Understanding the family care dyad: A comparison of two multilevel models

Abstract

Although there has been increasing attention on a dyadic perspective of illness, contemporary dyadic research methods are still rarely utilized in cardiovascular disease. The focus of this paper is to describe the advantages of two types of multilevel dyadic models (the matched pairs model and the lesser known incongruence model). Data exemplars in a sample of heart failure family dyads are used to illustrate the distinct advantages of these two related multilevel dyadic models with particular emphasis on alignment with research questions. The more commonly known matched pairs model examines separate outcomes for each member of the dyad, controlling for the interdependent nature of the data. By re-parameterizing this model into a univariate dyadic outcomes model, researchers can address distinct, and sometimes more appropriate, research questions (e.g. incongruent appraisals of the illness experience). This paper promotes greater application of these methods in cardiovascular research to further understanding of the dyadic experience and more appropriately target interventions.

A Multilevel Modeling Approach to Examine Incongruent Illness Appraisals in Family Care Dyads Over Time

“…Covariates (at both the individual and dyad levels) can be added to explain variability around average estimates of dyadic mean appraisal and dyadic incongruence in appraisal of patient pain interference. This cross-sectional approach to incongruence has been well described across illness contexts to examine dyadic incongruence in appraisals of patient symptoms, barriers to care, care values, and confidence to manage illness to name a few (Cano et al, 2005; Lyons, Sadowski & Lee, 2002; Lyons & Lee, 2020b; Miller et al, 2018; Reamy et al, 2011; Sayer & Klute, 2005).…”

“…Moreover, these models examine actor and partner effects on the health of both members of the care dyad to capture the transactional nature of the dyadic relationship (i.e., how members of the dyad influence each other) and have shed important light on the transactional nature of such conditions as dementia, cancer, heart failure, stroke, Parkinson’s disease, diabetes, and military couples to name a few (Bonds et al, 2020; Burns, 2020; Hooker et al, 2018; Lyons, Sadowski & Lee, 2020; Lyons, Winters-Stone, et al, 2016; Mavandadi et al, 2014; Miller et al, 2019; Moon et al, 2017; Oh & Ryu, 2019; Pucciarelli et al, 2017; Vellone et al, 2018; Zamir et al, 2017). Much has been written over the last 15 years regarding the advantages and nuances of MLM in dyadic research (Garcia et al, 2015; Gonzalez & Griffin, 1997; Kenny, 2018; Kenny et al, 2006; Lyons & Lee, 2020b; Lyons & Sayer, 2005; Rayens & Svavarsdottir, 2003; Sayer & Klute, 2005); few studies have employed an MLM approach to dyadic incongruence.…”

One year of methods corner: the way forward to innovate research in cardiovascular care

“…In the first year of the Methods Corner, seven papers have been published. [4][5][6][7][8][9][10] Six of these articles addressed quantitative research techniques. 4,5,[7][8][9][10] In the January 2020 issue, propensity weighting was discussed.…”

Chronic illness as a ‘family disease’: The need for appropriate scientific methods for dyadic research

“…14 One method is the matched pairs model, and the other is the incongruence model. 14 In sum, as the importance of the role of caregivers and families is increasingly acknowledged, it is critical that researchers apply appropriate statistical techniques that account for the interdependency between patients and their caregivers. Indeed, researchers should employ contemporary dyadic research methods in order to make their research truly dyadic.…”