Abstract Body

Background and purpose: Persistent somatic symptoms (PSS) are highly prevalent in primary care and are a burden to both the patient and healthcare. Data-based identification of patients with PSS could foster various improvements in care. However, identification is currently hampered by unambiguous registration of PSS. The present study aims to explore different theory-driven methods for data-based identification of patients with PSS.

Methods: A cross-sectional study was performed on routine primary care data from 169,138 patients in the Netherlands. Identification of PSS was based on (A) PSS-related syndrome codes, (B) PSS-related terminology, (C) PSS-related symptom codes, and (D) 4-dimensional symptom questionnaire (4DSQ) scores. Sample size, demographics, chronic conditions, and health care utilization (HCU) as extracted via the four methods were explored. Sensitivity between methods was examined.

Results: The percentage of cases identified varied between 0.3% and 7.0% across the methods. Over 58.1% of cases had comorbid chronic physical condition(s) and over 33.8% had comorbid chronic mental condition(s). HCU was generally higher for cases selected by any method compared to the total cohort. HCU was relatively higher for method B compared to the other methods. In 26.7% of cases, cases were selected by multiple methods. Sensitivity between methods was generally low.

Conclusions: The different methods yielded different patient samples within our cohort. Therefore, for the most comprehensive data-based selection of PSS cases, a combination of methods A, C, and D would be recommended. Additionally, advanced (data-driven) methods are needed to create a more sensitive algorithm for identifying the full spectrum of PSS.