Organization of HCSUS Research
Understanding the enormous cost of a national probability sample, the HCSUS Consortium attempted to cover a wide variety of research topics. A team of researchers was assigned to each of fourteen cognitive areas. These Research Teams were charged with:
- preparing specific aims for proposals
- outlining specific questions that they want answered by the data
- nominating measures that would collect the appropriate data
- rewriting measures to make them specific to HIV as needed
- drawing up a slate of proposed publications using these data
- ensuring that the purview of each team was respected in drawing up publication proposals
- overseeing the writing of the publications
- assisting other Research Teams in working with data in their respective cognitive areas
The following list provides information on the fourteen HCSUS Research Teams, each of which is a link to a larger discussion of its activities. These teams continue to perform their functions as needed for active projects related to the original HCSUS.
The goals of this team are to examine variations in access to care, unmet needs for services, and transitions in usual source of care across geographic location, type of health care system, and patient characteristics; and to examine the relationships between access and utilization, service mix, cost, and survival. The national Community Advisory Board assisted in the development of access measures specific to HIV.
Antiretrovirals and especially the combination therapies made great inroads into treatment regimens between the time the original proposal was written and the time of the first interview wave. Consequently, the Consortium sought and received additional funding to examine more closely access and barriers to highly active therapies, their short- and long-term effects on cost, and their impact on disease progression.
Data being used by this team include CD4 cell count, staging, symptoms, treatments, history, prevalence, and complications. Effects from a variety of covariates are being assessed, including biologic, clinical, health services, and demographic factors, and differences in treatment strategies, care setting, and insurance status. Findings from this team are used to help explain variations in quality of care, cost, and utilization.
This team examines the utilization rates of a range of medical and nonmedical services, the costs of such services, and the variation of utilization and costs geographically, across health care systems, and across patient clinical and demographic characteristics. Of particular interest are the effects of insurance status, assistance programs, and use of hospital emergency departments. Estimates in lifetime cost and changes in costs over time have been calculated.
Data collected by this team focuses on patient knowledge of causes, symptoms, and treatment of HIV disease and the sources of such knowledge, and how these influence decisions on when to seek care, use of traditional and alternative treatments, and risk-reduction practices. Connections of knowledge to satisfaction with care, health-related quality of life, social support, and end-of-life decisions are being analyzed.
Interviews included the short form of the Composite International Diagnostic Inventory (SF-CIDI) to assess depression, generalized anxiety disorder, panic attacks, and alcohol and drug dependence as defined by the DSM-IV and ICD-10 standards. From this, need for and actual use of mental health services (including pharmacotherapy) are being determined, as well as influences on utilization and cost of medical services.
Because of the landmark design proposed by the HCSUS, a team of statisticians, programmers, and sociologists was assembled to oversee the development of appropriate methodology for the multi-stage assembly of the national probability sample, as well as oversight of data collection, calibration, and weighting. The team continues to be a resource for researchers on appropriate use of data and analysis methods.
Because oral symptoms can be among the first to appear in HIV infection and because of the existence of anecdotal evidence suggesting denial of access to dental services to HIV-positive individuals, additional funding was sought and received to study the oral health of HIV patients. To avoid exacerbating possible problems with access to care, information on use and cost of dental services, prevalence of oral symptoms, and quality of care was collected only from patient self-reports.
The principal domains of this team are health-related quality of life, social support, and coping. In addition, the Patient team works with the Provider team to look at the influence of site and provider characteristics on patient satisfaction with care. In addition to gathering these data, variations in the questionnaires give the opportunity to examine how different wording, order, and scales affect response.
HCSUS originally proposed to directly collect data on children with perinatally acquired HIV infection, but this was not funded. However, the study was able to collect information on the children of the cohort, both those infected with HIV and those not. These data are used to assess the impact of HIV morbidity and mortality of parents on their children, including a national estimate of the number of AIDS orphans in the United States.
Because HIV care is provided in a variety of settings, separate studies of sites of care and individual providers were mounted to determine if these characteristics influenced quality of care and patient satisfaction. Data includes information on the type, size, affiliation, and service menu at sites of care as well as the age, education, training, experience, compensation plan, knowledge, and attitudes of individuals providing care.
While rates of HIV infection and treatment continue to be concentrated in urban areas, there are specific programs (such as Ryan White) targeted at rural populations with HIV. Differences in care between rural and urban areas are assessed, as well as migration patterns of patients as a result of their desires to receive higher quality of care and increased social support. Rural residents were oversampled to produce sufficient power for these analyses.
The herald cohort assembled for HCSUS presented many opportunities for data collection. One of the most exciting is the chance to collect blood samples, do a variety of analyses, and then place them in a frozen specimen bank. Tests initially performed include CD4 cell count, viral load, genotype, and phenotype. Analyses are being performed to shed light on clinical staging and viral strains resistant to highly active therapies.
Rate of HIV infection is growing among women faster than most other demographic groups, but most previous studies have focused on men. Initial symptoms may be very different for women than men, and might not be recognized as signs of HIV infection. The number of infected women is still small in the context of the total population, so to provide more power to calculations and estimates, women were oversampled at a rate twice that of men.