Our current prospective multi-site study has led to the development of a robust prognostic model among transfemoral (TF), transtibial (TT), and transmetatarsal (TM) amputees that we call AMPREDICT. AMPREDICT was developed to be used pre-operatively for treatment decision making by correlating outcome success with baseline demographic, physical and psychosocial measures. The ultimate goal of the proposed research is to validate AMPREDICT so that it can be recommended for widespread clinical use for the following purposes: (1) providing surgeons, rehabilitation teams, and patients with information about patient prognosis;(2) assisting surgeons, rehabilitation teams, and patients in planning the most appropriate level of amputation;(3) providing an objective tool for validly comparing treatment and rehabilitation outcomes;(4) assisting in the communication and exchange of information among clinicians and patients;and (5) assisting in VA-wide policy decisions to ensure sound judgment regarding the best level of care for each individual patient. To accomplish this, we plan to assess the performance of the prognostic model AMPREDICT by assessing its accuracy in different populations than the one it has been developed in. This research is critical before AMPREDICT is recommended for widespread clinical use. The primary aim for this study is to assess our current prognostic model's (AMPREDICT) performance for predicting both physical and psychosocial outcomes among TT and TM amputees in the same institution during a different time period (temporal transportability) and in different institutions (geographic transportability) so that it can be recommended for widespread clinical use.
Project Narrative About 10% of major amputations performed in the US are performed in Department of Veterans Affairs Medical Centers. Over 2600 major amputations were performed in DVA hospitals in fiscal year (FY) 2003. One of the primary determinants of functional outcome is the choice of amputation level. Amputation at the transfemoral level compared with the transtibial level significantly reduces the probability of attaining independent mobility and community ambulation. Choosing the optimum amputation level is complicated. The more distal the amputation the greater the potential for an enhanced functional outcome but the healing rate is reduced. A failure of healing may lead to prolonged or recurrent hospitalization with a need for additional surgery with its incumbent risks of additional morbidity and mortality. A recent VA Office of Patient Care Services Report entitled, """"""""Lower Extremity Amputations in VHA;FY 1997-2003"""""""", demonstrates this inconsistency. It Compared 20 Veteran Integrated Service Networks (VISN) with respect to the TF/TT amputee ratio. Seven VISN's performed more TF amputations than TT amputations. The report illustrates there are VISN's that perform nearly twice as many transfemoral amputations as they perform transtibial amputations, and VISN's that perform 1/2 the number of transfemoral amputations as they perform transtibial amputations. It is clear that there is widespread variability in surgical practice. Over the past four years, our group has conducted the first prospective study of factors associated with treatment """"""""success,"""""""" and has developed a prognostic model we call AMPREDICT. This robust prognostic model examines baseline demographic, medical, physical and psychosocial risk factors predicting medical and psychosocial outcomes. Implementation of AMPREDICT will provide a valuable clinical decision-making tool to providers to help them determine the appropriate level of amputation for their patients. However, before a prognostic model can be recommended for widespread clinical use, its performance needs to be established. This is accomplished through temporal and external validation by assessing a model's performance in patients recruited at a later time point (temporal transportability) and in medical centers different from those used for model building (geographic transportability). Since this research takes 4 years to complete, it is paramount that we embark on a second prospective multi-site trial to accomplish these tasks as soon as possible. The Veterans Medical Programs Amendments of 1992 (PL 102-405) identified veterans with amputations as a special disability group and emphasized the importance for VHA to provide the highest quality of care. This year the VA Office of Patient Care Services approved and funded the development of a national amputation system of care with the goal of enhancing the quality and consistency of care. As noted above, there is considerable variability in decision making regarding amputation level. This is in part related to the combined uncertainty of an individual's constellation of medical, psychological, and social factors on their functional outcome. The proposed work in conjunction with the previously funded work will specifically result in a model of outcome prediction that can be used clinically. The tool will allow the preoperative prediction of outcome based upon an assessment of key variables. This will significantly reduce variability in health care practices, allow more effective prediction of functional outcome as a rehabilitation planning tool, and allow for more effective patient education and consent. These are all key issues that will assist the VA in its mission to enhance the clinical care of patients with amputation.
