PHQ-9 Response Curve

Abstract

Major depressive disorder is common in primary care. Depression Improvement Across Minnesota—Offering a New Direction (DIAMOND), using a collaborative care model, was first implemented in March 2008 starting with 5 clinics and expanding to more than 70 clinics statewide by 2010. This was intended to improve depression management and to augment the relationship between the patient, the primary care provider, and the psychiatrist. Prior retrospective studies have demonstrated the clinical effectiveness of our program. This study was designed to examine those patients who were in clinical remission (defined as a Patient Health Questionnaire–9 [PHQ-9] score <5) at 6 months (180 days) after enrollment in collaborative care management. By determining the subsequent PHQ-9 data that were obtained, a PHQ-9 response curve was developed for those patients who did improve. The pilot study demonstrated that there appeared to be rapid response to depression treatment, evident by the first month of treatment and more pronounced in severely depressed patients. Also, it demonstrated that in the patients who did respond, there was no any difference in the remission rates over the study period when evaluated by the initial severity of the depression.

Major depressive disorder is common in primary care. Approximately 6% to 7% of adults will experience depression symptoms over the course of a year, and only half of these will be correctly diagnosed. ¹ Of those treated for depression, typically only 20% to 40% will achieve remission within 12 months.^2,3

Treatment guidelines for major depression recommend targeting remission as the goal of therapy.^4,5 Rush et al, ⁶ in the STAR*D trial, demonstrated a cumulative remission rate of 67% with multiple stepped therapy. With initiation of therapy by the primary care provider, guidelines and research have focused on follow-up of the patients after 6 or 8 weeks of therapy,^4,5,7 although improvement can be seen as soon as 2 weeks. ⁸ If stepped care in primary care involved treatment and follow-up at 6- to 8-week time periods, remission for the individual patient may take several months, prolonging the disease burden of major depression.

In the state of Minnesota in 2005, the Institute for Clinical Systems Improvement (www.icsi.org) convened a group to redesign care for depression in adults in the primary care setting. Depression Improvement Across Minnesota—Offering a New Direction (DIAMOND) was first implemented in March 2008 starting with 5 clinics, expanding to more than 70 clinics by 2010. DIAMOND was based on the Collaborative Care Model ⁹ and the IMPACT study, ³ with the effectiveness demonstrated in a meta-analysis of 37 randomized controlled studies by Gilbody et al. ¹⁰ The model consisted of 6 care components: a systematic screening and diagnostic approach, development of a patient registry, the introduction of a trained care manager, the utilization of treatment guidelines with standardized monitoring of improvement to ensure adequate therapy, and a consultative relationship with psychiatry and the primary care provider. This was intended to improve depression management and augment the relationship between the patient, the primary care provider, and psychiatrist. The collaborative care management (CCM) process involved weekly oversight by a psychiatrist, with medication or therapeutic changes managed by the primary care provider.

DIAMOND was implemented in Rochester, Minnesota, at 2 primary care clinics in 2008 and was expanded to the remaining 4 primary care sites in 2010 (patient population approximately 140 000). Prior retrospective studies have demonstrated the clinical effectiveness of CCM in our program,^11,12 and noted that increased anxiety and an abnormal screening test for bipolar disease had a negative impact on the odds of the patient getting to clinical remission by 6 months. ¹³ Alternatively, this study was designed to examine those patients who were in clinical remission (defined as a Patient Health Questionnaire–9 [PHQ-9] score <5) at 6 months (180 days) after enrollment in CCM. By determining the subsequent PHQ-9 data that were obtained, a PHQ-9 response curve was developed for those patients who did improve. This would give the primary care provider monthly targets for expected PHQ-9 results, similar to a goal target of hemoglobin A1C level for diabetes. PHQ-9 response curves were developed for those patients with moderate depression (PHQ-9 score of 10-14), moderately severe depression (PHQ-9 score of 15-19), and severe depression (PHQ-9 score ≥20).

Methods

Patients were enrolled into CCM for their depression based on clinical diagnosis of major depression or dysthymia and a PHQ-9 score of 10 or greater. The patient or the provider had the option of using CCM or treatment as usual. The study time frame was from March 1, 2008, through June 30, 2010, allowing for potential 180-day follow-up on all patients. The data were obtained by retrospective chart review. The frequency of subsequent PHQ-9 testing was based on clinical status of the patient and CCM guidelines. Only patients who were enrolled in CCM and successfully in remission by 180 days after diagnosis were included in the study. The study points were the index PHQ-9 on admission and then 30, 60, 90, 120, and 150 days following admission to CCM. The minimum PHQ-9 score for ±14 days of the study points (other than baseline) was used, since a PHQ-9 score is valid for this time frame. The percentage of patients who had data at each study point and were in remission (PHQ-9 score of <5) was also measured. The minimum PHQ-9 score was used (versus maximum score) to be consistent with calculation of the remission rate, which looked at a PHQ-9 score of <5.

Rates of remission were evaluated by χ ² analysis. MedCalc software was used for analysis (www.medcalc.org, version 11.6.1). The study was reviewed and approved by the Mayo Clinic Intuitional Review Board.

