UCLA

BACKGROUND AND OBJECTIVES: Headache syndromes are highly prevalent, disabling, and costly. Our goals were to (1) describe headache care delivery and costs in a system and (2) identify opportunities for the system to collect, organize, or analyze health care data to facilitate value-based headache care delivery. METHODS: We performed a descriptive, retrospective cohort study using data from a large integrated health system (July 2018-July 2021). We assigned individuals into a reference (REF) or headache group based on headache-related ICD diagnoses. The primary exposure variable, applied to the headache group, was the headache specialty seen most after the incident headache diagnosis: primary care (PC), neurology (NEU), or headache subspecialist (HS). Outcomes of interest were per member per month all-cause costs, per episode costs, all-cause utilization, and headache utilization. Variables included age, sex, insurance contract, and the Adjusted Clinical Groups (ACG) concurrent risk score. We calculated univariate statistics for clinical indicators and outcomes for each group. For outcome variables, we also report these statistics after adjustment for ACG risk score. RESULTS: We identified 22,700 (14%) individuals in the headache groups and 138,818 (86%) individuals in the reference group (REF). Within the headache groups, 84% received care from PC, 14% from NEU, and 2% from HS. The average ACG risk scores increased across exposure groups. In both unadjusted and after risk adjustment analyses, total cost of care (TCOC) was highest in NEU and HS, and the largest drivers of TCOC were outpatient facility costs, followed by inpatient facility costs. HS had the highest pharmacy and professional costs. After risk adjustment, all-cause inpatient admissions and headache-related ED visits were roughly similar, although there was increasing use of outpatient PC and NEU visits across exposure groups. DISCUSSION: Individuals seen by a NEU or HS had higher medical morbidity, higher health care utilization, and higher costs than those who receive care from PC. Outcome data were either not available or not structured to determine the value of neurologic expertise in headache care or within a particular headache care pathway. To clarify neurologys value in primary headache disorders, we encourage health system leaders to adopt an economic evaluation framework.