JK Rowling wrote: “It’s so difficult to describe depression…it’s that cold absence of feeling – that really hollowed-out feeling.” Most of us can’t express our feelings as articulately as this, but what if you can’t convey your feelings to others at all?
Within the realm of Intellectual Disability (ID) diagnostics can be tricky. Working with patients with ID is really rewarding, but it is not without its trials. Many patients can’t communicate their symptoms verbally and it is difficult to determine if a change in behaviour is a symptom of mental illness, physical illness or a situational stressor. If it is felt to be a mental illness, which mental illness is it? Symptoms of mania can be misinterpreted as challenging behaviour or ADHD. Furthermore, diagnostic overshadowing results in physical and mental illnesses being ignored and untreated (Cooper, Melville and Morrison, 2004; Simpson, Mizen and Cooper, 2016).
Mental illness in patients with ID is more than twice as common as in the general population (Cooper et al, 2007; Stansfeld et al, 2014). Yet there is surprisingly little literature on incidence and predictors of affective disorders in this population. Perhaps this is because of high resource requirements or difficulty in cohort retention, but either way the silence is deafening.
If we knew how frequently affective disorders occur in this population it would guide the weight with which it should be considered as a differential. This is why Professor Sally-Ann Cooper’s new paper in the BJPsych is music to our ears, as it outlines a prospective cohort study investigating incidence of affective disorders in this population.
The main aims of the study were to ascertain the incidence of affective disorders in adults with mild to profound intellectual disabilities and to investigate what factors predict incident unipolar depressive episodes. These findings were then compared with rates from the general population.
- All adults with intellectual disabilities (ID) registered with a GP were contacted
- Recruitment into a longitudinal cohort during the first time point (T1) from 2002-2004
- Assessments were repeated two years later between 2004-2006 (T2)
- At each time point, face to face interviews were completed:
- Psychiatric Assessment Schedule for Adults with Intellectual Disability (PAS-ADD) was completed for each person.
- C21st Health Check was completed to exclude physical causes and to identify and investigate possible predictors of unipolar depression
- Surveys assessing demographics
- PAS-ADD when completed with person’s carer gave 100% sensitivity and 58% false positive rate to pick up people meeting ICD10 criteria. If the threshold was triggered the person underwent a second interview with a consultant psychiatrist
- Incidence rates were compared with general population rates to calculate Standardised Incidence Ratios (SIRs) (Lloyd et al., 2005; Kroon et al., 2013)
- 17 factors were investigated including personal factors, lifestyle and supports and health and disabilities.
- The cohort size at T2 was 936 individuals. Cohort retention was 69.6%
- 42 people (6.5%) had incident unipolar depression
- 13 people (2.0%) had incident of bipolar affective episode
- Standardised Incidence Ratios
- Depression: 1.19 (95% CI 0.85 to 1.93)
- First episode mania: 41.5 (95% CI 5.0 to 149.8)
- Bipolar affective disorder (new and recurrent): 2 (95% CI 1.06 to 3.41)
- 51.9% of those with incident affective disorder experienced incidence and recovery within the 2-year period
- 22.4% of the cohort were prescribed mood stabilisers, mostly for epilepsy
- 8.2% of these had incident unipolar depression
- 1.4% of these had incident bipolar depression
- 2.1% of these had incident mania
- This was not statistically different from those in the cohort not taking mood stabilisers
- Factors that predicted incident unipolar depression
- Preceding life events (OR 1.30, 95% CI 1.02 to 1.65)
- Problem behaviours (OR 2.27, 95% CI 1.18 to 4.37)
Dr. Cooper previously reported a higher point prevalence of depression (3.8%) and bipolar affective disorder (2.3%) in the ID population compared to the general population (Cooper et al, 2007).
Incidence is similar but the prevalence of depression is higher in the ID population. There is a higher incidence and prevalence of bipolar affective disorder despite higher mood stabiliser use. This suggests affective disorders are either more enduring, treatment resistant or under-treated in this population.
Preceding life events and challenging behaviour increase the risk of an incident unipolar depressive episode. However, age, female gender, living in more deprived areas, no occupation, urinary incontinence and being a smoker were not predictive of unipolar depression.
Strengths and limitations
- This was a thorough study. We can be assured that the majority of people with ID were recruited from the GP, reducing selection bias. The prospective design gives us further information of the temporal relationship between factors and accurate incidence rates.
