in the USA, Canada and Europe.

Summary of quality assessment: Authors conducted a narrative synthesis of the findings, which seemed appropriate due to the vast differences of included studies, such as the difference in study population. Authors appropriately reported inclusion/exclusion criteria used for study selection. However, this review had some important limitations. The authors did not avoid publication and language bias in the search, only one relevant database was searched and only articles written in English were included in the review.

It is not clear from the review if two authors independently screened studies for inclusion and extracted data of included studies, avoiding bias. In addition, authors did not report methods used in quality assessment of included studies, nor did they discuss the quality and risk of bias of included studies. Therefore, there is low confidence in the conclusions about the effects of this study.