- The complexity of treatment options available for patients with depression has resulted in undertreatment and low response and remission rates, in both practicing outpatient psychiatric and primary care settings.
- In the STAR*D trial, utilization of a measurement-based care module with an automated feedback system to physicians increased physician adherence to treatment protocols and the quality of patient care. Note: This study/paper does not really address clinical outcomes.
The Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial, funded by the National Institute of Mental Health, National Institutes of Health, was the largest prospective, randomized trial to date to assess the treatment of depression in real-world outpatient psychiatry and primary care settings. The patient population consisted of 4041 self-referred outpatients, aged 18 to 75 years, who scored at least 14 on the 17-point Hamilton Rating Scale for Depression (HRSD17).
One important aspect of the STAR*D trial design was to evaluate which antidepressant agents were most effective in patients who had an unsatisfactory response to an initial agent. Patients with depression frequently show inadequate responses, due to underdosing of antidepressants, inadequate follow-up, or treatment abandonment prior to adequate duration of antidepressant therapy. In addition, antidepressant therapy may be switched prematurely or prolonged despite absence of symptom relief or remission. Furthermore, clinicians differ in their evaluation of treatment outcomes, including their assessment of side effects. One study reported that only 19% of patients received adequate treatment for depression in the primary care setting, that only two fifths reached the recommended dosage of antidepressant medication, and that only one third continued to take the prescribed agent past 1 month. Even when patients are treated by depression care specialists, remission rates are low, about 15% to 30% in public sector patients.
The STAR*D trial was designed to include a measurement-based care (MBC) database information system, which had three primary objectives designed to ensure adequate and safe treatment delivery to patients:
- Systematically determine adherence to the STAR*D protocol to identify deviations from protocol and provide opportunities for rapid correction
- Document physician adherence to treatment recommendations
- Show the feasibility of the feedback and prompt system, which included automated methods of enhancing adherence to a treatment algorithm, as well as showing that this system can be used in routine clinical practice
Attributes of the MBC system included criteria for study physicians to rate depression symptom severity, side effect frequency and intensity, and burden of disease.
Physicians and clinical care coordinators (CRC) were trained on the brief clinical tools and the clinical decision points employed in the study. STAR*D CRCs uploaded responses to physician and patient assessment tools, including:
- The Quick Inventory of Depressive Symptomatology-Clinician-rated (QIDS-C16)
- The self-report version of the QIDS-C16 (the QIDS-SR16)
- The Frequency, Intensity, and Burden of Side Effects Rating (FIBSER)
- The Patient-Rated Inventory of Side Effects
The MBC database provided study physicians with each patient’s medication, its dose frequency and duration, and the response to treatment. The MBC also included an automated feedback system (AFS), intended to alert study physicians when treatments deviated from STAR*D protocols. The AFS was implemented to guide physicians’ dose adjustments, monitor physician adherence to treatment duration protocols, and enhance treatment decisions. The AFS was used at point of care and in patient follow-up. For example, physicians received feedback on treatment decision points if a drug dose was being maintained too long without adequate response, or if the side effects of a drug were too burdensome and the dose needed to be lowered.
Investigators found physicians demonstrated an 85% clinical adherence rate to STAR*D treatment recommendations, suggesting that only 15% of visits were outside the study protocol (a mean of 3.48 visits per patient). Most AFS alerts were due to the extension of treatment at the 9- and/or 12-week visits, although the STAR*D protocol recommended moving to the next treatment level. Many deviations from treatment protocols were found to have acceptable rationales.
The MBC system may also have additional benefits for patients. This system may enhance patients’ understanding of the goals and objectives of treatment, allowing them to actively collaborate with their physicians and monitor their progress using objective criteria. The MBC system also enhanced quality of care at levels 1 and 2 of the STAR*D protocol.
The findings of this study suggest that the MBC component of the STAR*D study provides a shift toward point of care decision support and patient follow-up, thus optimizing pharmacotherapy. The authors concluded that MBC systems are feasible and can be applied to the treatment of depression in both the psychiatric and primary care settings. Unresolved issues that need to be addressed prior to broad implementation of MBC treatment algorithms include physician time for data entry, maintaining confidentiality, and portability of computer software. Acceptance of a guided MBC therapy system is more likely in centers with electronic medical records and computerized physician order-entry systems.
Trivedi M, Rush A J, Gaynes B, et al. Maximizing the adequacy of medication treatment in controlled trials and clinical practice: STAR*D measurement-based care. Neuropsychopharmacology.2007;32:2479-2489.