- The 2-year interval is short enough to pick up the majority of incident cases since there are no assurances people with ID will present to services when unwell.
- We can also be satisfied that if an affective disorder occurred it was likely to be picked up by the 100% sensitive PAS-ADD and collateral history. They dealt with the high false positive rate by performing a second interview with a consultant psychiatrist.
- They used a multivariate analysis which assessed individual predictive factors and reduced confounding bias.
- Cohort retention was adequate at 70%. This is much better than usual for this population and there was no difference between those who remained and those who left the study at T2. The greater Glasgow area is diverse in socio-economic status and urban/rural areas which makes the study generalisable to other high-income countries.
- This is a new area of research which means there is difficulty in assessing reliability of the results of this paper. All efforts were made to ensure a comprehensive study was completed and results compared to similar studies but we should be cautious in immediately accepting these results prior to replication. The Standardised Incidence Ratio (SIR) for first episode mania has a very wide confidence interval. This is likely due to low numbers and reduces our confidence in the reliability of this figure. The author states that “in spite of this the lower limit of the confidence interval is 5 which gives credibility to a higher incidence of mania in the ID population.” This is a valid point.
- The study may be underpowered. It is designed for all mood disorders, but it is only in the lower incidence ones that we see the increased rates. A repeat study would be worthwhile.
- A study is only as good as its methods. The assessment tools used in this study were different to those implemented in the general adult population. As such, SIRs become less accurate and introduce measurement bias.
- Although we can be satisfied that potential disorders are picked up by the PAS-ADD, the consultants assessing patients used their own judgement as well as the three different diagnostic criteria. How sure are we that they are similar in their scoring/assessment? There’s no mention of a Kappa to demonstrate non-heterogeneity. It might be overkill; but when relying on more than one assessor we should expect an assessment of how closely they agree with one another and its absence is a small oversight.
Implications for practice
This paper highlights the importance of identifying and treating mental illness thoroughly in people with intellectual disabilities. The high incidence rates of mania compared to the general population should prompt clinicians to consider this when making a differential diagnosis.
We are given new information on the factors that predict an incident depressive episode in an ID population which will be helpful for awareness in clinical practice. This affirms that generalisations cannot be made from the general population.
Additionally, more services are required to research affective disorders in this population to gather further robust evidence to support these results.
The famous quote “the worst thing about a disability is that people see it before they see you” is particularly relevant after reading this.
Conflicts of interest
Cooper S, Smiley E, Allan L, Morrison J. (2018) Incidence of unipolar and bipolar depression, and mania in adults with intellectual disabilities: prospective cohort study. The British Journal of Psychiatry (2018) 1–6. doi: 10.1192/bjp.2018.12 [Pubmed Abstract]
Cooper SA, Smiley E, Morrison J, Williamson A and Allan, L. (2007) ‘An epidemiological investigation of affective disorders with a population-based cohort of 1023 adults with intellectual disabilities’, Psychological Medicine, 37, pp. 873–882. [Pubmed Abstract]
Cooper SA, Smiley E, Morrison J, Williamson A and Allen L. (2007) ‘Mental ill-health in adults with intellectual disabilities: prevalence and associated factors’, British Journal of Psychiatry, 190, pp. 27–35. [Pubmed Abstract]
Cooper SA, Melville C and Morrison J. (2004) ‘People with intellectual disabilities.’, BMJ (Clinical research ed.). BMJ Publishing Group, 329(7463) [BMJ Abstract]
Kroon J. S. et al. (2013) ‘Incidence rates and risk factors of bipolar disorder in the general population: a population-based cohort study.’, Bipolar disorders, 15(3), pp. 306–13.[Pubmed Abstract]
Lloyd T. et al. (2005) ‘Incidence of bipolar affective disorder in three UK cities’, British Journal of Psychiatry, 186(2), pp. 126–131. [Pubmed Abstract]
Simpson N, Mizen L and Cooper SA. (2016) ‘Intellectual disabilities’, Medicine, 44(11), pp. 679–682. [Medicine Journal Abstract]
Stansfeld S, Clark C, Bebbington P, King M, Jenkins R and Hinchliffe S. (2014) Adult Psychiatric Morbidity Survey: Survey of Mental Health and Wellbeing, England, 2014 – NHS Digital.