Table 2_Beyond GCP: developing a differentiated competency framework for principal investigators in innovative drug clinical trials in China.docx

The second-round consultation was conducted using the streamlined indicators.Panel composition changed between rounds: some Round 1 experts did not participate in Round 2, and two regulatory experts were added to strengthen stakeholder representation and content validity. The Statistical Package for the Social Sciences (SPSS) 26.0 was utilized for quantitative analysis, encompassing descriptive statistics, the calculation of Kendall's W and the coefficient of authority (Cr).</p>Results<p>An exploratory framework comprising 30 competencies was established.

The mean expert authority coefficient was 0.87. The test of concordance was statistically significant (p < 0.001), but Kendall's W was 0.206, indicating limited inter-expert concordance. Indicators were categorized into three descriptive strata by coefficient of variation (CV): high-consensus (CV ≤ 7%; 4 items), medium-consensus (7% < CV ≤ 17%; 19 items), and low-consensus (high-divergence) (17% < CV < 25%; 7 items).

These strata describe variability in the panel's ratings and do not establish differences in competency importance or a developmental sequence.</p>Conclusions<p>This exploratory 30-item inventory provides an initial structure for evaluating PI competencies in China's innovative-drug trial context. Its differentiation logic offers preliminary perspectives to inform the future development of differentiated and context-sensitive PI competency models.

While this framework may inform future training system design, institutional human-resource management, and PI career development, its long-term validity, feasibility, and practical utility warrant prospective, real-world evaluation.</p>