Artificial Intelligence replication of District Hospital Rabies Post-Exposure Prophylaxis Regimen Completion Under Unmeasured Patient Travel-Distance Attrition: A Healthcare Causal Robustness Evaluation

District hospital rabies post-exposure prophylaxis (PEP) programmes in low- and middle-income countries depend on patients completing a multi-visit vaccine series, yet travel-distance attrition is rarely measured and never recorded in routine registers. Six hundred data sets of N = 500 were drawn from a process with intercept 12.0, exposure coefficient 0.42, group coefficient -3.1, severity coefficient 1.35 and error standard deviation 4.2, using seed 906793928. The estimator recovered the exposure coefficient with estimated bias 0.00017 and root mean squared error 0.01572, and the intercept with estimated bias 0.001 and root mean squared error 0.87611; empirical coverage of the nominal 95% confidence intervals was 1.0 for both parameters. The results indicate that under the stated process the estimator is unbiased and its interval coverage is conservative, so the unmeasured attrition mechanism does not by itself destabilise coefficient recovery. The contribution is a bounded robustness statement: the estimator is adequate for the declared process, and the paper specifies the conditions under which that adequacy would not survive.