Results

During the study time frame, there were 359 patients who had been successfully treated for their depression within 180 days while in CCM. All of these patients were in remission of their depression, with a PHQ-9 score of <5. Patients were categorized by the initial severity of their depression, with 194 patients (54.0%) with moderate depression (MD), 118 (32.9%) patients with moderately-severe depression (MSD), and 47 (13.1%) patients with severe depression (SD).

In the MD group, the baseline PHQ-9 averaged 11.80, 5.69 at 30 days (n = 174, 89.7%), 4.79 at 60 days (n = 156, 80.4%), 4.68 at 90 days (n = 141, 72.7%), 4.15 at 120 days (n = 102, 52.5%), and 3.10 at 150 days (n = 96, 49.5%; Figure 1). When the subsequent PHQ-9 scores were normalized to the baseline score of the patient, they were at 48.8% of index at 30 days, 41.6% at 60 days, 40.1% at 90 days, 36.6% at 120 days, and 26.2% at 150 days (Figure 2).

Figure 1.

PHQ-9 scores of patients who were in clinical remission by 180 days after admission to collaborative care management by time.

Figure 2.

Normative PHQ-9 scores of patients who were in clinical remission by 180 days after admission to collaborative care management by time.

Patients with MSD had a mean baseline PHQ-9 score of 16.5 with a 30-day mean of 6.6 (n = 105, 89.0%), 4.8 at 60 days (n = 97, 82.2%), 4.9 at 90 days (n = 87, 73.7%), 4.1 at 120 days (n = 57, 48.3%), and 3.7 at 150 days (n = 61, 51.7%; Figure 1). Normalized to the patient’s baseline PHQ-9 score, the MSD 30-, 60-, 90-, 120-, and 150-day scores were 40.4%, 29.5%, 29.9%, 24.6%, and 22.0%, respectively (Figure 2).

The patients with the initial diagnosis of SD had an average baseline PHQ-9 score of 21.87, with 6.09 at 30 days (n = 45, 95.7%), 4.47 at 60 days (n = 36, 76.6%), 3.45 at 90 days (n = 31, 66.0%), 3.35 at 120 days (n = 23, 48.9%), and 3.00 at 150 days (n = 21, 44.7%; Figure 1). Figure 2 shows the normalized follow-up PHQ-9 scores for this group of 28.0%, 20.7%, 15.6%, 15.1%, and 13.4%.

A targeted goal of treatment of depression is remission. Of the available data at each time frame (only those patients remeasured), there was no difference between the MD, MSD, or SD groups in the percentage of patients whose PHQ-9 scores were <5 (Figure 3). All the groups were near 40% remission rates by 1 month of treatment and more than 70% by 5 months. By study design, the groups were 100% in remission by 6 months.

Figure 3.

Remission rates while in collaborative care management, by time and severity of baseline severity of depression (all rates are P = NS for each study period).

Discussion

With an idealized patient population (those who were in clinical remission after 6 months of treatment in CCM), this pilot study demonstrated that there appeared to be rapid response to depression treatment. This was evident by the first month of treatment and more pronounced in the severely depressed patients. This study also demonstrated that in the patients who did respond, there was no difference in the remission rate over the study period, when evaluated by the initial severity of the depression.

Current clinical practice for depression in primary care varies in frequency and efficacy of follow-up. However, it is not unusual for the patient to have a return visit after 2 months of initiation of therapy. The data from this study would suggest that early follow-up (in this case, with a care manager) and management may decrease the burden of the disease and allow for earlier clinical improvement.

This study gives the clinician a response curve for their patients with depression. In a patient with moderately severe depression after 1 month of treatment, it would not be unrealistic to expect the follow-up PHQ-9 to be 40% of the initial PHQ-9 and by 2 months to be 30% of index. If these goals are not met, the data would suggest that the clinician should evaluate the patient to determine what further treatment modalities may be used. This study would suggest a more aggressive management of the clinical symptoms to the targeted goal of remission.

In prior studies, we have demonstrated that clinical comorbidities decreased the odds of the patient obtaining remission by 6 months. ¹³ This study evaluated only the rate of change of the PHQ-9 score and was not adjusted for demographic or other comorbidities. The patients were from a primary care practice with minimal exclusion criteria for entry into CCM and so may represent “real-world” patients. As a pilot study, the current evaluation was limited by the number of patients, especially those in the severe depression group. A limitation of this study was the absence of statistical evidence for the distinctiveness of our 3-sample normative continuum on depression. It is possible that further evaluation into the impact of demographic and mental health comorbidities would be interesting, but larger study samples would be needed. Another confounding factor was that there was less than 100% data collection at each time frame, which necessitates a prospective study designed with this is mind. Patients with recurrent depression may have a different response curve to depression therapy than those with their first depressive episode. This study included patients with both recurrent and first-time depression.

Conclusions

This study demonstrated a PHQ-9 response curve for patients who had their depression managed in collaborative care and in remission by 180 days. There was no difference in remission rates over the study period based on the initial severity of the depression. Future studies are needed to determine the utility of this in clinical practice.