Selection of neighborhood controls: logistics and fieldwork

This paper describes the logistics and fieldwork involved in assembling 123 neighborhood controls for a case-control study investigating the relationship between cholecystectomy and right-side colon cancer. All neighborhood controls were matched pairwise to cases on sex, age at diagnosis (+/- years), and race. Eighty-six percent of the neighborhood controls were also matched on street of residence and the remaining 14% on census tract. Study subjects were contacted by mail and then interviewed by telephone with regard to their gastrointestinal surgical history. The mean number of necessary households contacted to obtain each control was nine. When data were stratified by age, the mean number of necessary households contacted increased with increasing age of the cases. Only 3.3% (35/1065) of the residents from these households refused complete participation in the telephone interview. The distribution of fieldwork activities by hours worked is also reported. Similar findings were observed for males and females. Neighborhood controls are a commonly used comparison series. Data from this report may be used to modify the general impression that assembling neighborhood controls can be a difficult task. We urge other investigators to report fieldwork activities used in identifying their control series. These data will be valuable in planning and budgeting case-control studies.