Table 5.
Baseline characteristics and measles seronegativity before and after propensity score matching (PSM).
| Variable | Unmatched cohort | Model 1: total effect model | Model 2: controlled direct effect model |
|---|---|---|---|
| Demographic balance | |||
| Age (years, Mean ± SD) | 9.82 ± 6.14 vs. 8.45 ± 5.92(SMD=0.228) | 8.45 ± 5.92 vs. 8.51 ± 5.88(SMD=0.010) | 8.45 ± 5.92 vs. 8.48 ± 5.90(SMD=0.005) |
| Sex (Male, %) | 52.41% vs. 54.43% (SMD=0.041) | 54.43% vs. 53.80% (SMD=0.013) | 54.43% vs. 54.43% (SMD=0.000) |
| Region (Xinjiang, %) | 55.17% vs. 68.35% (SMD=0.273) | 68.35% vs. 68.35% (SMD=0.000) | 68.35% vs. 68.35% (SMD=0.000) |
| Ethnicity (Han, %) | 58.18% vs. 51.90% (SMD=0.126) | 51.90% vs. 52.53% (SMD=0.013) | 51.90% vs. 51.90% (SMD=0.000) |
| MCV Doses (2 Doses, %) | 74.49% vs. 73.42% (SMD=0.024) | 73.42% vs. 75.32% (SMD=0.043) | 73.42% vs. 73.42% (SMD=0.000) |
| Primary outcome | |||
| Seronegativity Rate (%) | 15.74% (303/1925) vs. 11.39% (18/158) | 15.82% (25/158) vs. 11.39% (18/158) | 15.19% (24/158) vs. 11.39% (18/158) |
| Effect estimates | |||
| Odds Ratio (OR/AOR) | Crude OR=0.69 (95% CI:0.41–1.14) | Total OR=0.68 (95% CI:0.36–1.31) | Direct AOR=0.72 (95% CI:0.37–1.38) |
| P-value | P=0.147 | P=0.252 | P=0.316 |
SMD < 0.10 indicates adequate balance between matched cohorts.
Model 1 assesses the Total Causal Effect by matching strictly on baseline confounders (age, sex, region, ethnicity) without adjusting for downstream vaccination history.
Model 2 assesses the Controlled Direct Effect by matching on baseline confounders plus MCV dose history (0/1/2/≥3 doses) to evaluate whether antibody levels differ given identical vaccination regimens.