Effect of Depression on Glycemic Control in Type Two Diabetes Mellitis

(My Original Blog Post: http://anxiety-helper.car-manuals.biz/effect-of-depression-on-glycemic-control-in-type-two-diabetes-mellitis)

Introduction:


The relationship between Diabetes and clinical depression is a subject of growing interest. The diabetic population has a 2-fold risk of depression1 compared to general population, 10-15% meeting the criteria for clinical depression2 and as high as 40% prevalence of sub-clinical depressive symptomatology3. Depression can severely impact medical management of Diabetes in terms of higher symptom burden4, increased functional impairment4,5, poorer glycemic control6 and more diabetic complications 7,8.



The course of depression is more chronic and severe in people with Diabetes9. 20% recover completely for 5 years and on average this population suffers 4.2 episodes in 5 years10. Symptoms of depression and diabetes may exacerbate each other at neuroendocrine level11. Depression in this population also remains under-diagnosed and under-treated12, 13 because of symptom overlaps14 and it commonly viewed as secondary to medical condition hence not independently important.



Compliance to medication5, diet15 and exercise4 is central to prognosis in this chronic illness. Depression has been most commonly associated with non-adherence in foreign16,17,18 and local literature19. Depressed mood leads to increased pessimism with respect to perceived benefits and reduced self-efficacy, both being crucial motivating factors for self-care and compliance 20.



Local literature shows high non-compliance21,22 and majority of diabetics having poor glycemic control23,24,25,26. Previous studies27,28,29,30 show various reasons for non-compliance, most being forgetfulness, belief of immediate cure and financial constraint. This study aims to find an association between depression and treatment compliance in the diabetic population of a tertiary care hospital in Karachi. Operational definitions are attached as Annexure 1.



Objectives:


1. To find the prevalence of depressive disorders in the diabetic population coming to a tertiary care hospital of Karachi.


2. To observe the effect of depression on glycemic control in the same population.



Hypothesis:


The group of diabetics having depression will have poorer glycemic control as compared to those who do not have depression.



Material and Methods:


Study Design: Cross-sectional observational study


Study Setting: The Aga Khan University Hospital is a 500-bedded tertiary care hospital



catering to middle and upper socio-economic strata suffering from Diabetes Mellitis. At the hospital there are 16 clinics every week specific to Diabetes and Endocrinology. American Diabetic Association recommendations of diet, exercise, glycemic monitoring and investigations are followed on all patients.



Duration Of Study: 6 months after protocol approval.



Sample Size: Sample size calculated by the formula for binomial variables in two groups:


h‘= [zα√(C+1)P(1-P) + zβ√CP1(1-P1)+P2(1-P2)]2


C(P2-P1)2


is 280 keeping p-value of 0.05 and power of the study 80%.



Sampling Technique: Non-probability convenience sampling will be done.



Sample Selection:



Inclusion Criteria:


1. Patients having Type II Diabetes and its related complications (Annexure 1)


2. Age between 18 and 60 (Adult population only)


3. Patients who have at least taken one appointment at the endocrinology clinic before



Exclusion Criteria:


1. Patients having history of psychiatric illnesses other than depressive disorders (Annex 1)


2. Patients having history of use of psychotropic drugs


3. Patients unable to give informed consent



Data Collection Procedure:



Outcome Variables:



Dependent variable:


1. HbA1c level (poor glycemic control 7%) done 3 months after the recruitment time


2. Number of diabetes-related complications (Coded in Annex1)



Independent variable:


Score on Hospital Anxiety and Depression Scale (Depression Subscale with cut-off ≥8)



Method of Data Collection:


Informed consent after nursing assessment will take place for those fulfilling the criteria; those agreeing will undergo a semi-structured clinical interview of 10 minutes before consultation with endocrinologist including demographic details, history of psychiatric illness or exposure to psychotropic, family history of diabetes and depression and self-care activities. The same sample will be followed on next visit in 6 months for HbA1c level measured using few drops of blood on High-Protein Liquid Chromatography method (HPLC).



Data Collection Tools:


1. Measurement of Depression: The Urdu Translation of Hospital Anxiety and Depression Scale31-depression subscale has been used in various studies. It has good sensitivity, specificity and receiver operating characteristics when compared to other measures of depression32,33. It will screen for case-ness of depression in this study (Annex2) with a cut-off score of ≥8


2. Measurement Of Compliance: A proxy measure of compliance will be used Glycosylated Hemoglobin (7% poor compliance).



Self-care activities will be part of the interview. Weight, blood pressure and diabetes-related complications that develop will be followed in case records for change indicating treatment compliance.



Data Analysis Procedure:


Statistical Software


Statistical Package for Social Sciences Version 13.0 (SPSS 13.0)


Statistical Tests


A 2x2 table will be constructed as follows:


DISEASE / HbA1c>7%



Exposure: Good Glycemic control: Depressed: a: b


HbA1c

Odds Ratio (ad/bc) will be calculated and Chi-square test of significant applied to primary variables of HAD-D score and HbA1c level.



Stepwise logistic regression will be applied using glycemic control and diabetes-related complications as dependent variables, controlling for demographic details and duration of illness. A bivariate analysis for depression and different self-care behaviors asked during the interview will be conducted to find the most effected self-care behavior.