Most of us will have at least one diagnosable mental disorder before we’re 45 years old. And most of us think that if we’re diagnosed with a disorder today, our future (and past) mental health concerns will be related to that disorder. But will they? New evidence is challenging the wisdom of focusing too much on the diagnosis, given that several recent studies demonstrate transdiagnostic aetiology (Antilla et al., Goodkind 2015; Sprooten 2017), and that transdiagnostic treatments are increasing in popularity (Meier & Meier 2017).

What if most people met the criteria, over time, for a succession of different disorders? One two-decade study with over two million Danish participants showed just that (Plana-Ripoll et al, 2019): every mental disorder that was diagnosed carried an increased risk that the patient would be diagnosed at another time with other disorders, both within and across internalising, externalising, and thought disorder diagnostic families. This study was conducted on patients on inpatient and outpatient clinical registers, thus excluding those being treated in the community and going untreated, and therefore increasing the likelihood that the research participants experienced more complex comorbidities and lengthier treatments than usual.

In this study, Caspi and colleagues (2020) report on replication and extension of the Danish research, using a birth cohort that was representative of the entire population of New Zealand, a cohort whose mental health has been repeatedly and systematically assessed for four decades. The investigators were interested in the stability of mental health diagnoses over time. If you’ve been diagnosed with depression, for example, what are the chances you’ll be diagnosed later with schizophrenia, anxiety, alcoholism, or one of the myriad other diagnostic labels?

Methods

The Dunedin Study is an ongoing longitudinal cohort study of 1,037 participants, all born in New Zealand in 1972 and 1973. When the participants were eleven (1983/84), it was the first cohort where mental health disorders were measured in children using standardised diagnostic interviews (Anderson et al, 1987). Subsequent diagnoses were made on 9 occasions with strong participant retention until participants turned age 45 years. This diagnostic time-series allowed the researchers to describe mental disorder life histories in terms of 3 developmental parameters: age of onset, duration, and comorbid diversity.

The research group applied confirmatory factor analysis to participants’ symptoms in order to summarise mental disorder life histories by means of a general factor of psychopathology, the p-factor (Caspi & Moffitt, 2018; Lahey et al, 2017). They tested the hypothesis that mental disorder life histories (p-factors), reflect compromised brain function, by examining associations with neurocognitive deficits at age 3, a subsequent cognitive decline from childhood to adulthood, and advanced brain age in midlife, as derived from neuroimaging.

Listen to Prof Terrie Moffitt talking about this research before her keynote talk at the 2021 #IoPPNfestival:

Results

• Approximately one-third (346 of 1,013) of the cohort experienced an initial onset of a disorder by age 15 years, and nearly two-thirds (600 of 1,013) experienced an initial onset of a disorder by age 18 years.
• Early onset was associated with more years with a disorder and more comorbid disorders. Participants with early-onset disorders subsequently met diagnostic criteria for more diverse disorder types (r = 0.64; 95% CI, 0.60 to 0.67; P < .001).
• A large majority of the participants eventually experienced a disorder: by age 45, 86% of the cohort met the criteria for at least one disorder. Although these findings suggest that a majority of the population has a mental health disorder at one or more points, it’s a relatively small minority who sustain enduring mental health disorders (14% in this cohort).
• Participants characterised by only one pure disorder were atypical. For example, among participants ever diagnosed with an internalising disorder, most (503/712) also experienced externalising or thought disorders and another 16% (113/712) had multiple kinds of internalising disorders.
• Longitudinal analyses showed that participants with extensive mental disorder life histories had poor neurocognitive functioning at age 3 years, experienced childhood-to-adulthood cognitive decline, and older brain age at midlife.

Conclusions

Most participants had one or more mental disorders by the time they are 45, and most of those who later had chronic or severe mental disorders showed evidence of problems at the age of three.

The authors concluded:

Mental disorder life histories are better described by age of onset, duration, and diversity of disorder than by any particular diagnosis. The finding that most mental disorder life histories involve different successive disorders helps to account for genetic and neuroimaging findings pointing to transdiagnostic causes and cautions against over-reliance on diagnosis-specific research and clinical protocols.

Applying a life-course framework to mental health problems orients research and practice away from looking for the cause of a single disorder at one point in time toward considering the dynamics of an individual’s mental disorder life history. This means encouraging researchers to design tools to assess an individual’s life-course vulnerability to psychopathology, identify causes of this vulnerability, explain why this vulnerability manifests in different diagnoses at different points in the life course, and develop transdiagnostic prevention.

Strengths and limitations

This is a report on a complex analysis of longitudinal data, part of an enormous multi-decade project involving over a thousand participants and hundreds of professionals conducting cognitive, physical, financial, psychological, and mental health assessments on these participants every few years. The data analysis was carefully considered and reconsidered from different angles, meeting a high standard of evolving best practice in psychological assessment techniques and statistical methods.

