Whole pathway modelling of depression in patients with diabetes (theme 2: mental health) part 2: independent economic evaluation: methods and results, discussion and conclusion

Background

Diabetes is associated with premature death, along with a number of serious complications such as amputation, blindness and heart disease. The presence of diabetes with comorbid depression increases both the risk of mortality and the prevalence of complications, as well as resulting in increased healthcare use and costs.

Historically care pathways for diabetes and for depression have been isolated from each other, resulting in a siloing of care. There is now interest in exploring methods by which the care pathways may be integrated, and the impact that this has on both patient outcomes and costs to the healthcare system.

Objectives

This project aimed to assess the health economic outcomes associated with having both diabetes and depression, and explore potential changes to the care pathways for these diseases that could be implemented to improve the health economic outcomes.

Data Sources

Data were drawn from a range of sources including published literature identified through a series of searches. A scoping review in MEDLINE identified pooled data, followed by subsequent searches in MEDLINE, PsychInfo and Cochrane electronic databases for individual studies where no pooled data was available. Searches were conducted from July to November 2013. These searches were supplemented by papers identified through consultation with experts and by papers known to the authors.

Methods

The objective of the literature review was to understand the relationship between diabetes and depression to inform the model development and to identify evidence that may be used to populate the economic model. The search aimed to identify information for multiple parameters. The review of the literature consisted of 3 steps. Firstly, a scoping review of reviews was conducted to inform the model conceptualisation. This aimed to identify all types of reviews (narrative, systematic and meta-analytic) relating to depression in diabetes. Secondly, after the model had been developed, targeted searches were conducted to identify studies for parameters where suitable data had not been identified in the reviews. Finally, a systematic search was conducted to provide a more comprehensive understanding of a “key” model parameter, namely the relationship between diabetes-related complications and depression.

A mathematical model of the depression care pathways experienced by people with diagnosed type-2 diabetes (T2DM) in England was created based on the conceptual model developed with the aid of an advisory group supplemented by a review of the published literature and a meeting with current service users. This model took the form of a discrete-event simulation, and was developed to assess the relative cost-effectiveness of proposed service changes from an NHS perspective, wider social benefits were also explored. The population considered for this project was adults with T2DM currently managed within primary care in England. Patients could have existing depression, develop depression, or remain depression free.

The health economic outcomes considered were morbidity, quality of life, mortality, and costs incurred by the healthcare system. The potential service changes (interventions) considered included: improvements in opportunistic screening for depression; collaborative care; both improvements in opportunistic screening for depression and collaborative care. The comparator was current standard care. An expert group of advisors assisted in the identification of relevant service changes and identification of relevant evidence.

Results

Sixty reviews of depression in diabetes were identified in the scoping search. Insufficient data was identified in the reviews to populate all of the model parameters. Targeted searches were therefore conducted to identify data from individual studies for: the prevalence of depression amongst T2DM in England; the natural history of depression in diabetes, including incidence, recurrence, relapse and persistence; and the effectiveness of screening for depression in diabetes. A review was also conducted to identify evidence on the link between the development of diabetes-related complications and depression, and the converse.

The model estimated that the proposed policies have the potential to reduce both the time spent with depression, and the number of diabetes-related complications experienced. All three policies were associated with an improvement in quality of life and an increase in depression-free years compared with current practice, but with an increase in health care costs. Overall Policy 3, which examined the effect of introducing both collaborative care and increasing opportunistic screening together, was estimated to produce the greatest benefits in terms of both events avoided and depression free years. However, the incremental cost-effectiveness ratio (£37,421) comparing Policy 3 to current practice is above the cost per QALY currently considered cost-effective. In addition, when comparing across the policies, Policy 2 (improvement in opportunistic screening) is dominated by Policy 1 (collaborative care), and comparing Policy 3 with Policy 1, this policy would again not be considered cost-effective (with an ICER of £68,017) when assuming a willingness to pay threshold of either £20,000 or £30,000 per QALY. However, these estimates do not take into account the uncertainty surrounding both parameters and structural assumptions. All three policies produced some benefits when looking at a wider societal perspective and were associated with a reduction in both the number of days off sick due to ill health and the need for informal care.

Additional research is required to decrease the uncertainty in the results presented, such as in the bi-directional relationship between diabetes and depression, and the natural history of depression in patients with diabetes.

While this study examined the effects of policies in patients with diabetes, the proposed changes to the pathway are potentially generalizable to patients with other long term physical conditions predominantly treated within primary care.

Using the evidence currently available, the results of this research suggest that policies targeted at identifying and treating depression early in patients with diabetes may lead to a reduction in diabetes related complications and depression, which in turn increase life expectancy and health related quality of life. Although there is an increase in overall health care costs, the results show that this is below the willingness to pay threshold currently considered acceptable in England.