Major depressive disorder (MDD) and type 2 diabetes are growing health problems, with high costs to both individuals and society. Most research focuses on either MDD or on type 2 diabetes. However, more people have both MDD and type 2 diabetes than we would expect and therefore more research is needed to understand the complex relationship between these two conditions.
Our research aims to use data from the CPRD, with over 200,000 individuals with both type 2 diabetes and MDD to test how: a) having both type 2 diabetes and MDD impacts an individual's health, b) MDD alters glucose levels for type 2 diabetics over time and c) the order and age of diagnoses alter outcomes.
Firstly, we will test how MDD alters blood glucose levels over time. All patients with type 2 diabetes are seen regularly (~ every 6 months) to have their glucose monitored. We will test if, and how, having a diagnosis of MDD as well as type 2 diabetes alters the changes in glucose levels over time and determine whether the relative timing of diagnosis (i.e. MDD before type 2 diabetes or vice versa) is important. Secondly, we will focus on individuals with MDD and test if having type 2 diabetes alters the progression of MDD. For example, do individuals with both conditions have more MDD episodes or are they more likely to develop more severe MDD?
Our findings will be of crucial importance to individuals with both depression and type 2 diabetes.
Aims. The global prevalence of both type 2 diabetes (T2D) and major depressive disorder (MDD) is increasing. Previous research has demonstrated that diagnosis with both conditions leads poorer diabetic control and an increased risk of treatment resistant depression. This study aims to understand the temporal relationship between T2D and MDD using longitudinal health records in two cohorts: one comprising individuals diagnosed with T2D (T2D cohort) and one containing individuals diagnosed with MDD (MDD cohort). Within the T2D cohort, we will explore the association between MDD and glycaemic control (HbA1c), as well as diabetes care, across T2D disease duration. Additionally, we will investigate the effect of poor glycaemic control and MDD on the accumulation of multiple long-term conditions (i.e., cardiovascular traits), frailty, and mortality. In the MDD cohort, we will evaluate the impact of a T2D diagnosis on MDD related outcomes including the severity of MDD, antidepressant prescribing patterns and adverse outcomes associated with depression (i.e., suicide). Within both cohorts, we will examine the potential role of mediators and moderators, such as obesity and inflammation. We will conduct subgroup analyses by sex, ethnicity, and socioeconomic status whenever feasible.
Primary exposures.
T2D cohort: MDD diagnosis.
MDD cohort: T2D diagnosis.
Primary outcomes.
T2D cohort: HbA1c.
MDD cohort: number of recorded episodes, age at first episode, treatment-related outcomes (i.e. psychiatric referral, anti-depressant switching).
Study Design. Retrospective, longitudinal cohort study.
Methods. Multivariable longitudinal analyses will be implemented using several statistical techniques, including latent class growth mixture models, Cox proportional hazards regression and conditional multinomial logistic regression.
Linked datasets. ONS Death registration data, HES data (inpatient, outpatient, A&E), Townsend Deprivation Index (patient and practice level), CPRD Ethnicity records.
Findings. This project will provide novel information about the complex relationship between MDD and T2D, thereby informing decisions on medical management and public health strategies for both conditions.
Cohort 1: Individuals diagnosed with type 2 diabetes.
Type 2 diabetes and related outcomes:
1. Glycated haemoglobin (HbA1c);
2. Age at type 2 diabetes diagnosis;
3. Type 2 diabetes medication and time on medication;
4. Adiposity measures including Body mass index (BMI) and changes in BMI over time;
5. Microvascular complications of diabetes (primary or secondary care);
6. Cardiovascular complications of diabetes (primary or secondary care);
7. Renal complications of diabetes (primary or secondary care);
8. Frailty measures coded using the electronic frailty index (eFI) from primary care data[1];
9. Alzheimer’s disease (progression to)- possible complication of MDD (primary or secondary care);
10. All cause mortality (CPRD death data).
Cohort 2: Individuals diagnosed with major depressive disorder.
Major depressive disorder (MDD) and related outcomes:
1. Major depressive disorder;
2. Number of MDD episodes;
3. Age at first episode of MDD;
4. Number of antidepressants prescribed;
5. Early antidepressant switching;
6. Antidepressant resistant depression;
7. Psychiatric referral;
8. MDD– recurrent episodes, psychiatric co-morbidity, psychiatric referral (primary or secondary care), symptom level information;
9. Frailty measures coded using the electronic frailty index (eFI) from primary care data[1];
10. Alzheimer’s disease (progression to)- possibly related to MDD (primary or secondary care);
11. All cause mortality (CPRD death data).
Jessica Tyrrell - Chief Investigator - University of Exeter
Jessica Tyrrell - Corresponding Applicant - University of Exeter
Alex Dregan - Collaborator - King's College London (KCL)
Alexandra Gillett - Collaborator - King's College London (KCL)
Jack Bowden - Collaborator - University of Exeter
HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;CPRD Aurum Ethnicity Record;Patient Level Townsend Index;Practice Level Townsend Index