Uniform SCI Publication mandates are associated with unequal clinical training burdens: a quasi-experiment comparing two medical training pathways in China

Many medical schools now require scientific papers for graduation, yet few studies have examined how such policies affect trainees’ clinical training time. This study investigated whether a uniform mandatory SCI publication policy was associated with unequal clinical training burdens on two medical training tracks that differed substantially in allocated research time. We conducted a mixed-methods quasi-experiment at a teaching hospital in Shanghai, covering 223 graduates from 2021 to 2026. The cohort included 173 students from an eight-year integrated medical track (123 before a 2017 policy change, 50 after) and 50 students from a four-year graduate-entry clinical track, who served as a contemporaneous control group. The 2017 policy required eight-year trainees to publish original Q1/Q2 (JCR Quartile 1/2) Science Citation Index papers as first or co-first authors to graduate; no equivalent requirement was set for the four-year track. Outcomes included individual SCI publication counts, cumulative impact factor, clinical exam scores, delayed graduation rates, and self-reported weekly time logs. We used Mann-Whitney U tests for two-group quantitative comparisons, Kruskal-Wallis H tests for multi-group contrasts with Bonferroni-adjusted Dunn post-hoc tests, and thematic framework coding for qualitative interview data. After the policy, eight-year trainees’ median SCI publications doubled from 0.5 to 1.0 ( P < 0.001), and cumulative impact factor rose from 2.8 to 6.5 ( P < 0.001). The four-year control group produced zero original SCI papers across all six graduating cohorts. Eight-year trainees’ weekly research hours increased from 8 to 18. While average clinical exam scores showed no meaningful difference (85.7 vs. 86.4, P = 0.34), the additional 10 weekly research hours likely displaced time from bedside clinical training. Interview records confirmed that only trainees with multi-year protected research blocks could complete full research projects; four-year trainees reported fully saturated clinical rotation schedules that left no capacity for original scholarship. Uniform publication requirements were associated with increased research output only for trainees who had protected research time. For those without such time—as reported by all interviewed 4 + 4 trainees and confirmed by their weekly time logs—the same rules imposed additional clinical training burden without scholarly gains. These findings suggest that one-size-fits-all graduation standards may inadvertently exacerbate educational inequities. Tiered criteria that match scholarly expectations to available research time could better preserve clinical training quality and warrant further investigation.