BOCA RATON, FLA. – Patients with major depressive disorder who provided their primary care physician with monthly ratings of their symptoms were more likely to respond to antidepressant treatment and to achieve remission compared with a usual care group, findings of a prospective, observational cohort study have shown.
In a multivariate analysis that controlled for significant differences in baseline characteristics, patients who provided monthly updates were twice as likely to respond to therapy (odds ratio, 2.02) and 59% more likely to achieve remission (OR, 1.59) compared with the usual care group.
Dr. Trina Chang
A total of 83 primary care physicians enrolled 915 patients with major depressive disorder in the Clinical Outcomes in Measurement-Based Treatment (COMET) trial. Physicians were alternatively assigned to an intervention group that received monthly patient symptom ratings on the 9-item Patient Health Questionnaire (PHQ-9) or a usual care group, with PHQ-9 feedback only at the 6-month study end. A total of 380 patients in the intervention group and 284 in the usual care group completed all measures and were assessed further.
Usual care may be insufficient to monitor patient response to antidepressant treatment in the primary care setting, Dr. Trina Chang said at the annual meeting of the New Clinical Drug Evaluation Unit, sponsored by the National Institute of Mental Health.
"We found there is not a lot of change going on in this first 6 months in these patients with major depression," Dr. Chang said in an interview. For example, 43% of 253 patients who received usual care had no treatment changes during the study.
"There are a couple of different ways to interpret it. The wishful thinking way ... would be [that] all our patients got better so they didn’t need treatment changes," said Dr. Chang, a psychiatrist at Massachusetts General Hospital and Harvard Medical School in Boston.
"Unfortunately, because the study also found [that] less than half of patients go into remission, probably more of them should have been having changes" to their treatment regimen.
In the usual care group, 58% scored 5 or higher on the PHQ-9 at 6 months and were considered nonremitters.
In a subset of the usual care group that had 6-month PHQ-9 scores, among participants who had less than a 25% improvement at that time compared with baseline, only 15% had a treatment change during the study. The findings "indicate an area where there could be a lot of improvement," she said. If the remaining 85% of patients were to get treatment adjustments, "we could do a lot to improve depression outcomes."
None of the participants had received treatment within 120 days prior to enrollment. The primary care physicians were free to choose any kind of antidepressant therapy. Most patients started monotherapy that included selective serotonin reuptake inhibitors for 63%, selective norepinephrine reuptake inhibitors for 25%, or bupropion for 10%. Only 0.4% of patients received an initial augmentation treatment.
Of the patients who received a change in treatment, 17% switched antidepressants; 16% had a dose titration of their initial medication (up or down); 5% had another type of medication added (for example, a benzodiazepine); 3% received augmentation with an antipsychotic, stimulant, or antiepileptic; and 2% received a second antidepressant. The remaining 57% of patients discontinued treatment.
Mean age was 47 years in the intervention group and 45 years in the usual care patients. Women made up 64% of the intervention group and 67% of the usual care group. There were some baseline differences between groups, including a significantly greater proportion of intervention group patients who had Medicaid insurance, were unemployed, and were black or African American. In contrast, the usual care group had a significantly greater proportion of participants who were employed full time, had a 4-year college degree, were married, and were white, Hispanic, or Latino.
Remission was achieved in 48% of the intervention group versus 43% of the usual care group. Complete response rates were achieved in 67% and 60% of the groups, respectively. At 6 months, partial responses were noted in 16% of the intervention and 18% of the usual care groups.
Potential solutions can involve both providers and patients, Dr. Chang said. "On the provider end, anything that could help them detect symptoms and follow up on patients better ... would help." In addition, symptom reporting coupled with treatment algorithms and/or a collaborative care model with a care manager who could more frequently and closely monitor patient treatment response could help, she said.
"On the patient end, anything that might help them to not discontinue their medications would help," she added. Making patients feel supported and answering their questions could all bolster compliance with treatment and possibly improve outcomes.