BACKGROUND: This paper describes a collaborative Medicare claims data linkage and sharing effort between the Baylor Health Care System (BHCS) and Texas Medical Foundation (TMF, the Texas Quality Improvement Organization) designed to assess the effect of three quality improvement interventions on care delivered to elderly patients with diabetes. The randomized controlled trial is being conducted among a network of primary care physician practices owned by BHCS and focuses on measures of care process and outcome. METHODS: Cohort definition and baseline measurement took place between January 1 and December 31, 2000. BHCS administrative data and TMF-supplied Medicare enrollment data were used to define the January 1, 2001 prevalence cohort of Medicare diabetic beneficiaries meeting study inclusion criteria. A total of 22 practices (with 92 physicians and 2,158 patients) were randomized to one of three interventions, each of which involved performance measurement feedback on three claims-based measures of care process. Physician profiles, generated by TMF using Medicare utilization files, were reported to study physicians via academic detailing sessions with a BHCS physician educator. RESULTS:The January 1 - December 31, 2000 baseline Medicare claims for the January 1, 2001 prevalence cohort were provided toHTPN by TMF in October 2001, representing a ten-month lag in the ability of Quality Improvement Organizations to provide Part B data relative to a specific episode of care time frame. Overall baseline rates for the claims-based process measures were: annual HbA1c testing (86.1%), annual eye examination (60.8%), and annual lipid profile (72.5%). As anticipated, medical-record based rates of annual eye examination were significantly underrepresented. Agreement between claims-based and medical record-based measures was very close for annual HbA1c and annual lipid profile. CONCLUSIONS: The use of Medicare claims data, through collaboration with a QIO, can help health care providers overcome a significant barrier associated with quality improvement initiatives. Limitations associated with the use of Medicare claims can impact implementation of intervention strategies, but do not prevent them from being a practical tool for improving care.
RCT Entities:
BACKGROUND: This paper describes a collaborative Medicare claims data linkage and sharing effort between the Baylor Health Care System (BHCS) and Texas Medical Foundation (TMF, the Texas Quality Improvement Organization) designed to assess the effect of three quality improvement interventions on care delivered to elderly patients with diabetes. The randomized controlled trial is being conducted among a network of primary care physician practices owned by BHCS and focuses on measures of care process and outcome. METHODS: Cohort definition and baseline measurement took place between January 1 and December 31, 2000. BHCS administrative data and TMF-supplied Medicare enrollment data were used to define the January 1, 2001 prevalence cohort of Medicare diabetic beneficiaries meeting study inclusion criteria. A total of 22 practices (with 92 physicians and 2,158 patients) were randomized to one of three interventions, each of which involved performance measurement feedback on three claims-based measures of care process. Physician profiles, generated by TMF using Medicare utilization files, were reported to study physicians via academic detailing sessions with a BHCS physician educator. RESULTS: The January 1 - December 31, 2000 baseline Medicare claims for the January 1, 2001 prevalence cohort were provided to HTPN by TMF in October 2001, representing a ten-month lag in the ability of Quality Improvement Organizations to provide Part B data relative to a specific episode of care time frame. Overall baseline rates for the claims-based process measures were: annual HbA1c testing (86.1%), annual eye examination (60.8%), and annual lipid profile (72.5%). As anticipated, medical-record based rates of annual eye examination were significantly underrepresented. Agreement between claims-based and medical record-based measures was very close for annual HbA1c and annual lipid profile. CONCLUSIONS: The use of Medicare claims data, through collaboration with a QIO, can help health care providers overcome a significant barrier associated with quality improvement initiatives. Limitations associated with the use of Medicare claims can impact implementation of intervention strategies, but do not prevent them from being a practical tool for improving care.
Authors: S F Jencks; T Cuerdon; D R Burwen; B Fleming; P M Houck; A E Kussmaul; D S Nilasena; D L Ordin; D R Arday Journal: JAMA Date: 2000-10-04 Impact factor: 56.272
Authors: R E Aubert; W H Herman; J Waters; W Moore; D Sutton; B L Peterson; C M Bailey; J P Koplan Journal: Ann Intern Med Date: 1998-10-15 Impact factor: 25.391