Phenotypic and genetic markers of psychopathology in a population-based sample of older adults

Abstract

Although psychiatric phenotypes are hypothesized to organize into a two-factor internalizing-externalizing structure, few studies have evaluated the structure of psychopathology in older adults, nor explored whether genome-wide polygenic scores (PGSs) are associated with psychopathology in a domain-specific manner. We used data from 6003 individuals of European ancestry from the Health and Retirement Study, a large population-based sample of older adults in the United States. Confirmatory factor analyses were applied to validated measures of psychopathology and PGSs were derived from well-powered genome-wide association studies (GWAS). Genomic SEM was implemented to construct latent PGSs for internalizing, externalizing, and general psychopathology. Phenotypically, the data were best characterized by a single general factor of psychopathology, a factor structure that was replicated across genders and age groups. Although externalizing PGSs (cannabis use, antisocial behavior, alcohol dependence, attention deficit hyperactivity disorder) were not associated with any phenotypes, PGSs for major depressive disorder, neuroticism, and anxiety disorders were associated with both internalizing and externalizing phenotypes. Moreover, the variance explained in the general factor of psychopathology increased by twofold (from 1% to 2%) using the latent internalizing or latent one-factor PGSs, derived using weights from Genomic Structural Equation Modeling (SEM), compared with any of the individual PGSs. Collectively, results suggest that genetic risk factors for and phenotypic markers of psychiatric disorders are transdiagnostic in older adults of European ancestry. Alternative explanations are discussed, including methodological limitations of GWAS and phenotypic measurement of psychiatric outcome in large-scale population-based studies.

Fig. 1. Within- and across-domain correlations among phenotypes in the Health and Retirement Study.

5873 < N < 5965. Associations that were not significant at p < 0.05 are marked with an “X”.

Fig. 2. High correlation between internalizing and externalizing factors suggests a one-factor model of psychopathology among older adults in the Health and Retirement Study.

INT internalizing, EXT externalizing. Standardized estimates are shown. A, B Confirmatory one-factor (model fit: χ²(9) = 70.37, p < 0.001; CFI = 0.980; TLI = 0.967; RMSEA = 0.052, 90% CI [0.041, 0.064]) and two-factor (model fit: χ²(8) = 46.71, p < 0.001; CFI = 0.988; TLI = 0.977; RMSEA = 0.044, 90% CI [0.032, 0.056]) phenotypic models in the test sample (n = 3002). C Confirmatory one-factor phenotypic model in the hold-out sample (n = 3001; model fit: χ²(9) = 61.96, p < 0.001; CFI = 0.983; TLI = 0.972; RMSEA = 0.048, 90% CI [0.037, 0.059]).

Fig. 3. Polygenic scores for internalizing, but not externalizing, disorders are associated with internalizing and externalizing behaviors in the Health and Retirement Study.

N = 3001. Associations between polygenic scores (PGS) and phenotypic outcomes, accounting for the top 10 ancestry principal components. Estimates are unstandardized and error bars are standard errors. A Individual PGSs as predictors; B Latent PGSs, where SNP weights were estimated using Genomic SEM. In both panels, error bars are standard errors around the estimate.

Fig. 4. Genomic SEM one-factor and two-factor model.

Confirmatory factor analyses were conducted on the GWAS summary statistics in Table 1, using the Genomic SEM package in R Statistical Software (Grotzinger et al., 2019). Standardized estimates are shown. See Supplemental Fig. 3 for unstandardized estimates. In both the one-factor and two-factor models, the residual variance of MDD was fixed to zero. Model fit comparisons between the one-factor model (χ²(14) = 76.762, p < 0.001, AIC = 104.762, CFI = 0.962, SRMR = 0.127) and two-factor model (χ²(13) = 46.072, p < 0.001, AIC = 76.072, CFI = 0.980, SRMR = 0.084) indicated superior model fit of the two-factor model (Δχ² = 30.69(1), p < 0.001, ΔCFI > 0.01, lower AIC). Single-nucleotide polymorphism effects were then integrated into the model to derive new SNP weights for the construction of latent polygenic scores (see Supplemental Methods).