The Relationship between Depression and Adherence in Diabetes
Depression is a common problem among people with either type 1 or type 2 diabetes. It has been consistently associated with poor self-management and health outcomes in diabetes, both cross-sectionally and over time.
Research suggests that depression may affect diabetes health outcomes through its relationship to suboptimal treatment adherence. Here we use the term ‘adherence’ to refer to the degree to which patients follow healthcare provider recommendations regarding prescribed medications and diabetes self-management activities. Although previous work often used the term compliance, this has fallen out of favor because of the passive role implied for the patient. Some may prefer the term ‘concordance’ to further emphasize the non-hierarchical nature of the relationship between doctor and patient in developing a self-management plan and the fact that patients must first agree with recommendations before they can be expected to follow them. Regardless of the term used, it is clear that many diabetes patients do not take medications as prescribed and struggle with self-management. It seems plausible that the symptoms of depression (e.g., concentration difficulties, loss of interest, pessimism about the future) would complicate the already difficult task of diabetes self-management for patients. However, the causal nature of this relationship has not been conclusively demonstrated. Furthermore, subclinical depressive symptoms and diabetes-related distress are more common than psychiatric presentations of depression, such as major depressive disorder, and are also consistently related to problems with adherence and glycemic control. Often, these non-psychiatric constructs may be confounded with measures of ‘clinical depression.’ Thus, assessment and treatment of emotional distress in people with diabetes should consider clinical depression, general distress, and diabetes-specific distress in relation to adherence. Intervention studies for depression in diabetes have failed to demonstrate consistent effects of depression amelioration on improved treatment adherence or diabetes self-management. Thus, treating depression may be necessary but not sufficient for improving adherence and glycemic control. Interventions for depression that also address skills for disease self-management show the most promise.
Major depressive disorder is one of the most common psychiatric disorders, with a lifetime prevalence of approximately 17%. Depression has been associated with reduced psychosocial functioning, increased health care costs and disability. Depression is significantly more common in patients with chronic illness. In particular, it is a significant problem for many people with diabetes and has been linked to poor treatment outcomes. People with type 2 diabetes have been estimated to suffer from major depressive disorder at rates 1.6 to 2.0 times higher than the general population. These estimates suggest that 15% to 20% of people with diabetes experience depression, compared to 2% to 9% of the general population. While type 2 diabetes is more common than type 1 diabetes, available research suggests that rates of depression are equivalently increased in both diseases. Depression in diabetes has been consistently associated with poor metabolic control, more diabetes-related complications and increased risk of mortality in individuals with diabetes. These associations may be accounted for, at least partially, by depression’s relationship to suboptimal treatment adherence.
Adherence to treatment recommendations is extremely important to reduce the risk of complications of diabetes. In addition to adherence to prescribed medications such as oral hypoglycemic agents and insulin, a healthy diet and exercise, self-monitoring of blood glucose (SMBG) and foot care are often recommended for successful disease self-management. Treatment adherence has been shown to reduce morbidity and mortality while enhancing glycemic control among people with diabetes. A meta-analysis of 47 independent studies examining the association between depression and treatment nonadherence in people with either type 1 or type 2 diabetes found that depression was signiﬁcantly associated with non-adherence to the diabetes treatment regimen . Effect sizes were largest for missed medical appointments and composite measures of self-care (i.e., instruments gauging a variety of self-care behaviors). The authors explained that the tendency for diabetes patients with elevated depressive symptoms to miss medical appointments is especially interesting given the interpersonal nature of depression.
Moderate but weaker relationships were observed between depression and diet and medication adherence. The effect for SMBG was small albeit more robust when examined in a continuous rather than categorical manner. Foot care was not signiﬁcantly associated to depression. However, only 2 of the 47 studies reviewed examined this aspect of diabetes self-care, pointing to the need for additional research in this area. No evidence was found to suggest that the association between depression and treatment adherence was different across types of diabetes, however, the effect was stronger in children and adolescents than among adults. A more recent cross-sectional study in pediatric type 1 diabetes showed that the relationship between depressive symptoms and glycemic control was partially accounted for by the indirect effect through SMBG, suggesting that decreased SMBG adherence may explain part of the relationship between depressive symptoms and glycemic control in pediatric type 1 diabetes.
