Composite diagnostic criteria are problematic for linking potentially distinct populations: the case of frailty

Abstract

Composite diagnostic criteria are common in frailty research. We worry distinct populations may be linked to each other due to complicated criteria. We aim to investigate whether distinct populations might be considered similar based on frailty diagnostic criteria. The Functional Domains Model for frailty diagnosis included four domains: physical, nutritive, cognitive and sensory functioning. Health and Retirement Study participants with two or more deficiencies in the domains were diagnosed frail. The survival distributions were analyzed using discrete-time survival analysis. The distributions of the demographic characteristics and survival across the groups diagnosed with frailty were significantly different (p < 0.05). A deficiency in cognitive functioning was associated with the worst survival pattern compared with a deficiency in the other domains (adjusted p < 0.05). The associations of the domains with mortality were cumulative without interactions. Cognitive functioning had the largest effect size for mortality prediction (Odds ratios, OR = 2.37), larger than that of frailty status (OR = 1.92). The frailty diagnostic criteria may take distinct populations as equal and potentially assign irrelevant interventions to individuals without corresponding conditions. We think it necessary to review the adequacy of composite diagnostic criteria in frailty diagnosis.

Figure 1

The characteristics of the 16 population groups categorized based on the four frailty domains of the Functional Domains Model. Note: the 16 groups are based on the combinations of presence or absence of deficiencies in the four frailty domains. The groups are labelled with serial numbers from one to 16 and code names that identify the deficiencies in the frailty domains. The four domains of the Functional Domains Model for the diagnosis of frailty are (1) physical functioning, (2) nutritive functioning, (3) cognitive functioning, and (4) sensory problems. Individuals with two or more deficiencies in the four domains are considered frail. For example, the code, 0011, represents the absence of deficiencies in the physical functioning and nutritive functioning domains with the presence of deficiencies in the cognitive functioning and sensory problems domains. The groups with one deficiencies in the four domains are in yellow shade and those with three deficiencies are in red shade.

Figure 2

Regression coefficients of the four frailty domains for estimating individual characteristics. (a) coefficients for age in years. (b) odds ratios for female proportions. (c) odds ratios for the proportions of white or Caucasian. (d) coefficients for education in years. (e) coefficients for per capita income. (f) coefficients for per capita wealth. Note: those without any deficiencies in the four frailty domains were the baseline group for comparison. The coefficients were estimated via linear regression and the four frailty domains were the independent variables. The odds ratios were estimated via logistic regression with the four frailty domains as independent variables. The ranges are 95% confidence intervals. *significantly different from the association with Domain 3) cognitive functioning.

Figure 3

The correlations between the four frailty domains in the Functional Domains Model. Numbers in the cells = Spearman’s correlation coefficients, ranging from 1 to −1; (1) physical = physical functioning domain in the Functional Domains Model; (2) nutritive = nutritive functioning domain; (3) cognitive = cognitive functioning domain; (4) sensory = sensory problems domain.

Figure 4

The survival curves of the 16 groups categorized based on the combinations of the four frailty domains in the Functional Domains Model. (a) Groups diagnosed with one deficiency in the four frailty domains. (b) Groups diagnosed with two deficiencies in the four frailty domains. (c) Groups diagnosed with three deficiencies in the four frailty domains. Note: p values derived from log-rank tests to examine the null hypothesis that there was no difference in the survival distributions of the eligible groups. The eligible groups are those labelled in the subtitles and two groups, those not diagnosed with any frailty domains (0000) and those diagnosed with all four frailty domains (1111), are not eligible for respective tests. Population groups are labelled with serial numbers and the codes that represent the combinations of the four frailty domains in the Functional Domains Model. Zero represents the absence of the frailty domains and one the presence. The frailty domains are in the following sequences: physical functioning, nutritive functioning, cognitive functioning, and sensory problems. For example, the code, 0011, represents the absence of domain (1) physical functioning and domain (2) nutritive functioning and the presence of domain (3) cognitive functioning and domain (4) sensory problems.

Figure 5

The p values of the comparisons in the survival distributions of the population groups categorized according to the 16 combinations of the four frailty domains in the Functional Domains Model. (a) Groups diagnosed with one deficiency in the four frailty domains. (b) Groups diagnosed with two deficiencies in the four frailty domains. (c) Groups diagnosed with three deficiencies in the four frailty domains. (d) Groups diagnosed with two or more deficiencies in the four frailty domains. Note: grey cells = adjusted p < 0.05. The p values derived from log-rank tests to examine the null hypothesis that there is no difference in the survival distributions of the eligible groups. The p values were adjusted for multiple comparisons. The blank cells represent insignificant survival differences. Population groups are labelled with serial numbers and the codes that represent the combinations of the four frailty domains in the Functional Domains Model. Zero represents the absence of the frailty domains and one the presence. The frailty domains are in the following sequences: physical functioning, nutritive functioning, cognitive functioning, and sensory problems. For example, the code, 0011, represents the absence of physical functioning and nutritive functioning domains with the presence of cognitive functioning and sensory problems domains.

Figure 6

The regression coefficients of the comparisons in the survival distributions of the population groups categorized according to the 16 combinations of the four frailty domains in the Functional Domains Model. (a) Groups diagnosed with one deficiency in the four frailty domains. (b) Groups diagnosed with two deficiencies in the four frailty domains. (c) Groups diagnosed with three deficiencies in the four frailty domains. (d) Groups diagnosed with two or more deficiencies in the four frailty domains. Note: regression coefficients derived from discrete-time survival analysis to examine the null hypothesis that there is no difference in the survival distributions of the eligible groups, while demographic characteristics are adjusted, including age, sex, race, years of education, per capita income and per capita wealth. The blank cells represent insignificant survival differences. The groups to the left of the graph were the baseline. If the coefficients are negative, the group to the left of the graph has higher mortality risk. Population groups are labelled with serial numbers and the codes that represent the combinations of the four frailty domains in the Functional Domains Model. Zero represents the absence of the frailty domains and one the presence. The frailty domains are in the following sequences: physical functioning, nutritive functioning, cognitive functioning, and sensory problems. For example, the code, 0011, represents the absence of physical functioning and nutritive functioning domains with the presence of cognitive functioning and sensory problems domains.

Figure 7

Odds ratios of the frailty index, frailty status, frailty domains, the interaction terms, and 16 frailty groups of frailty domains based on the Functional Domains Model for mortality prediction. (a) odds ratios of frailty index, frailty status, frailty domains, and the interaction terms of the frailty domains. (b) Odds ratios of frailty index, frailty status, frailty domains, and the interaction terms of the frailty domains. Note: odds ratios greater than one suggesting higher risk of mortality; less than one suggesting lower risk. The ranges represent 95% confidence intervals.