Journal of Gerontology: MEDICAL SCIENCES
Cite journal as: J Gerontol A Biol Sci Med Sci. 2010 September;65A(9):1012–1020
© The Author 2010. Published by Oxford University Press on behalf of The Gerontological Society of America.
All rights reserved. For permissions, please e-mail: email@example.com.
Advance Access published on June 7, 2010
to be responsible for more than 1.5 million fractures annu-
ally (1). these fractures are concentrated among older
adults, as frailty and declining bone mass associated with
aging lead to increased risk of falls and fractures (2–4). A
recent study by Cheng and colleagues (5) finds that nearly
30% of elderly medicare beneficiaries have osteoporosis
or have experienced an osteoporosis-related fracture.
Skeletal fractures in the elderly population have been as-
sociated with numerous adverse health outcomes includ-
ing disability (6), psychological deterioration (7,8), and
mortality (9–11). the resultant fractures are also expen-
sive to treat. Kilgore and colleagues (12) estimate that the
incremental costs of osteoporosis-related fractures in the
medicare population (in 2007 $) range from $7,788 (95%
confidence interval, $7,550–$8,025) for distal forearm
fracture to $31,310 (95% confidence interval, $31,073–
$31,547) for hip fracture.
pproximAtely 10 million Americans are esti-
mated to have osteoporosis, and the disease is thought
Although several recent studies have examined the use and
effectiveness of various forms of postacute care in the treat-
ment of fragility fractures (13–16), virtually no research has
examined patterns of treatment across the entire continuum
of clinical care. to our knowledge, the sole exception is work
by Wiktorowicz and colleagues (17), which examined 1997
data from a single Canadian region. even there, the focus of
the analysis was the derivation of cost estimates rather than
a description of the pattern of utilization per se. Using na-
tionally representative medicare claims data and a new al-
gorithm for identifying long-term nursing home residence,
this study provides more detailed estimates of the utilization
burden of fragility fractures in the elderly population.
the article has two objectives. First, it identifies the in-
cremental service utilization associated with fractures at
seven anatomical sites commonly linked to osteoporosis.
Following the empirical approach of Kilgore and col-
leagues (12) and lonnroos and colleagues (18), we use
person-specific changes in utilization before and after
Health Services Utilization After Fractures: evidence
David J. Becker,1 Huifeng yun,1 meredith l. Kilgore,1 Jeffrey r. Curtis,2 elizabeth Delzell,3
lisa C. Gary,1 Kenneth G. Saag,2,3 and michael A. morrisey1
1Department of Health Care organization and policy, 2Division of Clinical immunology and rheumatology, and
3Department of epidemiology, University of Alabama at Birmingham.
Address correspondence to David J. Becker, phD, Department of Health Care organization and policy, University of Alabama at Birmingham, 1665
University Blvd., rpHB 330H, Birmingham, Al 35294. email: firstname.lastname@example.org
Background. osteoporosis-related fractures impose a large and growing societal burden, including adverse health
effects and direct medical costs. postfracture utilization of health care services represents an alternative measure of the
resource costs associated with these fractures.
Methods. We use a 5% random sample of medicare claims data to construct annual cohorts (2000–2004) of beneficia-
ries diagnosed with incident fractures at one of seven sites—clinical vertebral, hip pelvis, femur, tibia/fibula, humerus,
and distal radius/ulna. We use person-specific changes in health services utilization (eg, inpatient acute/postacute days,
home health visits, physical, and occupational therapy) before/after fractures and probabilities of entry into (long-term)
nursing home residency to estimate the utilization burden associated with fractures.
Results. relative to the prior 6-month period, rates of acute hospitalization are between 19.5 (distal radius/ulna) and
72.4 (hip) percentage points higher in the 6 months after fractures. Average acute inpatient days are 1.9 (distal radius/
ulna) to 8.7 (hip) higher in the postfracture period. Fractures are associated with large increases in all forms of postacute
care, including postacute hospitalizations (13.1–71.5 percentage points), postacute inpatient days (6.1–31.4), home health
care hours (3.4–8.4), and hours of physical (5.2–23.6) and occupational (4.3–14.0) therapy. Among patients who were
community dwelling at the time of the initial fracture, 0.9%–1.1% (2.4%–4.0%) were living in a nursing home 6 months
(1 year) after the fracture.
Conclusions. Fractures are associated with significant increases in health services utilization relative to prefracture
levels. Additional research is needed to assess the determinants and effectiveness of alternative forms of fracture care.
