Abstract
Purpose
Informal caregivers provide ongoing assistance to a loved one with a health condition. No studies have compared caregiving intensity and perception of burden across chronic medical conditions.
Materials and methods
Databases were searched from inception through 11 September 2020 to identify studies that included the Level of Care Index or the Zarit Burden Inventory (ZBI) among caregivers for people with chronic diseases. Pooled mean ZBI scores and 95% confidence intervals by medical condition were calculated using a random effects model and heterogeneity with I2.
Results
Ninety-seven included articles reported on 98 unique samples across 21 chronic diseases. No study used the Level of Care Index. Among 12 disease groups with more than one study, heterogeneity was too high (I2 range: 0–99.6%, ≥76.5% in 11 groups) to confidently estimate burden. The percent of studies rated high risk of bias ranged from 0% to 98%, but all external validity items were rated as high-risk in >50% of studies.
Conclusions
Findings highlight the need for studies on caregiver burden to improve sampling techniques; better report sampling procedures and caregiver and care recipient characteristics; and develop a standard set of outcomes, including a measure of caregiving intensity. Systematic Review Registration: CRD42017080962
The amount of burden reported by caregivers to loved ones is associated with reduced physical and mental health.
We found considerable heterogeneity in perceived burden reported by informal caregivers across different studies within disease groups, which is likely related to methodological issues, including sampling techniques.
Health care providers who use research on caregiver burden should assess how representative study samples may be and exercise caution in drawing conclusions.
IMPLICATIONS FOR REHABILITATION
Acknowledgements
We thank Angie (Ye Eun) Lee, Nazli Inay, and Lydia Tao for the translation of Korean, Turkish, and Chinese publications.
Disclosure statement
No potential conflict of interest was reported by the author(s). ACJ was supported by a Canadian Institutes of Health Research (CIHR) Canadian Graduate Scholarship. DBR was supported by a CIHR Vanier Graduate Scholarship. BL was supported by a Fonds de recherche du Québec – Santé (FRQS) Postdoctoral Training Fellowship. AB was supported by a FRQS researcher salary award. BDT was supported by a FRQS researcher salary award and a Tier 1 Canada Research Chair. No funding body had any input into any aspect of this review.