The impact of follow-up type and missed deaths on population-based cancer survival studies for Hispanics and Asians.

摘要: The accurate estimation of cancer survival across racial and ethnic groups is an important aspect surveillance. It relies on completeness follow-up to identify deaths. In surveillance, there are two different methods follow-up. Active defined as any type that involves contact with the patient, his/her next-of-kin or physician ascertain vital status. Passive obtaining data through linkages without patient physician. Cancer registration in United States administered by one programs use types first, National Cancer Institute Surveillance, Epidemiology End Results (SEER), covers 28% US population employs resource intensive activities capture last date for each case. Cases matched national statewide databases such Death Index, Social Security, Medicare Medicaid, hospital discharge, well records physicians, pathology labs, registries. SEER registries also perform voter driver license data, some information obtained from research studies supplement registry data. standards require their report status current within 22 months annual submission a minimum 95% all registered patients, living deceased (1). can be either direct so combination active passive spite this, this paper we refer carried out “active” because available cases. Because availability contact, traditional method analysis “reported documented alive” method, which reported death alive if no available. This censors patients at contact. Researchers who analyses often its high quality. Moreover, was only source level until recently. The second surveillance program Program Registries (NPCR). NPCR established Centers Disease Control Prevention balance population. conduct matching cases in-state lists Index. However, do not systematically collect As result, statistics calculated using what known “presumed method. assumes person found dead, then he/she most recent covered Presumed appear less commonly literature (2–5), partly have recently become Both North American Association Central now recognize suitable approach population-based calculations (6). In States, Hispanics, Asians Pacific Islanders (APIs) fastest growing groups, comprising 22% (7). effects versus corresponding estimating been thoroughly examined among these groups. We hypothesized procedures equally effective four major racial-ethnic groups: whites, blacks, APIs. In study, five-year proportions were compared described methods, presumed alive, is, considering (active conducted SEER) assuming dead 5 years diagnosis (passive registries), respectively. addition, (whites, APIs) stage subgroups (eg, Mexican, Chinese, on) assess potential biases populations.