Clinical samples often have the disadvantages of oversampling care-recipients who are relatively more impaired and caregivers who are seeking services. However, it is often necessary to use such clinical samples to recruit caregivers of well-characterized care-recipients with specific conditions (e.g. Alzheimer’s) and to add specific instruments and questions to a project. One should be cautious when generalizing from findings based on caregiver samples recruited from clinics to the experience of those caregivers living in the community who may not utilize services that are the source of the sample.
With population-based studies, there is often much less detail available about the circumstances of caregiving (e.g. the diseases and disabilities of the care-recipient, how much and what kind of care is provided, and the availability of social support) but the samples may be more representative.
Comparisons of caregivers and non-caregivers have relied heavily on convenience sampling, which involves recruiting caregivers and non-caregivers in different ways. For example, a clinical or convenience sample of burdened caregivers might be recruited from a local clinical or support group, while a sample of healthy non-caregivers is recruited through a community program or bulletin board. Comparison groups obtained in this way may not represent their respective populations and may produce results that are limited in generalizability. Population-based sampling involves recruiting caregivers and non-caregivers with the same methods and may result in comparison groups that better represent their respective populations.
Compared to population-based studies, studies using convenience sampling methodology more commonly find negative mental health effects of caregiving (Pruchno et al., 2008). Pinquart and Sörensen (2003) found that although rates of depressive symptoms were indeed higher among caregivers, the effects sizes were small to medium and varied as a function of research design and sample characteristics. Across all studies, caregivers had depressive symptoms 0.58 standard deviations higher than non-caregivers; in non-convenience samples, the effect size dropped to 0.26. Caregivers from convenience samples also provided 9.9 hours of care more per week, indicating potentially higher levels of stressors (Pinquart & Sörensen, 2003).
References
Pinquart, M., & Sörensen, S. (2003). Differences between caregivers and noncaregivers in psychological health and physical health: A meta-analysis. Psychology and Aging, 18(2), 250-267. doi:10.1037/0882-7974.18.2.250
Pruchno, R. A., Brill, J. E., Shands, Y., Gordon, J. R., Genderson, M. W., Rose, M., & Cartwright, F. (2008). Convenience samples and caregiving research: How generalizable are the findings? The Gerontologist, 48, 820-827. doi:10.1093/geront/48.6.820
Roth, D. L., Sheehan, O. C., Haley, W. E., Jenny, N. S., Cushman, M., & Walston, J. D. (2019). Is family caregiving associated with inflammation or compromised immunity? A meta-analysis. The Gerontologist, [Advance online publication]. doi:10.1093/geront/gnz015