Every Step Counts: Incremental Mobility and Venous Thromboembolism Risk among Adult General Medicine Inpatients

Hospital-associated venous thromboembolism (VTE) is a leading but preventable cause of morbidity and mortality. Immobility is a key risk factor, yet commonly used scores such as Padua treat it as binary and static, overlooking the physiological and temporal continuum of mobility. The Johns Hopkins Highest Level of Mobility (JH-HLM) scale captures eight milestones across this continuum. We investigated whether lower JH-HLM levels are associated with progressively higher VTE rates and whether this relationship differs by prescribed pharmacologic prophylaxis.

We conducted a retrospective cohort study (2021-2024) of adult medicine inpatients admitted to two academic medical centers. Daily JH-HLM scores were averaged over three consecutive hospital days to define a time-varying mobility exposure. The primary outcome was first radiologically confirmed, symptomatic VTE during the same admission. A pooled Poisson regression model correlated time-varying mobility with subsequent VTE rates, adjusting for day of hospitalization, demographics, comorbidities, Padua elements (excluding mobility), payer, and prophylaxis status.

Among 25,763 patients, 169 developed VTE (0.7%). VTE rates increased stepwise with lower preceding mobility(p=0.001): 0.064, 0.088, 0.121, and 0.195 events per 1,000 patient-days when preceded by a three-day average JH-HLM of level 8 (walk ≥250 ft), 6 (walk ≥10 steps), 4 (move to chair/commode), and 1 (lying in bed), respectively. This relationship did not differ by prescribed chemoprophylaxis(p=0.37).

Lower inpatient mobility conferred a dose-response increase in VTE rates independent of prophylaxis. These findings highlight limitations of binary mobility items in current risk models and support integrating objective mobility assessments into prevention strategies.