Menzies, R. E., Richmond, B., Sharpe, L., Skeggs, A., Liu, J., & Coutts-Bain, D. (2024). The ‘revolving door’ of mental illness: A meta-analysis and systematic review of current versus lifetime rates of psychological disorders. British Journal of Clinical Psychology, 63, 178–196. https://doi.org/10.1111/bjc.12453
Key Points
- The meta-analysis found that rates of lifetime mental disorders were nearly double (1.84 times) those of current disorders, providing evidence for the “revolving door” phenomenon of mental illness.
- Treatment history moderated this relationship, with a smaller ratio of lifetime to current disorders in samples with >90% treatment rates compared to samples with minimal reported treatment. However, treatment did not entirely prevent the development of new disorders.
- Anxiety disorders typically preceded other disorders like mood and substance use disorders, except for panic disorder which usually developed after another anxiety disorder.
- The findings highlight the need for transdiagnostic treatments targeting shared underlying features, increased accessibility of psychological interventions, and a broader conceptualization of relapse prevention.
Rationale
Therapists have long observed the “revolving door” of mental health services, where individuals often develop, seek treatment for, and recover from multiple mental illnesses across their life (Iverach et al., 2014).
This phenomenon suggests that people who have had one disorder are likely to present with symptoms of multiple disorders across their lifespan, spurring explanations like dimensional approaches to mental illness, such as the Hierarchical Taxonomy of Psychopathology (HiTOP) model (Kotov et al., 2017).
However, this observation has not been empirically tested through a meta-analysis.
The present study addressed this gap, examining the mean number of current versus lifetime mental illnesses to determine if the “revolving door” could be accounted for purely by comorbidity or if people sequentially develop different disorders.
Method
The meta-analysis followed PRISMA guidelines and searched MEDLINE, PsycINFO, and Web of Science databases using terms like “lifetime disorders,” “mental health,” and “psychopathology.”
Inclusion criteria were: reporting numbers of current and past mental health diagnoses, adult participants, use of DSM-IV or later diagnostic criteria, and English language.
Two authors independently screened studies, and disagreements were resolved by consensus. Risk of bias was assessed using the Joanna Briggs Institute checklist for prevalence studies.
Search strategy and terms
The initial search yielded 5817 articles, with 1401 duplicates removed. In Step 1, four authors independently screened titles and abstracts of 4416 studies. In Step 2, two authors independently screened the full text of 1245 manuscripts, with almost perfect inter-rater reliability (κ = 0.87).
This process resulted in 38 studies included in the review, with 27 having sufficient quantitative data for meta-analysis.
Inclusion and exclusion criteria
Studies were included if they reported numbers of current and past mental health diagnoses in adults, used DSM-IV or later diagnostic criteria, and were published in English after 1994.
Reviews, case studies, dissertations, abstracts, and conference presentations were excluded.
Statistical measures
Comprehensive Meta-Analysis Version 3 was used for statistical analyses. A random-effects model was used due to significant heterogeneity across studies (Q35 = 772.17, p < .0001, I2 = 95.47).
Subgroup analyses and meta-regressions explored the impact of sample characteristics and diagnostic tools on the relationship between current and lifetime disorders.
Results
- Across 27 studies with 23,402 participants, the mean number of current disorders was 1.53 (SD = 1.02), while the mean number of lifetime disorders was 2.57 (SD = 1.77).
- Meta-analysis revealed that the average number of lifetime disorders was 1.84 times that of current disorders (95% CI [1.72, 1.96], p < .0001).
- Treatment history significantly moderated this relationship, with a smaller ratio in samples with >90% treatment rates (1.52 times, 95% CI [1.34, 1.68]) compared to samples with minimal reported treatment (1.99 times, 95% CI [1.76, 2.25]). Sample type (clinical vs. non-clinical) and diagnostic tool did not moderate the relationship.
