A New Approach to Estimate Age-Specific Mortality From Sibling Survival History Data in the Presence of Recall Errors
(2022) Annual Meeting of the Population Association of America — Location: Atlanta (6.April.2022)
Files
No attached file found for this publication.
Details
- Authors
- Abstract
- Sibling survival history (SSH) data can be used to estimate adult mortality rates. However, SSH data may be subject to substantial reporting errors in the form of omissions of siblings, or misreported years of birth and age at death. We present a new Bayesian survival modeling approach (B-Surv) to estimate agecohort-specific survival probabilities from SSH data while accounting for bias and uncertainty introduced by the age reporting errors. In the model, true survival by cohort and age is captured with a twodimensional splines function. SSH reported date of birth (DoB) and age at death (AaD) are the sum of the latent true outcome and a reporting error. A bivariate distribution of reporting errors in (DoB, AaD) is obtained from an analysis of such errors and its predictors, obtained from a comparison of SSH data with gold-standard data obtained in recent validation study conducted in a health demographic surveillance system in Senegal. We illustrate the approach using simulations and apply it to estimate adult survival in Senegal from DHS data.
- Affiliations
Citations
Zhengfan Wang, Emily Peterson, Stephane Helleringer, Masquelier, B., & Leontine Alkema. (2022). A New Approach to Estimate Age-Specific Mortality From Sibling Survival History Data in the Presence of Recall Errors. Annual Meeting of the Population Association of America, Atlanta. https://hdl.handle.net/2078.5/212439