Unexplained chronic fatigue (UCF) includes both chronic fatigue syndrome and idiopathic chronic fatigue. Few management strategies for UCF have proven effectiveness. However, practitioners of complementary and alternative medicine (CAM) and physicians specializing in the treatment of UCF (who often use non traditional practices) report that their treatments are often successful. We propose to test whether these providers have better results than community physicians in practice-based research networks (PBRNs). This prospective cohort study with six month follow-up will include 240 patients in four groups defined by the background of the treating clinicians: 1) MDs and DOs in 2 PBRNs (n=60), 2) MDs trained in CAM (n=60), 3) naturopathic doctors (n=60), and 4) 2 MDs who specialize in treating UCF (n=60). Study patients must be new to the practice of the clinician and have had debilitating fatigue for more than six months that cannot be explained by well-understood physical or psychological problems. Patients will provide baseline information about prognostic factors and baseline and follow-up information about fatigue, associated symptoms, and functional status. For each patient the participating clinicians will provide information about the patient's clinical characteristics, treatments given the patient, and clinician time spent with the patient. The clinicians will also provide information about their own attitudes concerning treatment of psychosocial problems. Differences between and within the four groups of patients will be determined for patient prognostic characteristics, treatments received, and outcomes (primarily changes in fatigue and number and severity of somatic symptoms). The influence of patient prognostic factors on outcome differences will be reduced by eliminating some patients on the basis of their propensity scores for being treated by UCF specialists and using regression methods to adjust for residual confounding due to differences in measured prognostic factors. Analyses will also be done in subgroups of patients such as those without evidence of major psychosocial problems or who have the somatic symptoms used to define chronic fatigue syndrome. The extent that differences in risk-adjusted outcomes can be explained by the time spent with the patient and clinician attitudes will be examined. If risk-adjusted outcomes for the four groups and for subgroups within these groups are apparently related more to treatment used than physician attitudes or time spent with the patient, then follow-up studies should be designed to identify specific approaches that are most effective for treating patients with UCF. The proposed study will evaluate whether complementary and alternative medicine treatments currently ? used for UCF are effective. Subsequent studies will be needed to identify which specific approaches are most effective. The proposed methodology is novel and might be useful for identifying effective treatments for other conditions. ? ? ?
