Comorbidity between major depression and alcohol use disorder from adolescence to adulthood

Abstract

Background

Limited information exists regarding the long-term development of comorbidity between Major Depressive Disorder (MDD) and Alcohol Use Disorder (AUD; abuse/dependence). Using a representative prospective study, we examine multiple aspects pertaining to MDD + AUD comorbidity, with a focus on the relation between disorders across periods (adolescence, early adulthood, adulthood) and cumulative impairments by age 30.

Method

816 participants were diagnostically interviewed at ages 16, 17, 24, and 30.

Results

Rates of comorbid MDD + AUD were low in adolescence (2%), but increased in early adulthood (11%) and adulthood (7%). Rates of cumulative comorbidity were elevated (21%). Most individuals with a history of MDD or AUD had the other disorder, except for women with MDD. Prospectively, adolescent AUD predicted early adult MDD, while early adult MDD predicted adult AUD. Compared to pure disorders, MDD + AUD was associated with higher risk of alcohol dependence, suicide attempt, lower global functioning, and life dissatisfaction.

Conclusions

Lifetime rates of comorbid MDD + AUD were considerably higher than in cross-sectional studies. Comorbidity was partly explained by bidirectional and developmentally-specific associations and predicted selected rather than generalized impairments. Clinically, our findings emphasize the need to always carefully assess comorbidity in patients with MDD or AUD, taking into account concurrency and developmental timing.

Introduction

Comorbidity between Major Depressive Disorder (MDD) and Alcohol Use Disorder (AUD; abuse or dependence) represents one of the most prevalent and disabling psychiatric combinations in adolescence and adulthood. In the last decades, clinical and epidemiological studies have documented the rates and characteristics of MDD + AUD comorbidity, especially in adulthood [1], [2], [3], [4], [5]. Although longitudinal studies have investigated associations between the two disorders, none has comprehensively examined the development and implications of comorbidity from adolescence to adulthood. Here, we use a prospective sample of participants followed from adolescence to age 30 to investigate several key issues of comorbidity [6] applying to MDD + AUD.

Our first focus is on the relation between the two disorders. Using two complimentary angles, we analyze the temporal ordering and predictions between MDD and AUD across key developmental periods. Evidence on temporal ordering from retrospective studies has been mixed, with studies reporting MDD to most often precede AUD [1], [7], [8], AUD to most often predece MDD [9] or no clear ordering [2], [10]. Similarly, predictions between disorders have been conflicting. Prospective studies have found MDD to predict AUD [8], [11], AUD to predict MDD [12], [13], bidirectional relations [9], [14], [15] and no association between the two disorders after controlling for confounders [16]. A key limitation of the literature is that most evidence comes from adult samples of wide age ranges [8], [9], [11], [14], [15], [16], making it difficult to determine whether associations vary across developmental periods.

Our second focus is on “cumulative comorbidity”, which indexes lifetime rather than point co-occurrence of disorders [17]. We examine the implications of having lifetime patterns with or without overlapping disorders (concurrent vs. successive comorbidity) [6]. Very little information is available to document these types of comorbidity. We also investigate clinical and psychosocial impairments associated with lifetime comorbid MDD + AUD compared to pure MDD and AUD. Greater impairments have typically been reported in comorbid individuals (e.g., more persistent and severe disorders, suicide risk, lower social adjustment) [1], [4], [18], [19], [20], but other studies found no poorer or less adverse outcomes of comorbidity, notably lower AUD severity compared to pure AUD [21], [22]. We also consider whether impairments vary as a function of both concurrency and temporal ordering (MDD-first vs. AUD-first). To our knowledge, no prospective study has previously examined cumulative impairments of MDD + AUD in adulthood.

We use data from the Oregon Adolescent Depression Project (OADP) [23] to address these questions. The OADP is unique in combining 4 prospective assessment waves from adolescence to adulthood paired with retrospective recall between assessments, providing diagnostic coverage up to age 30. This design compares favorably to most prospective studies, which have more limited coverage, as well as retrospective studies, which are impacted by long-term recall bias. In keeping with the new DSM-5 [24], we investigate AUD as a single disorder, rather than abuse and dependence as separate disorders. We do, however, consider alcohol dependence as a marker of AUD severity.