In August of 1998, Rehabilitation Clinical Services sponsored a conference entitled "Outcomes to Measure Quality and Value in Disability Management" designed to kick off an ongoing review of current databases and outcomes initiatives in the very rich data environment of the Veterans Health Administration (VHA). VHA has a high capability to do national rollups, but they are segmented in various places around the country. Also under assessment is the feasibility of developing a global disability index, which could be used for a variety of patient diagnostic groups and functional groups in order to better learn how to case match the total population for function.
The participants included specialty professionals from medical rehabilitation, spinal cord injury, audiology and speech pathology, long-term care, blind rehabilitation, vocational rehabilitation services, and behavioral health rehabilitation. Representatives from other federal agencies, private sector researchers, VSOs, outside patient advocacy groups, and the ICIDH, an international classification system that looks at various outcomes and divides them into various groups also were included. Such a widely diverse group of attendees was important to help define needs and commonalities of all and begin to combine some of these measures and develop a global disability index.
Attendees were given three tiers of questions, the first of which was from a national perspective:
2. Can these measures or indicators be key to critical customer audiences such as veterans, VSOs, and Congress, but also meet the needs of medical professional audiences?
3. Can outcome measurement help the VA to become more consumer oriented and thus more competitive with the private sector?
The second tier of questions related to the stakeholder perspective—the clinician, the hospital, the network, and the patient—everyone with a stake in outcome measurement.
1. What would a stakeholder want to know about a program's performance in order to assess its quality and choose among programs?
2. How do we benchmark?
3. How do we decide on best practices and what should we look at for best practice.
4. For what current systems do we have to define quality?
5. Can a system be integrated to improve the accuracy and the predictability of outcome measures?
6. Does any system of evaluation allow for cross comparison over a variety of patient populations?
7. Can we come up with a global disability index? If none exists, can a system of measurement be agreed upon that will improve the ability to predict functional outcome?
Questions in the third tier were focused on success criteria.
1. What are the key components of an outcome measure that would tell us whether a program is good or not?
2. How can it be measured and when should it be measured.
3. At what point is it going to be most predictable and have its best accuracy?
4. When will an outcome measure have its least influence from compounding variables and what determines it?
5. What are the key components of that outcome measure and what should be done in response?
6. Do we need to look at quality as compared with cost effectiveness and as compared with access?
7. Should we look at all those values in the VA and come up with something that meets all the needs of the veteran population?
Appropriate outcome measures related to any diagnosis should measure the effect of care on:
1.) stabilizing or reducing impairment;
2.) preventing complications;
3.) improving functional/vocational/community status;
4.) reducing social role limitations;
5.) improving quality of life; and
6.) improving health care satisfaction.
When outcomes are evaluated, the ICIDH international classification system is essential. ICIDH divides its outcome measures into three groups: organ level, person level, and social level. In the standardized ICIDH criteria, this is devised as impairment, and impairment measures are traditional medical model measures that are simple, very objective, reliable, and single-system focused. Impairment measures, however, really don't tell you whether or not the patient is going to be able to function in the community.
The next ICIDH classification level is disability or, under ICIDH-2, activity, which is defined as the combined effect of the various impairments on the patients' ability to care for themselves in their daily activities. This is a good measure to look at, but all the compounding variables that also affect their ability make it a more difficult measure than impairment.
The third level is handicap or participation measures, defined as the ability of patients to function in their community. The psychosocial influences involved with participation measures make it very difficult to tease out some of the compound variables. Other domains to be considered include environment, of significance within the veteran population because of the high incidence of homeless, indigent, and single veterans. Satisfaction and health related quality of life are also considered.
There are several administrative databases within VA from which to glean information. For example, the Patient Treatment Files (PTF) is a standard in-patient treatment file that shows who was admitted, for how long, what their admitting diagnosis was, what their discharge diagnosis was, and whether they got readmitted. The Standard Administrative File, predominantly in-patient, is now moving to outpatient.
The National Surgical Quality Improvement Program (NSQIP) is the primary quality assurance monitor for surgery. By looking at morbidity and mortality risk factors, it tries to benchmark for surgery.
The National Pharmacy Database is one of the most accurate databases. It tells us what medications patients are receiving. The NPD has helped identify a large number of patient groups, particularly diabetics, which can't be identified anywhere else.
The National Prosthetics Product Database is just beginning, but at this stage it is not very accurate and needs a lot of validation.
The Decision Support System is a billing system and looks closely at the cost of care for anyone—both in-patient and outpatient.
The Functional Status Outcome Database measures disability and activity. It is predominantly an ADL scale and is very valid. It has a national rollup so sites can be compared within the VA. Comparisons can be made with the private sector as well.
The Minimal Data Set (MDS) is another functional measure, used for long-term care, but also uses some components of behavior and medical illness. It is used in the private sector now to provide an allocation model for long-term care by HICFA. It is predominantly a snapshot database, not one that can track patients over a long period of time.
The Global Assessment of Function is a behavioral database and has a good reputation.
Finally, the VA has recently implemented the SF-36V database. It has five major domains, including pain and role responsibility, as well as some functional components. Every veteran in the entire system will be receiving this. The intent will be to use that data to identify those patients that might need more intense focus and more intervention.
Future directions in gathering and utilizing outcomes data within VA include extending measures to assess participation and environmental indexes, concentration on the development of patient stratification models, developing divergent set of outcome measures to accommodate patient cohorts with widely divergent functional levels and evaluating the capacity of currently mandated outcome measures to identify groups requiring more advanced evaluation.
Go to TOP.
|
|
|