One of the strength (as well as limitations) of this report is that the participants comprise a complete cohort of people born in a single place (Dunedin, New Zealand) in a single year, 1972/73. They were not chosen because of a propensity to mental disorders, an important strength, but they all come from the same city in the same year, a limitation. They are predominantly White, and lived through the same historic period. The fact that the findings replicate a very large Danish study (Plana-Ripoll, et al, 2019) mitigates the geographic homogeneity of the participants, although, in both studies, the participants were predominantly White.

I found this a compelling and fascinating report. I did not realise that mental health disorders are as prevalent as they are, or that diagnoses vary so dramatically across time within a single individual. The biggest strength of this research may be its importance in informing professional practice, research, and public health education. The biggest limitation as I see it is the ethnic homogeneity of the research participants. It is important to investigate whether or not the same findings apply to people from other cultures and races.

Implications for practice

Currently, most research programmes, treatment protocols, specialist clinics, and specialist journals are oriented to presenting diagnoses, on the assumption that a person’s diagnosis provides information about causes and prognosis. This study indicates, however, that a current diagnosis should be seen as a starting place for understanding a dynamic and temporary situation; one that will probably change over time. Instead of focusing on the current diagnosis as a static label of a person’s permanent mental health status, then, mental health practitioners and others would do well to focus on the current symptoms and the complexities of the individual’s personal life situation and journey.

This study’s findings argue against too much reliance on diagnostic labels, and for considering each person’s mental health as dynamic, changing over time with changing situations, experiences, and support. We should pay attention to alleviating troublesome symptoms as they show up, and avoid categorising people as mentally ill, or not. The authors write:

Therapy cannot just mitigate the presenting symptoms, but must also build skills for maintaining enduring mental health. The life-course approach makes transdiagnostic interventions a high priority.

Perhaps the most urgent implication is the fact that all of those who go on to have serious or chronic mental disorders show up as needing help as children and teens, many as early as age three. These findings underline the importance of investing in children’s and adolescents’ mental health, both prevention and treatment, especially because too few children receive effective treatment in a timely fashion.

The results of this investigation encourage researchers to design tools to assess an individual’s life-course vulnerability to psychopathology, identify causes of this vulnerability, explain why this vulnerability manifests in different diagnoses at different points in the life course, and develop transdiagnostic prevention.

Finally, this study highlights the prevalence of mental illness in the population and the importance of public health education concerning this. Knowing that 86% of us will have at least one mental disorder by the time we’re 45 and that 85% of this group (73% of the population) will have at least one other (different) diagnosis should go a long way toward removing the cultural stigma against mental illness.

Statement of interests

Dona has written a series of blogs on The Origins of You, a book co-authored by the principal author and two of the other authors of this study. She is also writing a book that refers to this material and other references. However, she was not involved in this study or has any other potential conflicts of interest.

Links

Primary paper

Caspi, A, Houts, R, Ambler, A et al (2020) Longitudinal assessment of mental health disorders and comorbidities across 4 decades among participants in the Dunedin birth cohort study. JAMA Netw Open 3(4):e203221. doi:10.1001/jamanetworkopen.2020.3221

Other references

Anderson JC, Williams S, McGee R, Silva PA (1987) DSM-III disorders in preadolescent children: prevalence in a large sample from the general population. Arch Gen Psychiatry. 44(1):69-76. doi:10.1001/archpsyc.1987. 01800130081010

Anttila V, Bulik-Sullivan B, Finucane HK, et al (2018) Brainstorm Consortium. Analysis of shared heritability in common disorders of the brain. Science. 2018;360(6395)eaap8757. doi:10.1126/science.aap8757

Caspi A, Moffitt TE (2018). All for one and one for all: mental disorders in one dimension. Am J Psychiatry.;175 (9):831-844. doi:10.1176/appi.ajp.2018.17121383 25.

Goodkind M, Eickhoff SB, Oathes DJ, et al (2015) Identification of a common neurobiological substrate for mental illness. JAMA Psychiatry 72(4) 305-315 doi:10.1001/jamapsychiatry.2014.2206

Lahey BB, Krueger RF, Rathouz PJ, Waldman ID, Zald DH. (2017) A hierarchical causal taxonomy of psychopathology across the life span. Psychol Bull.;143(2):142-186. doi:10.1037/bul0000069

Meier M, Meier M (2017) Clinical implications of a general psychopathology factor: a cognitive-behavioral transdiagnostic group treatment for community mental health.J Psychother Integration 28(3) 253-268. doi:10. 1037/int0000095

Plana-Ripoll O, Pedersen CB, Holtz Y, et al (2019) Exploring comorbidity within mental disorders among a Danish national population. JAMA Psychiatry 76(3) 259-270. doi:10.1001/jamapsychiatry.2018.3658

Sprooten E, Rasgon A, Goodman M, et al (2017) Addressing reverse inference in psychiatric neuroimaging: meta-analyses of task-related brain activation in common mental disorders. Hum Brain Mapp 38(4)1846-1864. doi:10.1002/hbm.23486

Photo credits

• Photo by pawel szvmanski on Unsplash
• Photo by Blake Connally on Unsplash
• Photo by Kelly Sikkema on Unsplash