Sources of Cost Data
Understanding where costs come from (e.g., the intervention, morbidity associated with disease, and side effects from treatment) and the types of costs helps to frame the collection of cost data. Cost data can be obtained in several different ways. Most clinical trials do not directly include cost endpoints nor are they powered to detect differences in costs. For clinical trials that do capture costs, statistical modeling may be needed to detect significant differences (Briggs, Sculpher, & Claxton, 2006). If a cost-effectiveness analysis is being planned alongside a clinical trial, identifying the types of costs is an important first step. In the planning phase of the study, it is often helpful to think of all of the different types of costs that can be incurred (e.g., hospitalizations). When identifying resources, it is important to include those resources that are expensive (e.g., hospitalizations) or that are not expensive but are used by a large number of people (e.g., screenings). For cost-effectiveness, cost-utility, and cost-benefit analyses, if resource utilization is the same in all groups tested, then these categories of costs can be ignored.
A limited number of instruments are available for assessing health care utilization. One such tool is the research utilization and dementia (RUD) questionnaire, which includes a comprehensive set of questions regarding the participants’ use of inpatient care, outpatient medical care, home health, formal caregiving, and social services (Wimo & Winblad, 2003). RUD also captures health care utilization and work loss experienced by the informal caregiver. Another relevant tool is called the Service Use and Resources Form (SURF), which includes detailed questions on health care utilization, social services, caregiving, and medical supplies (Schneider et al., 2001). In addition to existing survey tools, an investigator can implement his or her own survey questions to capture utilization (Mahoney et al., 2003). For example, an investigator may include the following question to capture hospitalizations (health utilization): How many times were you hospitalized in the past month owing to your condition?