Administrative Burden Documented in Medicaid Care Coordination

Vollständiger Abstract

Worum geht es in dieser Arbeit?

Importance Administrative burdens, the time, effort, and stress of navigating bureaucratic requirements, impede Medicaid beneficiaries and persist even when care is obtained, yet their frequency and patient time costs are largely invisible in claims data. Objective To measure administrative burdens documented in Medicaid care coordination notes by type, quantify the patient time costs they entail at multiple wage valuations, and characterize differences by race and ethnicity. Design, Setting, and Participants Retrospective cohort study of Medicaid managed care beneficiaries in a single community-based care coordination program across Washington, Virginia, and Ohio from January 2023 through November 2025. Natural language processing classifiers identified administrative burdens in encounter notes. Exposures Four documented administrative burden types: scheduling difficulties, transportation problems, paperwork and documentation requirements, and prior authorization delays. Main Outcomes and Measures Patient-level prevalence and incidence by burden type; encounter-normalized rates; patient time costs at the federal minimum wage ($7.25/h), a state-weighted living wage ($22.00/h), and the Resource-Based Relative Value Scale conversion factor ($33.40/h); and differences by race and ethnicity. Results Of 142 473 Medicaid beneficiaries (mean [SD] age, 31.6 [18.5] years; 55.3% female; 44.7% male) enrolled in the 4 participating managed care plans during the study period, 49 282 beneficiaries (34.6%; mean [SD] age, 33.7 [18.8] years; 60.5% female; 26.2% African American; 1.1% American Indian or Alaska Native; 3.8% Asian; 7.6% Hispanic; 0.9% Native Hawaiian or Other Pacific Islander; 35.0% White; 2.2% other race and ethnicity; 23.2% unknown ethnicity), completed at least 1 care coordination encounter. Patient-level prevalences were 6.1% (95% CI, 5.9%-6.3%) (transportation), 25.3% (95% CI, 24.9%-25.7%) (paperwork), 16.2% (95% CI, 15.8%-16.5%) (scheduling), and 9.8% (95% CI, 9.6%-10.1%) (prior authorization). Per affected patient, the mean time cost at the clinician-equivalent valuation was highest for transportation ($47.58; 95% CI, $45.75-$49.52) and lowest for scheduling ($12.35; 95% CI, $11.99-$12.73); documented burdens corresponded to 18 822 patient-hours cohortwide ($628 665). Encounter-normalized rates preserved this ordering, which was also robust to classifier misclassification. After excluding the 23.2% of beneficiaries with unknown ethnicity, a higher unadjusted burden prevalence was found in African American beneficiaries than White beneficiaries (rate ratio, 1.22; 95% CI, 1.18-1.25); payer-stratified analysis indicated this aggregate difference reflected enrollment in higher-burden plans rather than within-plan differences (<2 percentage points). Beneficiaries who engaged with care coordinators were older, more often female, and higher acuity than eligible beneficiaries who did not engage. Conclusions and Relevance In this cohort study, administrative burdens in Medicaid varied markedly by type in both prevalence and patient time cost. Natural language processing of existing care coordination notes offers a scalable way for managed care plans and state Medicaid agencies to monitor which burdens are most common and most time-consuming.