Cancer incidence and prevalence estimates are necessary to inform health policy, to predict public health impact and to identify etiological factors. Registers have been used to estimate the number of cancer cases. To be reliable and useful, cancer registry data should be complete. Capture-recapture is a method for estimating the number of cases missed, originally developed in ecology to estimate the size of animal populations. Capture recapture methods in cancer epidemiology involve modelling the overlap between lists of individuals using log-linear models. These models rely on assumption of independence of sources and equal catchability between individuals, unlikely to be satisfied in cancer population as severe cases are more likely to be captured than simple cases.
To estimate cancer population and completeness of cancer registry, we applied Mth models that rely on parameters that influence capture as time of capture (t) and individual heterogeneity (h) and compared results to the ones obtained with classical log-linear models and sample coverage approach. For three sources collecting breast and colorectal cancer cases (Histopathological cancer registry, hospital Multidisciplinary Team Meetings, and cancer screening programmes), individual heterogeneity is suspected in cancer population due to age, gender, screening history or presence of metastases. Individual heterogeneity is hardly analysed as classical log-linear models usually pool it with between-"list" dependence. We applied Bayesian Model Averaging which can be applied with small sample without asymptotic assumption, contrary to the maximum likelihood estimate procedure.
Cancer population estimates were based on the results of the Mh model, with an averaged estimate of 803 cases of breast cancer and 521 cases of colorectal cancer. In the log-linear model, estimates were of 791 cases of breast cancer and 527 cases of colorectal cancer according to the retained models (729 and 481 histological cases, respectively).
We applied Mth models and Bayesian population estimation to small sample of a cancer population. Advantage of Mth models applied to cancer datasets, is the ability to explore individual factors associated with capture heterogeneity, as equal capture probability assumption is unlikely. Mth models and Bayesian population estimation are well-suited for capture-recapture in a heterogeneous cancer population.
BMC Medical Research Methodology 04/2015; 15(1):39. DOI:10.1186/s12874-015-0029-7 · 2.17 Impact Factor
Revue d Épidémiologie et de Santé Publique 09/2012; 60:S70. DOI:10.1016/j.respe.2012.06.095 · 0.66 Impact Factor
Cancer population studies require reliable and complete baseline data, which should theoretically be available by collecting histopathology records. The completeness of such a collection was evaluated using capture-recapture analysis based on three data sources concerning breast and colorectal cancers over an identical period and within the same geographical area.
The total number of breast and colon cancer cases was estimated using capture-recapture analysis based on the number of cases which were common or not between sources recording screened, diagnosed and treated cancers in the French Alpes Maritimes district.
The estimated total number of new cases of breast cancer diagnosed among Alpes Maritimes residents women aged 50-75 was 791 (95% CI: 784-797) in 2008. Of these 791 cases, 729 were identified through histopathology records, thus amounting to 92.2% completeness (95% CI: 91.5-93.0%). The total estimated number of new cases of colorectal cancer diagnosed among Alpes Maritimes residents aged 50-75 was 527 (95% CI: 517-536). Of these 527 cases, 481 were identified through histopathology records, thus amounting to 91.3% completeness (95% CI: 89.7-93.0%).
The estimated completeness of cancer records collected from histopathology laboratories was higher than 90% for new cases of breast and colorectal cancer within the age range concerned by the screening programme. A verified and validated histopathology data collection may be useful for cancer population studies.
08/2011; 35(6):e62-8. DOI:10.1016/j.canep.2011.05.017
Revue d Épidémiologie et de Santé Publique 09/2010; 58. DOI:10.1016/j.respe.2010.06.030 · 0.66 Impact Factor