Key Words: medicare—Fractures—osteoporosis—Utilization.
Received March 15, 2010; Accepted May 7, 2010
Decision Editor: Luigi Ferrucci, MD, PhD
by guest on December 22, 2015
UTILIZATION AFTER FRACTURES
fractures to estimate the incremental service burden for
each fracture type. For consistency with earlier work, we
compare utilization in the 6 months prior and subsequent
to the incident fracture diagnosis date. the second objec-
tive is to apply a newly developed claims-based algorithm
for identifying nursing home residence (19). the algo-
rithm distinguishes between short-term (skilled nursing
facility [SNF]) and long-term nursing care, and through-
out this article, we refer to patients receiving long-term
nursing care as “nursing home residents.” We use the algo-
rithm to examine transitions into nursing home residency
following fractures and differences in incremental utiliza-
tion by nursing home residency status at the time of the
the purpose of this study is to provide detailed esti-
mates of the average health services utilization associated
with fragility fractures. We use person-specific compari-
sons to control for underlying variation in health/utiliza-
tion and identify the incremental impact of fractures on
the use of various health services. Characterizing these
patterns of health services utilization is an initial step to-
ward studying the determinants of these patterns of care,
exploring the trade-offs in treatment options, and identify-
ing opportunities to improve the quality of care for this
broad set of fragility fractures.
this study was approved by our university’s institutional
review Board and the Centers for medicare and medicaid
Services privacy Board. All our analysis is based upon a 5%
random sample of medicare beneficiaries taken from the
Chronic Conditions Warehouse (20). the Chronic Condi-
tions Warehouse provides beneficiary demographic charac-
teristics and comprehensive longitudinal administrative
claims data (inpatient, SNF, outpatient, physician, and home
health) for 1999–2005.
the claims data are used to construct annual cohorts of
patients diagnosed with an incident occurrence of one of
seven skeletal fractures—clinical vertebral, hip, pelvis, fe-
mur, tibia/fibula, humerus, and distal radius/ulna. We distin-
guish incident fractures from prevalent cases by requiring
that patients had remained continuously enrolled in medi-
care during the previous 12 months without any claims con-
taining diagnosis codes for the specific fracture site.
Following earlier work, we employ a highly specific case
definition algorithm to identify each of the non–vertebral
(21–24) and vertebral (25) fractures in our study. our case
definitions require either an inpatient hospitalization with an
appropriate international Classification of Disease, Version 9,
diagnosis code or a hospital outpatient or physician service
claim containing Healthcare Common procedure Coding
System codes relevant to the specific fracture. to focus on
fragility fractures that are more likely to be related to osteo-
porosis, for anatomical sites other than distal radius/ulna, we
restrict our analysis to closed or pathologic fractures (26).
Additionally, we restrict our analysis to elderly (older than
65 years) U.S. residents (50 states + District of Columbia),
who were continuously enrolled in fee-for-service medicare
(parts A and B) in the 12 months prior to their fracture.
Utilization/Health Status Measures
We use the date of service from the case qualifying claim
(admission or from date) to establish a precise index diagno-
sis date for patients in each of our fracture cohorts. We then
use data from the inpatient, outpatient, skilled nursing, physi-
cian, and home health claims files to construct measures of
health services utilization in the 6-month periods before and
after (including the index diagnosis date) the incident frac-
ture. the use of this specific window is motivated by earlier
work, showing that average monthly health expenditures re-
turn to baseline approximately 6 months after the fracture oc-
currence date (12). Because of this 6-month follow-up
window and the 12-month incident exclusion criterion, our
analysis is limited to fractures occurring in 2000–2004.
the claims data are used to construct three broad types of
utilization measures in the 6 months prior and subsequent to
the fracture diagnosis date. First, we calculate the total num-
ber of acute inpatient days, which allows us to identify the
share of fracture patients who were treated in the inpatient
setting and the average length of stay conditional on an acute
hospitalization. Next, we construct measures of postacute
care utilization, including total days in various postacute in-
patient settings—SNFs, inpatient rehabilitation facilities, and
long-term care hospitals—and total hours of home health
care. Finally, we construct an array of more detailed utiliza-
tion measures, including hours of physical and occupational
therapy (pt/ot) by setting, and counts of outpatient physi-
cian visits by specialty. these physician visit counts will not
capture postoperative care provided in the inpatient setting.
