Identification of patients with moderate medically unexplained physical symptoms in primary care with a five years follow-up

Background: Patients with medically unexplained physical symptoms (MUPS) are common in primary care, with a spectrum from mild to moderate and chronic MUPS. The burden of chronic MUPS is high, and early identification of moderate MUPS patients is important to prevent chronicity. The PRESUME screening method to identify moderate MUPS patients in primary care was developed, but insight in prognostic accuracy is needed. Therefore, our objective is to determine the prognostic accuracy for identification of moderate MUPS patients using the screening method with 5 year follow-up. Methods: The PRESUME screening method consists of three subsequent steps based on consultation frequency, exclusion of medical/psychiatric diagnosis and identification of MUPS. In a random 10% sample of patients from the Julius General Practitioners Network (n = 114.185), patients were identified with mild, moderate or chronic MUPS in 2008 (index year), using routine care data. In 5 years follow-up we calculated predictive values and odds ratio's for sustained MUPS related symptoms. Results: In 2008, 789 patients (6.9% of the patient population) were identified as having mild, moderate or chronic MUPS. On average 55.5% of the moderate MUPS patients in 2008, still had MUPS related symptoms or developed chronic MUPS in 5 year follow-up. Positive predictive values for maintaining MUPS related symptoms or worsening was 67% after 1 year, and 48.7% after 5 years for moderate MUPS patients. Conclusion: The prognostic accuracy of the PRESUME screening method using electronic medical record data for identification of moderate MUPS patients is moderate. However, it might be a useful method to identify patients at increased risk of moderate MUPS, if combined with a validity check by the GP.