Abstract
Patient navigation is an innovative type of care management designed to reduce cancer disparities. As many sites adopt this model of care, a frequent query is how to assess appropriate caseload for these new positions. While data on the effectiveness of patient navigation interventions appears promising, there is a lack of methods to assess appropriate caseload for optimal effect of navigator intervention. Boston University and its affiliated Community Health Centers (CHCs) is one of nine sites participating in the NCI CRCHD National Cancer Institute Patient Navigation Research Program (PNRP), which aims to evaluate whether navigation improves cancer outcomes.
The purpose of this study is to develop and describe the caseload across six CHCs participating in the Boston PNRP. Navigators target female patients with abnormal breast or cervical cancer screening. Review of the literature identified several concepts to consider in assessing caseload. We examined the following four factors at each navigation site: (1) absolute number of cases navigated, (2) total patient days in navigation, (3) total time spent navigating, and (4) total documented patient-navigator encounters.
Eligible participants were women over 18 with an abnormal mammogram or Pap test at the participating CHCs from 7/1/2007 through 6/31/2008. Patients were navigated until their screening abnormality was resolved with a definitive diagnosis (cancer or noncancer) or until the end of cancer treatment.
Across the six CHCs, the median number of patients navigated was 145 (range: 19-216). The median number of encounters per patient at each of the CHCs was 5 (range: 3-11). The median time spent navigating per patient was 135 minutes (range: 77-240). The median number of days in navigation per patient was 56 (range: 73-142). The number of active patients a navigator was tracking, on average, ranged from 7 to 53 patients per day (median: 38).
We conclude that measuring caseload is more complex than counting heads. The four descriptions of caseload provide different representations of navigator workload that vary among the six CHCs. Patient characteristics (age, race/ethnicity, language, insurance) were not associated with these measures of caseload. Due to the different constructs assessed in each measure, a single caseload score may not adequately reflect navigator workload. Next research steps will involve assessing the predictive ability of these constructs on outcomes to help determine an appropriate navigator caseload.
Citation Information: Cancer Epidemiol Biomarkers Prev 2010;19(10 Suppl):A26.
RSS Feeds