Brian Saylor and Sharman Haley
August 31, 2007
The Alaska Native Tribal Health Consortium (ANTHC) is implementing new paraprofessional and professional health aide programs in rural Alaska. The Community Health Aide (CHA) Program instituted decades ago has been a resounding success in improving access to primary health care in rural Alaska. Following this model, the ANTHC is now implementing three new paraprofessional and professional health aide programs to improve access to primary health services, lower health costs, and improve health outcomes in rural communities in three fields of care with large, unmet needs: dental health, behavioral health, and elder care. Two of these developing programs (dental and behavioral health) closely mirror the CHA model for service delivery, training and supervision; the third program (Personal Care Assistance Program) is an enhancement of existing elder service delivery infrastructure.
The common objective of each of these programs is to increased access to needed care in rural Alaskan communities by training and supervising paraprofessionals to deliver services at the local level. Using highly trained health care providers to deliver sophisticated health and medical services in rural communities is both costly and ineffective. Providers brought in to rural communities seldom remain in the community, thus disrupting ongoing treatment regimens and compromising continuity of health care. Skilled highly trained health care professionals command high salaries, and therefore have a high cost per unit of service provided in small populations. Additionally, providers brought in from outside the community are unfamiliar with local customs, and may not understand the social and cultural meaning of health and health care. Recruiting and training local providers from rural Alaskan communities helps to address all these problems; local providers are less costly, understand the local customs and tend to remain in the community.
All three programs within this scope of work share these common characteristics. With a grant from the Ford Foundation, ANTHC contracted with the Institute of Social and Economic Research and the Institute for Circumpolar Health Studies at the University of Alaska Anchorage to assess the clinical and economic impacts of these three programs designed to train and supervise local residents of rural Alaskan communities to provide needed health services. The desired work product was a template or generic model of impacts that could guide data collection and analysis for these and future paraprofessional and professional health programs. The model was to be initially applied to the Dental Health Aide (DHA) Program and then to the Behavioral Health Aide (BHA) and the Personal Care Assistance (PCA) Programs. The work began in February 2005.
The original scope of work contemplated two background papers, designs for clinical and economic models, and a pilot study on the Dental Health Aide (DHA) program. The aim of the evaluation team was to develop an evaluation protocol and test its utility in evaluating the services provided by the behavioral health aides and personal care attendants. The two background papers - one an analysis of the scope of practice of the various levels of DHAs and the other a literature review of similar programs in other countries - followed the task descriptions quite closely. The clinical impact model and pilot study, however, were much more complex than expected.
Evaluating clinical outcomes of any activity requires a data set with extensive data on the characteristics of the patient, the nature and extent of services provided, and the clinical outcomes of those services on the health status of the patient. Because the DHA program was in the early stages of implementation, there was little treatment process or clinical outcome data upon which to base an evaluation. Furthermore, the clinical benefits of preventive dental care affect the patients’ oral health and function over a lifetime. These long time lags make an observation-based evaluation of outcomes impractical.
Given the limitations in the clinical data, the evaluation team decided that the best approach would be to build a 20-year computer simulation model comparing the clinical outcomes of two alternative models of service delivery - the old, pre-DHA service delivery model, and the new DHA model that was being implemented. The innovation in the DHA program is not in the types of services delivered, but how and where and by whom the services are delivered. The DHA program provides increased access to basic preventive and restorative dental services by village residents. This change in access to oral health services drives the change in clinical outcomes.
The clinical efficacy of the preventive and restorative oral health services themselves is well established in the dental research literature. Our model of different patterns of service delivery was built using these research-based treatment-effects relationships, coupled with the best clinical advice of dentists practicing in rural Alaska.
The evaluation team selected a discrete event simulation approach to modeling oral health services. This was an ambitious and innovative undertaking. We found no examples of comparable oral health modeling projects in the literature. Dr. Dan Kiley, a practicing dentist, was retained to work on the clinical impact model as part of his Masters of Public Health degree. Dr. Kiley collaborated with a simulation systems programmer, Ben Saylor, to build a village-level clinical effects model. We believe that the use of the simulation modeling technique yielded a model that is far more detailed, complex and powerful tool than could ever have been anticipated with the modest project budget. It is also appears to be more advanced than anything that has been done before in field of oral health services modeling, and deserves national exposure and recognition.
Unfortunately, the evaluation team was not able to finish the model. The village-level clinical effects model is complete and usable, but legal and political concerns interrupted our work before we could tackle the regional-level clinical and fiscal effects models. The Alaska Dental Society filed a lawsuit in state court challenging the legality of the scope of practice of Dental Health Aide Therapists—the highest of six levels of DHAs - and other elements of the DHA program. Concerns by ANTHC attorneys about the release of program information as well as heightened sensitivity among program staff caused the client to ask us to stop all work on the DHA component of the project. This stopped our work in developing a regional model that would interface with the village-level model.
As part of the process of developing the model, the team created a flowchart to represent how patients flow through the oral health care system. The model itself operates on a flowchart based on this one.
The village level dental health aide model is made available for download here. It is a Java application developed using the simulation software AnyLogic 5 from XJ Technologies. It requires Microsoft Windows, Excel, and a Java runtime environment to run. More detailed information is available in the User's Guide below.