Does the moderator matter? Identification of multiple moderators of the association between peripheral inflammatory markers and depression severity in a large racially diverse community cohort

On the pursuit of reproducibility: the importance of large sample sizes in psychoimmunology

Peripheral inflammatory markers (PIMs), such as C-reactive protein (CRP) or white blood cell count (WBC), have been associated with depression severity in meta-analyses and large cohort studies. However, in typically-sized psychoimmunology studies (N < 200) that explore associations between PIMs and neurobiological/psychosocial constructs related to depression and studies that examine less-studied PIMs (e.g., interferon gamma), significant concerns about reproducibility of results exist. For the well-characterized association between PIMs (CRP/WBC) and depression severity, we examined statistical errors as a function of sample size in a large community cohort (n = 24,550). We further assessed how statistical errors varied as related to analytic decisions (e.g., number of covariates) and characteristics related to study design (e.g., relationships within subgroups of patients). Only large samples (e.g., n = 1000 to n = 10,000) were sufficiently powered to detect PIM-depression associations and minimized overestimation of effect sizes (e.g., effect size inflation), and greater sample sizes were required as more covariates were included in analytic models. Moderately sized samples (n > 500) generally ensured the correct directionality of effect sizes (e.g., low rates of sign reversal). Sample sizes required for 80% power also varied widely depending on study design characteristics (e.g., N = 350 to N = 10,000+). Typically-sized psychoimmunology studies examining PIM-depression associations (N < 200) are likely underpowered and at high risk of overestimation of effect sizes. Study design characteristics also notably influence power and statistical error rates. Use of large sample sizes (e.g., N > 7000) and consideration of analytic decisions (e.g., number/choice of covariates) will maximize reproducibility of psychoimmunology studies related to depression to enhance development of treatments for depression or to help understand pathophysiological mechanisms of depression.

Replicable and robust cellular and biochemical blood marker signatures of depression and depressive symptoms

BACKGROUND

Identification of replicable and robust peripheral blood-based markers associated with depression remains elusive, given that studies frequently identify potential biomarkers that ultimately fail to replicate in other studies, impeding progress in psychiatric research. Peripheral biochemical and cellular markers (PBCs; e.g., albumin) may play an important role in depression.

METHODS

Using a test-replication design including participants from the NHANES community cohort (ntest=17,450, nreplication=17,449), we examined 42 PBCs to identify PBCs that were both replicably and robustly associated with either overall depression severity or individual symptoms of depression across both cohorts across a wide range of possible combinations of analytic decisions (n's = 17,000+).

RESULTS

We found that a small set of PBCs (e.g., bilirubin) were robustly and replicably associated with overall depression severity, with unique signatures of PBCs linked with individual symptoms of depression when stratified by gender. A varying degree of correlation was found between measures of replicability.

CONCLUSIONS

We identified replicable and robust cellular biochemical blood marker signatures associated with both overall depression severity and individual symptoms of depression. Our findings can be used to enhance other researchers' abilities to better understand factors associated with depression and potentially drive the development of effective treatments for depression.

Exploring the multiverse: the impact of researchers' analytic decisions on relationships between depression and inflammatory markers

In recent years, a replication crisis in psychiatry has led to a growing focus on the impact of researchers' analytic decisions on the results from studies. Multiverse analyses involve examining results across a wide array of possible analytic decisions (e.g., log-transforming variables, number of covariates, or treatment of outliers) and identifying if study results are robust to researchers' analytic decisions. Studies have begun to use multiverse analysis for well-studied relationships that have some heterogeneity in results/conclusions across studies.We examine the well-studied relationship between peripheral inflammatory markers (PIMs; e.g., white blood cell count (WBC) and C-reactive protein (CRP)) and depression severity in the large NHANES dataset (n = 25,962). Specification curve analyses tested the impact of 9 common analytic decisions (comprising of 58,000+ possible combinations) on the association of PIMs and depression severity. Relationships of PIMs and total depression severity are robust to analytic decisions (based on tests of inference jointly examining effect sizes and p-values). However, moderate/large differences are noted in effect sizes based on analytic decisions and the majority of analyses do not result in significant findings, with the percentage of analyses with statistically significant results being 46.1% for WBC and 43.8% for CRP. For associations of PIMs with specific symptoms of depression, some associations (e.g., sleep, appetite) in males (but not females) were robust to analytic decisions. We discuss how multiverse analyses can be used to guide research and also the need for authors, reviewers, and editors to incorporate multiverse analyses to enhance replicability of research findings.

Racial Differences in C-Reactive Protein, Depression Symptoms, and Social Relationships in Older Adults: A Moderated Network Analysis

INTRODUCTION

We introduce moderated network analysis as an integrative approach to assess the moderation effects of race on the relationship between C-reactive protein (CRP) and depression symptoms in older adults. This study further explores how the observed relationships differ adjusting for social relationships.

METHODS

This secondary analysis of cross-sectional data from the National Social Life, Health, and Aging Project (2010-2011) includes 2,880 older adults. We used different depression symptom domains (depressed affect, low positive affect, somatic symptoms, and interpersonal problems) from the Center for Epidemiologic Studies-Depression Scale. Social relationships were assessed with measures of social integration, social support, and social strain. The moderated networks were constructed using the R-package mgm. The racial moderator was coded as White/African American racial groups.

RESULTS

In the moderated networks of CRP and depression symptoms, CRP-"interpersonal problems" edge was present only among African Americans. CRP-"somatic symptoms" edge was present in both racial groups with equal edge weights. After adjusting for social relationships, the aforementioned patterns remained the same, but the edge weights were attenuated. We additionally observed CRP-social strain and social integration-"depressed affect" edges only in African Americans.

DISCUSSION

Race may moderate the relationship between the CRP and depression symptoms in older adults and social relationships might be important covariates to consider while analyzing them. This study as an initiation point; future network investigations would benefit from leveraging more contemporary cohorts of older adults, gaining a large sample size with diverse racial/ethnic backgrounds, and important covariates. Several important methodological issues of the current study are addressed.

Does the moderator matter? Identification of multiple moderators of the association between peripheral inflammatory markers and depression severity in a large racially diverse community cohort

Depressive symptomology has been linked to low-grade peripheral inflammatory markers (PIMs), specifically C-reactive protein (CRP) and white blood cell count (WBC). However, such associations may be affected by multiple moderators (including race/ethnicity), though few well-powered and racially diverse studies have examined this. We examined 31 moderators of PIM-depression relationships in a large racially diverse cohort (n = 21,570). We also examined if associations between PIM and depression severity were dependent on clinical cutpoints for moderate depressive symptoms and elevated CRP. We found several positive moderators of PIM-depression relationships for both WBC and CRP: ongoing medication use (antidepressant, statin, or any prescription drug), presence of sleep concerns, and poor health status (β's = 0.06-0.21, p's < 0.05). For both WBC and CRP, individuals of non-Hispanic White race/ethnicity were found to have stronger PIM-depression associations overall relative to minoritized groups (B's = 0.14 to 1.01, p's < 0.05). For CRP, stronger PIM-depression relationships existed for individuals with moderate (or greater) depression severity or elevated CRP (B's = 0.27 to 0.49, p's < 0.05). Thus, a wide range of moderators appears to affect PIM-depression associations. These results could help identify participants with strong coupling of PIM-depression severity, to guide future research and personalized treatments for depression and to indicate gaps in the applicability of widely referenced theoretical models among racial/ethnic minoritized groups.