• Conference Proceeding

Mortality following diagnosis of vertebral compression fracture (VCF) in the Medicare population

Citation

Lau, E., Kurtz, S., Schmier, J., Halpern, M., & Edidin, A. (2005). Mortality following diagnosis of vertebral compression fracture (VCF) in the Medicare population. In [5], p. S102. .

Abstract

BACKGROUND CONTEXT: Vertebral compression fracture (VCF) is
known to increase mortality in female patients, but the generality of
this finding at a national level, as a function of age, gender, ethnicity, and
region in the United States remains unclear.
PURPOSE: To analyze mortality associated with vertebral compression
fractures for Medicare beneficiaries from 1997 to 2001.
STUDY DESIGN/SETTING: Retrospective data analysis of the Medicare
5% beneficiary encrypted files (BEF).
PATIENT SAMPLE: All individuals with a new VCF diagnosis in the
Medicare Part B, Outpatient Facility, or Inpatient Facility from 1997 through
2001 BEF data were identified, corresponding to 51,600 patients in this
study period. Controls were identified and matched for age group, gender,
and state of residence with a 10:1 ratio to VCF patients.
OUTCOME MEASURES: Start and end date of Medicare entitlement,
as well as date of death, if applicable, was obtained from the Medicare
denominator file. Mortality rate was calculated from date of initial VCF
diagnosis, for both cases and their controls, until end of Medicare enrollment
or until death, and at various intermediate intervals (eg, 6 month, 1 year).
METHODS: Ratios of mortality rates between the VCF and the control
groups were compared using descriptive statistics and regression methods.
Sub-population analyses were performed by gender, by race/ethnicity, by
age group, and by regions. In addition, the Charlson comorbidity index
was calculated and used as another dimension in the stratified analyses.
RESULTS: Patients with VCF had an overall mortality rate approximately
twice that of matched controls (p.001). Survival rates following VCF
diagnosis estimated by Kaplan-Meier method at 3, 6, 9, 12, 24, and 36
months were 92%, 87%, 84%, 81%, 71%, and 62% respectively. For
controls, the corresponding survival fractions were higher at 98%, 96%,
94%, 92%, 85%, and 78%. Male patients were at a greater mortality risk
than female patients following VCF diagnosis. Difference in mortality
between cases and controls was greatest following VCF when VCF was
diagnosed in younger patients (eg, 65–69), but this difference declined as the
age of the patient increased. The mortality ratios among different races and
ethnic groups were similar, as were the mortality ratios for patients from
different geographic regions. The higher mortality risk among VCF patients
may be due to poorer overall health, and not necessarily directly related to
VCF. However, the mortality ratio was reduced only slightly and remained
statistically significant after controlling for the presence of comorbidities.
CONCLUSIONS: This study establishes the mortality risk associated with
VCF for patients of all ages, genders, and ethnicities in the Medicare
population. The difference in mortality between VCF cases and controls
is higher than previously reported, even controlling for comorbidities. In
addition, this study suggests VCF is especially detrimental to men and to
younger patients. Maintenance of bone health and early intervention when
VCF is diagnosed may be important steps to reduce the burden of this
disease in the elderly population.
DISCLOSURES: No disclosures.
CONFLICT OF INTEREST: Authors (EL, SK, JS, MH) Grant Research
Support: Kyphon, Inc. Project MT00054.0MD; Author (AE) Other: Employee
of Kyphon, Inc.