Publication:
Intra-cluster correlation coefficients in primary care patients with type 2 diabetes and hypertension

Simple item page
dc.contributor.authorYi Lin Lee
dc.contributor.authorYvonne Mei Fong Lim
dc.contributor.authorKian Boon Law
dc.contributor.authorSheamini Sivasampu
dc.date.accessioned2024-07-30T07:58:58Z
dc.date.available2024-07-30T07:58:58Z
dc.date.issued2020
dc.description.abstractIntroduction: There are few sources of published data on intra-cluster correlation coefficients (ICCs) amongst patients with type 2 diabetes (T2D) and/or hypertension in primary care, particularly in low- and middle-income countries. ICC values are necessary for determining the sample sizes of cluster randomized trials. Hence, we aim to report the ICC values for a range of measures from a cluster-based interventional study conducted in Malaysia. Method: Baseline data from a large study entitled Evaluation of Enhanced Primary Health Care interventions in public health clinics (EnPHC-EVA: Facility) were used in this analysis. Data from 40 public primary care clinics were collected through retrospective chart reviews and a patient exit survey. We calculated the ICCs for processes of care, clinical outcomes and patient experiences in patients with T2D and/or hypertension using the analysis of variance approach. Results: Patient experience had the highest ICC values compared to processes of care and clinical outcomes. The ICC values ranged from 0.01 to 0.48 for processes of care. Generally, the ICC values for processes of care for patients with hypertension only are higher than those for T2D patients, with or without hypertension. However, both groups of patients have similar ICCs for antihypertensive medications use. In addition, similar ICC values were observed for clinical outcomes, ranging from 0.01 to 0.09. For patient experience, the ICCs were between 0.03 (proportion of patients who are willing to recommend the clinic to their friends and family) and 0.25 (for Patient Assessment of Chronic Illness Care item 9, Given a copy of my treatment plan). Conclusion: The reported ICCs and their respective 95% confidence intervals for T2D and hypertension will be useful for estimating sample sizes and improving efficiency of cluster trials conducted in the primary care setting, particularly for low- and middle-income countries.
dc.identifier.doi10.1186/s13063-020-04349-4
dc.identifier.urihttps://trialsjournal.biomedcentral.com/articles/10.1186/s13063-020-04349-4
dc.identifier.urihttps://repository.nih.gov.my/handle/123456789/754
dc.language.isoen
dc.publisherBMC
dc.relation.ispartofTrials
dc.relation.issn1745-6215
dc.relation.journalTrials
dc.subjectIntra-cluster correlation
dc.subjectType 2 diabetes
dc.subjectHypertension
dc.subjectPrimary care
dc.subjectSample size
dc.titleIntra-cluster correlation coefficients in primary care patients with type 2 diabetes and hypertension
dc.typejournal-article
dspace.entity.typePublication
oaire.citation.issue1
oaire.citation.volume21
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
NIH20200183.pdf
Size:
954.24 KB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Selected Article of Journal (ICR)