|
BACKGROUND/RATIONALE:
Lessons learned from the Foot SDR study included: facilities with higher levels of coordination were associated with fewer amputation rates; there was widespread variation in implantation of coordinated strategies; and patients at high-risk for lower extremity complications related inability to care for their feet and wanted more frequent care. HQ has recently created an ICD-9 based high risk foot registry. There are potential limitations with this approach. The registry relies on outpatient coding that could be a significant source of error. This approach is also limited by simply providing a list of patients with high risk conditions not stratified by increasing risk. The list could also present a daunting challenge to larger facilities with large diabetic populations. For example, the VA West Los Angeles with ~12,000 diabetics would then have a list of ~3,600 patients with high risk conditions.
OBJECTIVE(S):
The Veterans Health Administration (VA) has prioritized amputation prevention and produced a number of tools to assist this priority. These include continuing medical educational programs, hospital-level amputation prevention teams, distribution of monofilaments and patient education materials, establishing clinical reminders, amputation registries, and high risk registries. The objective of this study was to measure the reliability of the national foot risk registry.
METHODS:
Random stratified samples based on diabetes-related foot risk were drawn on 50 charts at two hospitals. We measured reliability of the current national registry using chart abstraction as the gold standard. Kappa coefficients were calculated for the documented chart notion and coding for foot visits over one fiscal year. Local reports will include the proportion of patients screened by foot risk and clinic enrollment type.
FINDINGS/RESULTS:
At one center, the reminders are typically filled out by health techs and nurses in primary care (kappa = -0.01 to 0.54). At another center, podiatrists typically complete the reminders (kappa = 0.00 to 0.88). There was variation in the reliability of coding each condition among the two centers: diabetes (kappa =0.00, 0.88); foot ulcer (kappa = 0.70, 0.24); amputation (kappa = 0.77, 0.30); end-stage renal disease (kappa = 0.00, 0.32); peripheral neuropathy (kappa =0.33, 0.20); peripheral arterial disease (kappa = -0.01, 0.30); and foot deformity (kappa =0.72, 0.54).
Implications:
Reliability exhibited a great deal of variation by condition, hospital, and provider type. Based on our small sample of records at two VA hospitals, there may be limited reliability of the national high risk foot registry based on administrative data.
IMPACT:
The findings highlight an improvement opportunity for the coding of foot exams and creation of a registry that is reliant on exams rather than coding. A secondary ongoing study is investigating the implementation of a clinical reminder based registry at one of the hospitals. The current diabetes foot exam clinical reminder has been replaced with the International Foot Risk Classification System using class 3 software. The check box format automatically populates the visit file with a Health Factor. A different Health Factor is created for each classification.
PUBLICATIONS:
None at this time.
DRA:
Chronic Diseases
DRE:
Communication and Decision Making
Keywords:
Diabetes
MeSH Terms:
none
|
