Predicting prosthetic prescription after major lower-limb amputation

We describe prosthetic limb prescription in the first year following lower-limb amputation and examine the relationship between amputation level, geographic region, and prosthetic prescription. We analyzed 2005 to 2010 Department of Veterans Affairs (VA) Inpatient and Medical Encounters SAS data sets, Vital Status death data, and National Prosthetic Patient Database data for 9,994 Veterans who underwent lower-limb amputation at a VA hospital. Descriptive statistics and bivariates were examined. Cox proportional hazard models identified factors associated with prosthetic prescription. Analyses showed that amputation level was associated with prosthetic prescription. The hazard ratios (HRs) were 1.41 for ankle amputation and 0.46 for transfemoral amputation compared with transtibial amputation. HRs for geographic region were Northeast = 1.49, Upper Midwest = 1.26, and West = 1.39 compared with the South (p < 0.001). African American race, longer length of hospital stay, older age, congestive heart failure, paralysis, other neurological disease, renal failure, and admission from a nursing facility were negatively associated with prosthetic prescription. Being married was positively associated. After adjusting for patient characteristics, people with ankle amputation were most likely to be prescribed a prosthesis and people with transfemoral amputation were least likely. Geographic variation in prosthetic prescription exists in the VA and further research is needed to explain why.