Visualization of Midwives' Professional Work Time a Comparative Analysis Between Obstetric Internal Medicine and Surgical Nursing.

BACKGROUND: Despite growing specialization, the domain-level structure of nurses' and midwives' daily activities remains under-characterized, and few studies compare obstetric with general wards using a single 37-subcategory taxonomy. In Japan, census- and acuity-based staffing under the Medical Care Act/DPC-PDPS captures headcount but underrepresents indirect care such as documentation and coordination. Quantifying domain-level time is therefore needed to align ward staffing with actual workload. We aimed to quantify obstetric midwives' day-shift work time using the JNA 37 - subcategory framework and to compare domain and subcategory profiles, together with direct versus indirect care proportions, with internal medicine and surgical wards. METHODS: We conducted a prospective time-motion study in a Japanese university hospital. Midwives in the obstetric ward and nurses in internal medicine and surgical wards were continuously observed during weekday day shifts (08:30-17:15) using the Japanese Nursing Association's 37-subcategory framework. Eligible staff were licensed midwives/nurses assigned to direct ward care for the full weekday day shift (08:30-17:15) who provided consent; exclusions were float/temporary, supernumerary/administrative, orientation-only, declined consent, and shifts curtailed by emergencies/extraordinary events. Three trained midwife observers performed continuous, one-to-one focal-follow observation, coding activities every 10 seconds with the JNA 37-subcategory taxonomy. To limit observation bias, observers rotated across staff and days, maintained an unobtrusive distance, and emphasized unit-level evaluation and de-identification during consent. The analytic unit was the day shift (obstetric 4 [29%], internal medicine 5 [36%], surgical 5 [36%]). For each domain and subcategory, minutes per shift were summarized as median (IQR). Domain-level differences were tested with Kruskal-Wallis, followed by Holm-adjusted Mann-Whitney pairwise tests. Subcategory differences were tested using one-way ANOVA with Games-Howell pairwise tests. RESULTS: Fourteen day shifts were analyzed. Ward differences were observed for Domain II (Kruskal-Wallis p = 0.0368) and Domain IV (p = 0.0200); only the Surgery vs. Internal Medicine contrast in Domain IV remained significant after Holm adjustment (p = 0.0476). Documentation & Coordination (Domain III) had the highest median in obstetrics, but the omnibus test did not reach significance (p = 0.0888). CONCLUSIONS: Obstetric day shifts showed higher median time in Documentation & Coordination (Domain III), although the omnibus test was not significant; surgery had the highest median in Clinical Settings (Domain II; omnibus p = 0.0368); and Operational Management (Domain IV) differed by ward with only the Surgery vs. Internal Medicine contrast remaining significant after Holm adjustment (adjusted p = 0.0476). These patterns may inform staffing models that explicitly weight indirect-care tasks-especially documentation/coordination-alongside direct care when allocating resources. Future work should extend to nights/weekends and incorporate patient acuity and staffing mix. Generalizability is limited by the single-center, weekday day-shift design.