- A systematic review of 12 studies indicated that anxiety disorders typically preceded other disorders, except for panic disorder, which usually developed after another anxiety disorder. Depressive, eating, and trauma-related disorders were often secondary to other diagnoses, while impulse control and bipolar disorders tended to present first.
Insight
This meta-analysis provides the first empirical evidence for the “revolving door” phenomenon, suggesting that individuals are likely to experience multiple different mental health conditions across their lifespan.
The findings extend previous research by quantifying the ratio of lifetime to current disorders and identifying treatment history as a moderating factor.
While treatment appears to partially reduce the likelihood of future disorders developing, it does not entirely prevent this risk.
The study also offers insights into common temporal sequences of disorders, with anxiety disorders often preceding mood and substance use disorders.
Strengths
- Strict inclusion criteria ensuring the use of structured diagnostic interviews.
- Wide scope of studies screened, and diverse populations from various countries included in the final meta-analysis.
- Using two independent reviewers for study selection and risk of bias assessment also enhances the reliability of the findings.
Limitations
- Not all studies included all disorders, particularly common childhood disorders and less frequently assessed conditions like eating and personality disorders, potentially leading to an underestimation of the true ratio of lifetime to current disorders.
- The majority of included studies were cross-sectional rather than longitudinal, raising the possibility of inaccurate recall selectively affecting estimated rates of lifetime disorders. However, longitudinal studies excluded due to reporting prevalence rates showed similar patterns, somewhat allaying these concerns. The cross-sectional designs also limit conclusions regarding causality and the identification of shared etiologies across lifetime disorders.
Clinical Implications
The results of this meta-analysis have significant implications for clinical psychology practice.
They underscore the importance of transdiagnostic treatments targeting overlapping constructs common to numerous disorders, consistent with recommendations from dimensional approaches like HiTOP
Increasing the accessibility of psychological treatments through stepped care models, such as offering online self-help interventions for mild to moderate symptoms, may be crucial to meet the growing demand for mental health services.
A greater emphasis on relapse prevention, not only for the individual’s current disorder but also for future disorders, should be a priority in treatment. Scheduling follow-up appointments or encouraging the use of smart technologies for symptom self-monitoring post-treatment may prove beneficial.
References
Primary reference
Menzies, R. E., Richmond, B., Sharpe, L., Skeggs, A., Liu, J., & Coutts-Bain, D. (2024). The ‘revolving door’ of mental illness: A meta-analysis and systematic review of current versus lifetime rates of psychological disorders. British Journal of Clinical Psychology, 63, 178–196. https://doi.org/10.1111/bjc.12453
Iverach, L., Menzies, R. G., & Menzies, R. E. (2014). Death anxiety and its role in psychopathology: Reviewing the status of a transdiagnostic construct. Clinical Psychology Review, 34, 580–593. https://doi.org/10.1016/j.cpr.2014.09.002
Kotov, R., Krueger, R. F., Watson, D., Achenbach, T. M., Althoff, R. R., Bagby, R. M., Brown, T. A., Carpenter, W. T., Caspi, A., Clark, L. A., Eaton, N. R., Forbes, M. K., Forbush, K. T., Goldberg, D., Hasin, D., Hyman, S. E., Ivanova, M. Y., Lynam, D. R., Markon, K., … Zimmerman, M. (2017). The hierarchical taxonomy of psychopathology (HiTOP): A dimensional alternative to traditional nosologies. Journal of Abnormal Psychology, 126(4), 454–477. https://doi.org/10.1037/abn0000258
Keep Learning
- How might the “revolving door” phenomenon inform our understanding of the etiology and course of mental disorders?
- What are some potential transdiagnostic constructs that could be targeted in treatment to address the development of multiple disorders across the lifespan?
- How can mental health services be restructured to better accommodate the need for long-term, intermittent care suggested by the “revolving door” phenomenon?
- What are the ethical implications of discussing the likelihood of future disorder development with patients, and how can this be done in a sensitive and empowering manner?