Of note, most of the studies included in the 2008 meta-analysis used measures of depressive symptoms or depressive affect that are not adequate for the identification of major depressive disorder or other psychiatric presentations of depression. In fact, one study that did examine major depressive disorder using the gold standard approach to diagnosis – a structured clinical interview – did not find evidence for a consistent relationship to poorer self-management; instead, general distress and diabetes-specific distress were more closely related to problematic self-management. Furthermore, depressive symptoms remain significantly associated with poorer diabetes self-management even when probable cases of major depressive disorder are excluded from analysis. Thus, despite the consistent association between depression and poor diabetes self-management, there is considerable evidence to suggest that this relationship extends to various presentations of depressive affect and emotional distress and is not limited to cases of major depressive disorder.
Recent research has highlighted the importance of understanding diabetes-specific distress as opposed to more general affective distress. Diabetes-related distress refers to the negative emotional reactions specific to having a diabetes diagnosis, feeling at risk of developing complications, and struggling with the demands of daily self-management, as well as negative appraisals of diabetes treatment providers, and feeling that diabetes negatively impacts one’s social relationships. Diabetes-related distress has been shown to be more prevalent, more long-standing, and more closely tied to treatment non-adherence and poor glycemic control than clinical depression. Although diabetes-related distress can often be confused with clinical depression, it reflects a distinct emotional construct that does not imply psychopathology and is framed within the context of coping with the burdens of diabetes and its treatment. As such, it is likely to benefit from distinct approaches to treatment from those used to treat clinical depression .
Treating depression in people with diabetes is a relatively new focus for clinical research and the current literature is, therefore, limited. Markowitz et al (2011) reviewed 17 independent treatment studies and determined that cognitive behavioral therapy (CBT) and pharmacotherapy with antidepressant medications, such as selective serotonin reuptake inhibitors (SSRIs), were generally effective in ameliorating depressive symptoms in people with diabetes. However, there were mixed results concerning how well these interventions improved glycemic control. Furthermore, no study has demonstrated a clear impact on improved diabetes self-management as a result of depression treatment. Since none of the studies that were reviewed specifically targeted adherence to the diabetes treatment regimen or self-care behaviors, the authors concluded that treating depression may be necessary but not sufficient for maximizing diabetes health outcomes such as glycemic control.
Similar implications can be gleaned from the few more recent randomized controlled trials (RCTs) that have tested integrated models of care for people with poorly controlled type 2 diabetes and comorbid depression. Katon et al. (2010)  evaluated the merits of an intensive intervention for adult primary care patients with significant depression and poorly controlled diabetes, coronary heart disease, or both. Participants were randomized to either a usual care control group or a 12-month intervention aimed at managing depression and improving glycemic control, blood pressure, and lipid control. The intervention employed a “treat-to-target” approach and integrated elements of disease management interventions with collaborative care for depression. It was led by nurse educators and supervised by primary care physicians, a psychiatrist, and a psychologist. Interventions included medication augmentation, prescription of anti-depressants, use of techniques including individualized motivational coaching, goal-setting, problem-solving, and coaching on medication adherence and self-care strategies. These interventions were delivered during structured visits in each patient’s primary care clinic every 2 to 3 weeks, with monitoring of patient data relevant to depression and disease control, with medication adjustments made as necessary. Once participants met treatment goals, they continued to be followed by the nurse every 4 weeks to review depression, adherence, and laboratory test results over 12-months.
The intervention group demonstrated significantly reduced HbA1c, LDL cholesterol, systolic blood pressure, and depressive symptoms relative to usual care controls. They were also more likely to have one or more adjustments in insulin, antihypertensive medication and antidepressant medication. It is noteworthy, however, that adherence to medications, diet, and exercise recommendations was not impacted by the intervention. However, at 12 months post-intervention, both blood glucose and blood pressure monitoring were significantly improved compared to controls. Devices and training in self-monitoring were provided to participants as part of the intervention .
