Abstract
The Cohort Size Shrinkage Index (CSSI) is the most widely used intensity measure in studies of the Chinese famine of 1959-1961. We examined requirements to construct a valid CSSI measure: reliable information on birth cohort size; a stable cohort size trend outside the famine years; and limited migrations. We used census data to examine the cohort size trend and concentrated on the time window of 1950-1970 to exclude events with a large impact on cohort size trends other than the famine. We observed a significant difference in cohort size trends comparing pre-famine and post-famine births. Compared to pre-famine based CSSI (p-CSSI), pre- & post-famine based CSSI (pp-CSSI) tends to overestimate famine intensity at higher mean CSSI levels and underestimate intensity at lower mean levels. In Sichuan province, we demonstrated a less pronounced dose-response relation between famine intensity and tuberculosis outcomes using p-CSSI as compared to pp-CSSI. These observations demonstrates that the CSSI is not a robust measure as had been assumed previously. We recommend the use of p-CSSI and encourage researchers to re-examine their results of Chinese famine studies.

This publication has 40 references indexed in Scilit: