Lisa Balboa, Maryse Nashime, Serena Soong and Joe Wilson assess how data use could help to progress actuarial understanding of mental health
Mental health is complex. There are more than 200 classified forms of mental illness, spanning a wide spectrum of symptoms, severity and treatment approaches. Further intricacies are introduced by co-morbidities, as the presence of a mental health condition together with other mental or physical health conditions doesn’t always lead to an additive impact on risk. The IFoA Mental Health Working Party has been exploring how the use of data could provide deeper actuarial insights into some of the complexities of mental health.
Remote health services
The UK government’s COVID-19 mental health and wellbeing surveillance: report reveals a deterioration in the UK population’s mental health during the pandemic. Mental health worsened during lockdown periods, in particular, and this was seen most prominently among those who contracted COVID-19 or who had financial worries. At the same time, accelerated digital adoption and increased public awareness during the pandemic have provided new opportunities to improve actuarial understanding of mental health.
The impacts of COVID-19 fast-tracked the adoption of telehealth support services, including in the insurance sector. Digitally accessible remote health services facilitated patient access to mental health services when in-person healthcare was disrupted, and telehealth services are likely to continue to be important in improving access to healthcare.
Digital health data generated from such services, as well as the more general global increase in the adoption of electronic health records, could be leveraged to gain further insights into mental health. Avenues for further research include a deeper exploration of the impact of co-morbidities and various treatment approaches on risk. These larger data sources and newer modelling techniques could be valuable for actuaries who are looking to assess risks arising from lower frequency mental health conditions.
The impacts of COVID-19 fast-tracked the adoption of telehealth support services
Wearable technology
The wearable technology sector also experienced high levels of growth during the pandemic. Wearables can capture everyday health in areas such as sleep and activity levels. There is a link between lifestyle factors and mental health: research carried out by AIA Australia and Quantium Health, for example, found that 30% of depression risk is connected to risk factors such as diet, sleep and exercise.
So far, only a few life and health insurers are incorporating wearable technology into their core insurance products; for those insurers, it is typically used to encourage healthy behaviours by offering rewards. With the public’s growing use of wearables and health apps, such detailed and specific data from outside the health service could provide new associations between lifestyle and health. These could lead to an expansion of insurance coverage, including coverage regarding mental health. However, considerations around policyholders’ privacy and data rights remain central for the adoption of digital health data by insurers.
An evolving landscape
Public engagement with mental health is at an all-time high, and researchers are poised to collect more and better data in this area. New population studies would be useful to refine our assessment of current risk factors. One example of a risk factor is age at diagnosis. The research from AIA Australia and Quantium Health suggests that older individuals are more likely to experience depression due to their higher likelihood of comorbidities, while Kessler et al.’s paper ‘Age differences in the prevalence and co-morbidity of DSM-IV major depressive episodes: results from the WHO World Mental Health Survey Initiative’ finds that incidence of co-morbid depression falls with age. More data could help to unpack the relationship between age and depression.
There has also been significant progress in recent years in terms of communication around mental health in insurance. The Association of British Insurers’ Mental Health and Insurance Standards emphasise that insurers should regularly review their underwriting approaches using up-to-date, relevant and statistically credible evidence, and they have improved transparency around underwriting approaches for mental health conditions.
Best practices for validating the reliability and accuracy of digital health data needs to be established
Where do we go from here?
When introducing additional risk factors, it’s important to assess the limitations of the data and how these limitations might be addressed. With the rapid growth in digital health data availability, best practices for validating the reliability and accuracy of this data need to be established. Historic data for some of these additional risk factors may also be limited, which may present a barrier to carrying out traditional longitudinal analyses that support long-term risk assessment for most life insurance products. When modelling the impact of mental health on morbidity and mortality risks, it’s also relevant to consider how past and future changes in disclosure rates, diagnosis and treatment of mental health conditions might impact results. Cross-industry discussion and close collaboration with the medical profession could help to overcome some of these challenges.
The hope is that by incorporating these additional data sources and risk factors into mental health modelling, we can drive better understanding of the interactions between mental health and insurance. Analysis could be enhanced right across the insurance value chain to shape the insurance product design, underwriting processes and health support services offered by insurers.
During the past two years, the Mental Health Working Party has focused on facilitating cross-industry conversations around mental health in insurance, and we have recently published a paper focused on data and modelling considerations for mental health in life insurance. This paper is available on our working party page at bit.ly/MentalHealthWP
We encourage and welcome input into the work we’re doing, so please get in touch if you have comments or suggestions.
Additional risk factors
A growing awareness of mental health, coupled with technological change, is contributing to richer digital health data sources. Further insights into risk factors from these datasets could allow insurers to advance portfolio-level risk assessment and use a more in-depth view of customers’ mental health needs to guide insurance product development. These additional risk factors are considered in more detail in the Mental Health Working Party paper ‘Data and Modelling Considerations for Mental Health in Life Insurance’.
Age at diagnosis
Research links age and incidence of some mental health conditions, particularly when analysed together with co-morbidities.
Individual’s support network
An individual’s personal support network can be crucial in the treatment and management
of mental health conditions. This bio-psychosocial context can be challenging to assess quantitatively, but the integration of digital mental health tools into clinical practice could create new opportunities here.
Engagement in managing mental health
Adherence to treatment is important to most conditions, whether mental or physical. Attendance rates at counselling, cognitive behaviour therapy or other healthcare appointments could be factors to consider.
Lifestyle factors
Research finds that lifestyle factors such as nutrition, sleep and exercise are associated with mental health outcomes. These factors might be further investigated using wearable technologies and other health apps.
Lisa Balboa is a business development actuary at Hannover Re
Maryse Nashime is a senior actuary at Partner Re
Serena Soong is head of risk and solvency planning at Legal & General
Joe Wilson is a pricing actuary at RGA
