HSR&D Home » Research » RRP 10-105 – HSR&D Study
Determinants of Guideline-Concordant Depression Care for COPD Patients
Vincent S. Fan, MD MPH
VA Puget Sound Health Care System Seattle Division, Seattle, WA
Funding Period: October 2010 - March 2012
The prevalence of depression in COPD patients is estimated between 25-50%. The rate of depression is significantly higher in COPD patients than in patients with heart disease, stroke, diabetes, arthritis, and cancer. There is evidence that depressed COPD patients are less likely to get adequate antidepressant therapy compared to patients with heart disease, diabetes and osteoarthritis.
To assess quality of depression care for Veterans with COPD, and patient and facility-level factors associated with guideline-concordant depression care.
This was a prospective observational cohort study in which Veterans with depression were identified from the VHA External Peer Review Program (EPRP) data collected by the Office of Quality and Performance. Patients in EPRP represent a random sample of patients in primary care clinics from all VA facilities. We included outpatient groups that are oversampled: diabetes, women, prior myocardial infarction, spinal cord injury, heart failure, post-traumatic stress disorder (PTSD), and depression. EPRP data was linked to national VA administrative data including pharmacy (Corporate Data Warehouse), inpatient (Patient Treatment File) and outpatient (Outpatient Care Files) data.
Patients were excluded if they had a diagnosis of bipolar disorder or schizophrenia, died during the 6 months follow-up period, had no primary care visits during the year before, or no primary care/mental health visits in the year after the depression diagnosis.
The focus of this analysis was on patients with chronic depression, and we excluded those with a new diagnosis of depression. The primary outcomes during the 6-month follow-up period were adequate medication treatment (80% antidepressant refill adherence) or adequate psychotherapy (6 psychotherapy visits). The primary predictors were medical conditions from the AHRQ Clinical Classification Software (CCS), which classifies all inpatient and outpatient ICD-9 codes into 259 categories. We removed psychiatric diagnoses from the CCS. Psychiatric comorbidity was identified from the CCS categories (e.g. anxiety, substance abuse), as well as EPRP data (PTSD).
Additional covariates included race, socioeconomic status, gender, age, marital status, current drinking and smoking behavior. Using administrative data, we calculated the number of mental health and primary care visits in the last year, primary care location (CBOC vs VA facility), and distance to primary care. Because patients who receive depression treatment in primary care only may differ significantly from those who receive care in mental health, we performed separate analyses for each treatment setting.
Separate logistic regression models were created, clustering on VA facility, to model the association between COPD and other comorbid medical illness with adequate medication or psychotherapy treatment for depression. Comorbidity (CCS score) was modeled as quintiles.
Of the 302,025 patients in EPRP during FY2009 and FY2010, 91,952 patients had a recent visit or a positive screen for depression. 21,352 patients failed to meet the inclusion criteria or only had a positive PHQ 2/9 screen, leaving 77,421 subjects. 79.7% had chronic depression of which 79.3% received some care in mental health clinic during the treatment period, and 21.7% were treated in primary care only. Patients treated in mental health were younger (mean age 56 vs. 62, p<0.0001), more likely female and non-White than those treated in primary care. Although more mental health patients were in the highest quintile of CCS score compared to primary care (20.6% vs. 16.7%; p <0.0001), primary care patients were more likely to have chronic illnesses such as COPD (18.7% vs. 16.5%; p<0.0001), CHF (10.9% vs. 5.8%, p<0.0001), and diabetes (38.5% vs. 34.5%, p<0.0001) than those in mental health.
Compared to primary care only, mental health patients were more likely to receive appropriate medication and psychotherapy depression care. 55.6% of those in mental health had > 80% antidepressant adherence, compared to 35.2% in primary care (p<0.0001). 16.3% of mental health patients had 6 psychotherapy visits during the 6-month follow-up treatment and 7.4% had both.
In adjusted logistic regression models for patients treated in primary care, only the highest quintile of comorbidity (CCS score >20) was associated with better antidepressant adherence (OR 1.27, p<0.01). In mental health, higher quintiles of medical illness was associated with better antidepressant adherence (e.g. compared to a CCS score of 0-4, a CCS score 11-14 had an OR= 1.13, p<0.0001; a CCS 15-19 had an OR=1.17, p<0.0001; and CCS 20+ had an OR= 1.19, p<0.0001). Overall comorbidity score was not associated with adequate psychotherapy among those treated in Mental Health.
Increasing baseline primary care visits was associated with better medication adherence in both primary care and mental health settings. Increasing baseline mental health visits was strongly associated with both adequate medications and psychotherapy among those in mental health. There was no difference in medication therapy for depression for individual medical comorbidities (COPD, coronary artery disease, heart failure, stroke, and diabetes), although slightly lower psychotherapy adherence was seen for COPD (OR 0.90, p=0.003), diabetes (OR 0.93, p=0.06) and heart failure (OR 0.80, p= 0.02) patients treated in Mental Health Clinics.
Data from this study has shown that chronically depressed patients with COPD don't significantly differ in receipt of appropriate depression care when controlling for all other comorbidities and patient factors. In addition, we found that increasing comorbidity was associated with better adherence to antidepressants. Increasing contact with both primary care and mental health clinics are also associated with adequate medication and psychotherapy. This suggests that increased medical comorbidities, and increased contact with providers provide more opportunities to initiate treatment and provide follow-up for patients with chronic depression.
External Links for this Project
Dimensions for VADimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.
If you have VA-Intranet access, click here for more information vaww.hsrd.research.va.gov/dimensions/
VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address. Search Dimensions for this project
DRA: Mental, Cognitive and Behavioral Disorders, Cardiovascular Disease
DRE: Treatment - Observational, Prognosis
MeSH Terms: none