Practitioners' Prognoses: Potential Clinical and/or Research Tool?

Dear Editor:

There are many challenges to consider when designing complementary and alternative medicine (CAM) research, including how to measure outcomes; how to recruit patients—biomedical or CAM-specific diagnoses; placebo effects; intervention complexities; patients' previous experiences, attitudes, preferences, and expectations; and the patient–practitioner relationship. To overcome these challenges, there appears to be a growing consensus that pragmatic studies with nested qualitative components and flexible treatment protocols offer potential solutions. Additionally, a report by Sherman and Cherkin recommends that studies provide information on practitioners' perceptions of patients' suitability for treatment. ¹ Inspired by this recommendation, I asked practitioners treating patients with irritable bowel syndrome (IBS) in a pragmatic trial of acupuncture to identify patients' prognoses. The following paragraphs present an overview of the analyses and findings ² that may be used to generate discussion on the potential usefulness of prognoses as clinical and/or research tools.

Traditional Chinese Medicine practitioners in a pragmatic trial of acupuncture for IBS were encouraged to deliver treatments similar to their routine care. ³ At the end of each patient's initial treatment, the acupuncturists identified his/her prognosis (e.g., good, poor, don't know). Of the 105/113 patients to receive a prognosis, 47 (45%) received a “don't know” prognosis, 42 (40%) received a “good” prognosis, and 16 (15%) received a “poor” prognosis. To determine whether the prognoses were predictive of outcome, data were analyzed using multiple linear regressions with “poor prognosis” as the reference category.

The exploratory analysis suggests that prognosis is a significant predictor of outcome (Table 1). Patients with a “good prognosis” achieved a five-point increase in the Physical health Composite Score (PCS) and a seven-point increase on the Mental health Composite Score (MCS) of the SF-12 in comparison to patients with a “poor prognosis.” Patients with a “don't know prognosis” also scored seven points higher on the MCS than patients with “poor prognosis.” The findings are the first empirical evidence to suggest that acupuncturists' prognoses may be a useful predictor of outcome.

Although discussion worthy, the analyses have several limitations including small sample sizes that may reduce power of the calculations. Acupuncture trials are also subject to clustering of outcomes for patients treated by the same practitioner. ⁴ Therefore, further analyses with larger samples and that account for clustering are needed to confirm these exploratory findings.

With regard to potential applicability and/or usefulness, the prognoses' ability to predict outcome suggests that practitioners may provide important information about whether patients are likely to benefit from acupuncture. Therefore, prognoses may serve a practical role in the referral process such as recommending that patients with poor prognoses be offered a different intervention. Prognoses may also be used in research during allocation wherein practitioners assess patients prior to allocation and based on their prognoses, allocate patients to different study arms. Overall, the recommendation by Sherman and Cherkin that practitioners should report a patient's suitability for treatment ¹ and the prognosis data present two areas that may provide information that leads to more effective patient care.