Statistical reliability analysis for a most dangerous occupation: Roman emperor
Level 4 - case-series / case-control
Historical cohort survival analysis (by design analogy, not clinical CEBM)
OpenAlex W2994920274 · doi:10.1057/s41599-019-0366-y
What was done
The author applied survival data analysis and reliability engineering statistical tools (nonparametric and parametric techniques, including mixture Weibull distributions) to evaluate the time-to-violent-death of 69 rulers of the unified Roman Empire from Augustus (d. 14 CE) to Theodosius (d. 395 CE).
What was found
Among the 69 emperors, 62% suffered violent death. The hazard rate of violent death displayed a bathtub-like curve: risk was significantly high during the first year of rule, decreased in subsequent years, and increased again after 12 years of rule. This stochastic process was well fitted by a (mixture) Weibull distribution.
Why it matters
This study shows that regicide and violent death among Roman rulers followed a structured temporal hazard pattern analogous to reliability models in engineering rather than purely random stochastic events.
Limits
The dataset is small (n = 69) and restricted to the unified Roman Empire up to 395 CE. Cause-of-death determinations rely on historical records with inherent uncertainties, and the abstract does not report adjustment for historical, political, or military covariates.
Cited by
- context During the Crisis of the Third Century, 26 Roman emperors were murdered over a period of 50 years.
- context During the Crisis of the Third Century, 26 emperors were murdered over a 50-year period.