Affective disorders and chronic inflammatory conditions: analysis of 1.5 million participants in Our Future Health

Abstract

Background: Chronic inflammation is associated with psychiatric disorders. If inflammation is linked mechanistically to mental health, people living with chronic inflammatory conditions may experience mental health issues at higher rates than others.

Objective: To test this hypothesis, we analysed data from 1 563 155 adults living in the UK within the newly launched UK-wide Our Future Health research cohort.

Methods: Participants were split between two groups: people with self-reported lifetime diagnoses of six autoimmune conditions (n=37 808) and those without these diagnoses (n=1 525 347).

Findings: Lifetime prevalence (95% CI) of self-reported lifetime diagnoses of any affective disorder (depression, bipolar disorder, anxiety) was significantly higher (p<0.001) among people with autoimmune conditions (28.8% (28.4% to 29.3%)) than in the general population (17.9% (17.8% to 18.0%)), with similar trends observed for individual affective disorders. Prevalence of current depressive symptoms (9-item Patient Health Questionnaire (PHQ-9) ≥10, 31.7% vs 23.4%) and current anxiety symptoms (7-item Generalised Anxiety Disorder Scale (GAD-7) ≥8, 28.1% vs 21.6%) was also higher among people with autoimmune conditions. Odds of experiencing affective disorders, calculated using logistic regression models, were significantly higher in this group compared with the general population (OR (95% CI) = 1.86 (1.82 to 1.90), p<0.001), and these odds remained elevated when adjusting for the effects of age, sex, ethnicity (OR=1.75 (1.71 to 1.79), p<0.001) and additionally, for household income, parental history of affective disorders, chronic pain status and frequency of social interactions (OR=1.48 (1.44 to 1.52), p<0.001).

Conclusions: Overall, the risk of affective disorders among people living with autoimmune conditions was nearly twice that of the general population.

Clinical implications: Although the observational design of this study does not allow for direct inference of causal mechanisms, this analysis of a large national dataset suggests that chronic exposure to systemic inflammation may be linked to a greater risk for affective disorders. Future work should seek to investigate potential causal mechanisms for these associations.

Keywords: Anxiety disorders; Depression; Depression & mood disorders.

Figure 1. Prevalence with 95% confidence intervals (CI) of self-reported lifetime diagnoses of any affective disorder, of each of these disorders separately, and of current depression (PHQ-9 score ≥10) and current anxiety (GAD-7 score ≥8). Prevalence estimates are shown for participants with at least one of six autoimmune conditions (any autoimmune), those with each of these six autoimmune conditions separately, and those with no autoimmune conditions (none) representing the general population. Horizontal dashed lines mark the prevalence of each affective disorder in the any autoimmune (blue lines) and none (red lines) groups for ease of comparison with other groups. GAD-7, 7-item Generalised Anxiety Disorder Scale; IBD, inflammatory bowel disease; PHQ-9 9-item Patient Health Questionnaire.

Figure 2. Prevalence with 95% confidence intervals (CI) of self-reported lifetime diagnoses of affective disorders and of current depression (PHQ-9 score ≥10) and current anxiety (GAD-7 score ≥8), separately for participants self-identifying as male and female. Prevalence estimates are shown for participants with at least one of six autoimmune conditions (any autoimmue), those with each of these six autoimmune conditions separately, and those with no autoimmune conditions (none) representing the general population. GAD-7, 7-item Generalised Anxiety Disorder Scale; IBD, inflammatory bowel disease; PHQ-9 9-item, 9-item Patient Health Questionnaire.

Figure 3. OR with 95% confidence intervals (CI) for experiencing affective disorders among people with any autoimmune disorder compared with the general population. ORs for any affective disorder, depression, bipolar disorder and anxiety were calculated using self-reported lifetime diagnoses, whereas ORs for current depression and current anxiety were calculated using total scores on the PHQ-9 and GAD-7, respectively. Model 1 estimated the unadjusted OR, model 2 was adjusted for age, sex and ethnicity, and model 3 was adjusted for age, sex, ethnicity, household income, parental history of psychiatric illness, chronic pain experience (true/false) and frequency of social interactions. Horizontal dashed line represents OR=1.00. GAD-7, 7-item Generalised Anxiety Disorder Scale; PHQ-9 9-item, 9-item Patient Health Questionnaire.