The economic burden of untreated OCD: A Q&A with Naman Kochar

Obsessive-Compulsive Disorder (OCD) exacts a toll beyond the obvious emotional and psychological burdens. Beneath the surface lie hidden direct and indirect costs, often overlooked in discussions about this debilitating condition. 

Naman Kochar is a Senior Analyst at Costello Medical and the lead author of a paper entitled,  “A cost-of-illness analysis of the economic burden of obsessive-compulsive disorder in the United Kingdom.” In this work, published in the journal Comprehensive Psychiatry, Kochar not only uncovers the hidden financial strains borne by those grappling with OCD in the UK but also illuminates pathways to more effective interventions. Kochar’s analysis is part of a broader effort to reshape perspectives and catalyze meaningful change in the landscape of OCD care.

Note: the figures in this Q&A are in British Pounds Sterling (£). At the time of writing, (10/25/2023) £1 is equal to $1.21. 

Q: Can you elaborate on the methodology used to estimate the cost-of-illness (CoI) of OCD in the UK? For instance, how did you categorize the population by age group and symptom severity?

In the model, all people with OCD in the UK were divided into subpopulations based on criteria that may affect the total associated cost per year in treating individual or societal costs arising from OCD symptoms. The number of individuals in each subpopulation for an average year was estimated and multiplied by the average annual associated costs for that subpopulation to give a total cost for each subpopulation. 

Then, the overall cost of illness was obtained by summing costs over all subpopulations. The specified criteria for subdividing the population of OCD patients were established in line with NICE recommendations for OCD treatment, as well as with considerations of contributions to societal costs.

Age: It was necessary to consider children and adults separately, as NICE recommends different treatment pathways for these groups. Furthermore, societal costs are different for these groups; societal costs for adults will be focused on unemployment and lost productivity, whereas for children, these will be concentrated on costs for caregivers. Another age grouping was included for elderly patients since recommended drug dosages may differ for older adults, and elderly individuals are also less likely to contribute to lost productivity costs since they are less likely to be employed.

Symptom severity: Population subdivision by symptom severity was established based on the NICE stepped-care model, which determines treatment routes and, hence, treatment-related costs. Symptom severity may also determine associated indirect costs. For example, more severe symptoms in adult patients lead to greater out-of-pocket expenses. This subdivision was based on the baseline symptom severity experienced by patients before treatment.

Epidemiological inputs for different age and symptom severity groups were obtained from up-to-date UK-based sources where possible, including surveys (such as the Adult Psychiatric Morbidity Survey, conducted in 2014, and the Mental Health of Children and Young People Survey, conducted in 2017) and national databases (such as the Office for National Statistics). Where this was not possible, literature estimates had to be used; these estimates were validated by expert clinician opinion to ensure appropriateness.

Q: What specific cost inputs were derived from national databases, and how did you incorporate additional inputs from literature searches or expert clinician opinion?

The specific cost inputs identified from national databases included the following:

• Drug acquisition unit costs of SSRIs and clomipramine, which were obtained using the drugs and pharmaceutical electronic Market Information Tool (eMIT)
• Healthcare resources (e.g. therapy costs, including the additional costs associated with supervisors and non-medical care coordination, and residential care costs), which were obtained from the most recent available PSSRU reference costs, respectively
• Out-of-pocket costs, which were sourced from an analysis that collected data from participants taking part in the OCTET multi-center randomized controlled trial
• Lost productivity costs, which were obtained from the average weekly earnings in the UK

Aside from Professor Fineberg, the opinions of three expert clinicians were sought to validate assumptions from the literature and gain insights into real-world treatment patterns. In particular, clinicians estimated the proportion of patients with mild, moderate, and severe OCD on each type of treatment, and an average of their estimates was used to inform the treatment’s “market shares.” Additionally, clinicians validated key model assumptions—e.g., the twelve-month prevalence of OCD, the durations of treatment, and the increase in the frequency of treatment for a patient with comorbid OCD and depression, compared to a patient with just OCD.

Q: Could you explain the significance of conducting scenario analyses, particularly about comorbid depression treatment and presenteeism? How did these factors impact the overall cost estimates?

*Note: Presenteeism refers to working while affected by illness, often with reduced productivity.

Scenario analyses were conducted to assess the impact of using alternatives for key model inputs. The model inputs varied were those with the greatest uncertainty; for instance, as we anticipated, there was little literature on the proportion of patients that remained untreated and those that were treated inadequately, and a range of values for these parameters were tested.

Additionally, due to the conservative approach we took during the analysis, only those cost categories that could be robustly captured were included in the base case of the model—we understood that our analysis would not be able to capture the full extent of the economic burden of OCD, and therefore aimed to obtain a ‘lower bound’ estimate of the cost-of-illness in the base case. Some scenario analyses were conducted to include some of these additional costs. For example, due to the strong established links between OCD and depression, a simple scenario analysis was included in the model to account for extra therapy needs for patients with comorbid OCD and depression. 

Including the costs associated with depression more than doubled the treatment-related costs, highlighting that comorbid depression amplified the need for psychotherapy amongst OCD patients. Meanwhile, the 71.9% increase in lost productivity costs when the impact of presenteeism was included reflects the debilitating impact of the condition even on people with OCD who continue to go to work.

From the scale of the burden revealed by these findings, it is clear that more research is needed on the interaction between OCD and depression and how this affects both the symptom burden on people with OCD and the treatment patterns amongst patients treated for both conditions. Equally, research to identify the impact of presenteeism amongst people with OCD in quantitative terms would be valuable to more clearly communicate the impact of this relatively unexplored area of burden.

Q: The study indicates that societal costs were primarily driven by lost productivity through absenteeism. Could you discuss the implications of this finding on both individuals with OCD and their caregivers?