Bogner et al (2012) demonstrated encouraging findings from a less intensive intervention delivered over 12 weeks, which was dually aimed at promoting adherence to diabetes medications and prescribed antidepressants in people with type 2 diabetes. Integrated care coordinators received training on pharmacotherapy for depression and type 2 diabetes and worked closely with primary care physicians to provide guideline concordant treatment recommendations and to monitor adherence and clinical status. This intervention resulted in significantly higher rates of adherence to oral hypoglycemic agents and antidepressants, as well as reduced HbA1c levels and fewer depressive symptoms immediately post-intervention, relative to usual care. However, participants were not followed over time following the conclusion of treatment. Safren et al. (2013)  also used an integrative approach to depression management in type 2 diabetes and showed that 10-12 sessions of one-on-one Cognitive Behavioral Therapy for Adherence and Depression (CBT-AD) - in which clinically people with type 2 diabetes and depression were treated by a collaborative team, which comprised a psychologist, a nurse educator, and a dietitian, led to increased medication adherence immediately post-intervention and across 8-and 12-month follow-ups compared to enhanced usual care controls. SMBG was also significantly improved among intervention participants relative to controls, post-treatment. While there was some decrement in SMBG following 8-month and 12-month follow up assessments, intervention participants continued to exhibit superior rates of SMBG compared to the control group over the follow-up. Finally, HbA1c levels were significantly lower for the intervention group compared to controls immediately post-intervention, with gains maintained over the follow-up. Although improvements in depression severity were significant at post-treatment for intervention participants, relative to controls, the significance of this difference was not maintained over the follow-up. Thus, this provides further evidence of some level of independence between improvements in self-management and glycemic control on one hand and changes in depression on the other.
In conclusion, depression is a common problem among people with either type 1 or type 2 diabetes. It has been consistently associated with poor self-management and health outcomes in diabetes, both cross-sectionally and over time. Research suggests that depression may affect diabetes health outcomes through its relationship to suboptimal treatment adherence. However, the causal nature of this relationship has not been conclusively demonstrated. Furthermore, subclinical depressive symptoms and diabetes-related distress are more common than psychiatric presentations of depression, such as major depressive disorder, and are also consistently related to problems with adherence and diabetes control. Often, these non-psychiatric constructs may be confounded with measures of ‘clinical depression.’ Thus, assessment and treatment of emotional distress in patients with diabetes should consider clinical depression, general distress, and diabetes-specific distress in relation to adherence. Intervention studies for depression in diabetes have failed to demonstrate consistent effects of depression amelioration on improved treatment adherence or diabetes self-management. Thus, treating depression may be necessary but not sufficient for improving adherence and glycemic control. Interventions for depression that also address skills for disease self-management show the most promise.
^ Gonzalez, J.S et al.(2008). Depression and diabetes treatment A meta-analysis. Diabetes Care, 31(12), 2398-2403. 10.2337/dc08-1341.
^ McGrady, M.E et al (2009). Depressive symptoms and glycemic control in adolescents with type 1 diabetes. Diabetes Care, 32(5), 804-806. 10.2337/dc08-2111.
^ Fisher, L et al (2007). Clinical depression versus distress among patients with type 2 Not just a question of semantics. Diabetes Care, 30(3), 542 -548. 10.2337/dc06-1614
^ Gonzalez, J.S et al (2007). Depression, self-care, and medication adherence in type 2 relationships across the full range of symptom severity. Diabetes Care, 30(9), 2222-2227. Epub 2007 May 29.
^ Gonzalez, J.S et al. (2011). Depression in Have we been missing something? Diabetes Care, 34(1), 236-239. 10.2337/dc10-1970.
^ Markowitz, S.M et al. (2011). A review of treating depression in Emerging findings. Psychosomatics, 52(1), 1-18. 10.1016/j.psym.2010.11.007.
^ Katon, W.J et al. (2010). Collaborative care for patients with depression and chronic illness. The New England Journal of Medicine, 363(27), 2611-2620. 10.1056/NEJMoa1003955.
^ Lin, E.H.B et al(2012). Treatment adjustment and medication adherence for complex patients with diabetes, heart disease, and A randomized controlled trial. Annals of Family Medicine, 10(1), 6-13. 10.1370/afm.1343.
^ Bogner, H.R et al (2012). Integrated management of type 2 diabetes mellitus and depression treatment to improve medication A randomized controlled trial. Annals of Family Medicine, 10(1), 15-22. 10.1370/afm.1344.
^ Safren, S.A et al (In Press). A randomized controlled trial of cognitive behavioral therapy for adherence and depression (CBT-AD) in patients with uncontrolled type 2 diabetes._ Diabetes Care_.