Abstract
Context: The National Research Agenda for Public Health Services and Systems Research states the need for research to determine the cost of delivering public health services in order to assist the public health system in communicating financial needs to decision makers, partners, and health reform leaders.
Objective: The objective of this analysis is to compare 2 cost estimation methodologies, public health manager estimates of employee time spent and activity logs completed by public health workers, to understand to what degree manager surveys could be used in lieu of more time-consuming and burdensome activity logs.
Design: Employees recorded their time spent on communicable disease surveillance for a 2-week period using an activity log. Managers then estimated time spent by each employee on a manager survey. Robust and ordinary least squares regression was used to measure the agreement between the time estimated by the manager and the time recorded by the employee.
Main Outcome Measures: The 2 outcomes for this study included time recorded by the employee on the activity log and time estimated by the manager on the manager survey.
Setting: This study was conducted in local health departments in Colorado.
Participants: Forty-one Colorado local health departments (82%) agreed to participate.
Results: Seven of the 8 models showed that managers underestimate their employees' time, especially for activities on which an employee spent little time. Manager surveys can best estimate time for time-intensive activities, such as total time spent on a core service or broad public health activity, and yet are less precise when estimating discrete activities.
Conclusions: When Public Health Services and Systems Research researchers and health departments are conducting studies to determine the cost of public health services, there are many situations in which managers can closely approximate the time required and produce a relatively precise approximation of cost without as much time investment by practitioners.