This finding reveals a considerable humanistic burden of OCD, reflecting the high symptom burden and consequent need for better management of the condition. Lost productivity amongst individuals with OCD forces them to miss considerable time from work, resulting in financial loss to themselves and/or to their employers, and methods that alleviate the symptom burden and increase the ability of these individuals to work at full capacity would help to mitigate these losses.

Lost productivity amongst caregivers was not included within the base case of the model and, therefore, was not a driver of the base-case results. However, we did explore the impact of lost caregiver productivity in a scenario analysis—the high values identified for family members and caregivers reflect:

• The high symptom burden of disease; symptoms in severe cases may be so debilitating that family members/informal caregivers may have a considerably reduced capacity to work.
• The high treatment burden of the disease on the informal caregiver network: particularly amongst younger people with OCD, caregiver lost productivity may stem from the regular need to accompany individuals with OCD to periodic appointments, for example.

It can be hypothesized that high caregiver productivity stems from a combination of the above two factors, although other considerations may also definitely play a part.

The regular need for caregivers to take time away from work to care for individuals with OCD, as indicated by the considerable productive time lost, may also result in a considerable emotional and mental burden to these caregivers. Factors such as stress, depression, and reduced relationship satisfaction have been reported in the literature (Lee 2015); these factors may amplify the reduction in caregiver quality of life stemming from lost productivity.

Q: You mentioned that the base-case results likely underestimated the true economic burden of OCD. Can you explain why this might be the case and what factors were considered in this underestimation?

The likely underestimation of the economic burden of OCD reflects the conservative approach that we took in the model. Other than taking conservative estimates for key model inputs where there was a range of input values available (e.g., OCD prevalence), we also excluded cost categories for which there was no robust data in the analysis. Aside from the impacts of presenteeism, reduced productivity amongst caregivers, and the need for additional treatment due to comorbid depression (which were all explored in scenario analyses), we did not consider the following:

• Adverse event costs associated with the delivery of SSRIs were not considered: there was limited data on the frequency of these events amongst OCD patients.
• Missed opportunity costs were not considered: education and employment are influenced by a wide array of factors that may be independent of OCD, and hence, differential education and employment levels cannot be attributed solely to the impact of the condition.
• Indirect costs to family members and informal caregivers beyond loss of productivity were not considered: There is a lack of data to inform the impact on family members or informal caregivers of other factors that may impact healthcare or societal costs, such as poorer mental health.
• Other factors that may affect an OCD patient’s living circumstances (such as increased chance of marital separation or interactions with law enforcement) were not considered. The financial impact of these factors is not well-defined.

It is unclear whether, to what extent, and over what timeframe these factors result in economic ramifications for people with OCD. However, given their exclusion from the current model, it can be concluded that the current value of the economic burden is likely to be an underestimate.

Q: Were there any surprising or unexpected findings during this study that stood out to you? How did these findings influence your understanding of the economic impact of OCD?

The finding that stood out most was the discrepancy between the average cost per patient to the NHS (£525 per annum) and the average cost per patient to themselves as an individual and society (£6,551 per annum), giving a total cost of £7,077 per patient. The societal cost and burden were even higher if depression is supervised, as it commonly does if OCD is inadequately treated. We also learned that around half of individuals with OCD are estimated not to be receiving any form of treatment. Taken together, this suggests a huge discrepancy between what we as a nation spend on healthcare for this illness and the actual cost and burden of disease to our nation, which is over ten times higher than the current investment in healthcare.

Q: Given the substantial economic burden identified in the study, what are the potential implications for policymaking and resource allocation in mental health care, specifically for individuals with OCD in the UK?

The implications are that OCD is a major serious mental disorder that healthcare providers have relatively overlooked for too long. OCD needs to be better understood, diagnosed, treated, and maybe one day prevented across all age groups. 

Strategically, we need greater investment in research into OCD, which, according to recent figures (MQ Lancet Psych 2019), receives less than 1% of the total mental health research expenditure. We also need greater investment in health and social care, i.e., more individuals with OCD treated earlier and more effectively before the disease becomes entrenched so that they achieve maximum potential in the educational and work environment. As OCD tends to arise early in life, and having a longer duration of untreated illness predicts a poorer response to treatment, for many, this means better recognition, diagnosis, and treatment in childhood or adolescence. 

By investing resources into research and effective treatment, which would, on the one hand, increase direct NHS and social care costs, we would be expecting to reap dividends in terms of better health and social outcomes for those with OCD and, therefore, a reduction in the massive individual, societal cost and burden of this pernicious disorder.

Could you suggest any adjustments to current treatment approaches that might be explored to address the economic burden of OCD, particularly in light of the study’s findings?

We should first take a careful look at the existing evidence-based approaches and ensure they are delivered efficiently.

First, OCD needs to be recognized in its early stages—by parents, individuals, doctors, and teachers; and treated promptly and efficiently to obtain better outcomes. This requires a major investment in public health education.

Then, we need ready access and available treatments that work without undue delay. This means access to healthcare professionals who understand OCD and can refer individuals to the right clinical services that can deliver the treatments promptly and effectively.

You can access effective treatment for OCD and related conditions

Promising scientific research about OCD continues to be published, but one thing researchers have known for decades is just how treatable the condition is with exposure and response prevention (ERP) therapy. 

This evidence-based form of treatment is now more accessible than ever, through live, face-to-face video therapy sessions with NOCD Therapy. Research has shown that NOCD’s virtual ERP is just as effective as in-person OCD treatment, and leads to results in half the time.

If you or someone you love may be struggling with OCD, we encourage you to learn more about NOCD’s accessible approach to OCD and related conditions.