Following yun and colleagues (19), we use claims data
to characterize each patient’s (long-term) nursing home
residency status in each month of our study period (1999–
2005). the algorithm initially uses SNF claims and the
place of service and Healthcare Common procedure Cod-
ing System codes from the outpatient and physician claims
to classify each person month into one of three categories:
(a) “not nursing facility resident”—no nursing facility re-
lated claims in the month; (b) “short-term nursing facility
resident”—one or more SNF claims in the month; and (c)
“long-term nursing home resident”—nursing facility
claims, but no SNF claims in the month. Ultimately, indi-
viduals are defined as (long-term) nursing home residents
in all observation months subsequent to the initial month
containing evidence of a long-term nursing facility stay.
by guest on December 22, 2015
BECkER ET AL.
6. Braithwaite rS, Col NF, Wong JB. estimating hip fracture morbidity,
mortality and costs. J Am Geriatr Soc. 2003;51:364–370.
7. Dolan mm, Hawkes WG, Zimmerman Si, et al. Delirium on hospital
admission in aged hip fracture patients: prediction of mortality and
2-year functional outcomes. J Gerontol A Biol Sci Med Sci.
8. lenze eJ, munin mC, Skidmore er, et al. onset of depression in
elderly persons after hip fracture: implications for prevention and
early intervention of late-life depression. J Am Geriatr Soc. 2007;55:
9. Biluc D, Nguyen ND, milch Ve, et al. mortality risk associated with
low-trauma osteoporotic fracture and subsequent fracture in men and
women. JAMA. 2009;301:513–521.
10. Center Jr, Nguyen tV, Schneider p, et al. mortality after all major
types of osteoporotic fracture in men and women: an observational
study. Lancet. 1999;353:878–882.
11. Cooper C. the crippling consequences of fractures and their impact on
quality of life. Am J Med. 1997;103:12S–17S.
12. Kilgore ml, morrisey mA, Becker DJ, et al. Health care expenditures
associated with skeletal fractures among medicare beneficiaries,
1999–2005. J Bone Miner Res. 2009;24:2050–2055.
13. Deutsch A, Granger CV, Fiedler rC, et al. outcomes and reimburse-
ment of inpatient rehabilitation facilities and subacute rehabilitation
programs for medicare beneficiaries with hip fracture. Med Care.
14. Harada ND, Chun A, Chiu V, et al. patterns of rehabilitation utilization
after hip fracture in actue hospitals and skilled nursing facilities. Med
15. Buntin mB. Access to postacute rehabilitation. Arch Phys Med Reha-
16. Buntin mB, Garten AD, paddock S, et al. How much is postacute care
affected by its availability? Health Serv Res. 2005;40:413–434.
17. Wiktorowicz me, Goeree r, papaioannou A, et al. economic implica-
tions of hip fracture: health service use, institutional care and cost in
Canada. Osteoporos Int. 2001;12:271–278.
18. lonnroos e, Kautiainen H, Sund r, et al. Utilization of inpatient care
before and after hip fracture: a population-based study. Osteoporos
19. yun H, Kilgore ml, Curtis Jr, et al. identifying types of nursing facil-
ity stays using medicare claims data: An algorithm and validation.
Health Serv Outcomes Res Methods. 2010. in press.
20. iowa Foundation for medical Care. Chronic Condition Data Ware-
house User Manual. 2008. http://www.ccwdata.org/downloads/CCW
%20User%20manual.pdf. Accessed march 1, 2010.
21. Baron JA, Karagas m, Barrett J, et al. Basic epidemiology of fractures
of the upper and lower limb among Americans over 65 years of age.
22. Barrett JA, Baron JA, Karagas mr, et al. Fracture risk in the U.S.
medicare population. J Clin Epidemiol. 1999;52:243–249.
23. ray WA, Griffin mr, Fought rl, et al. identification of fractures from
computerized medicare files. J Clin Epidemiol. 1992;45:703–714.
24. Sporer Sm, Weinstein JN, Koval KJ. the geographic incidence and
treatment variation of common fractures of elderly patients. J Am
Acad Orthop Surg. 2006;14:246–255.
25. Curtis Jr, mudano AS, Solomon DH, et al. identification and valida-
tion of vertebral compression fractures using administrative claims
data. Med Care. 2009;47:69–72.
26. Warriner A, patkar Nm, Curtis Jr, et al. Defining Osteoporosis Frac-
tures Using a Systematic Literature Review and Expert Panel Process
[Abstract]. American College of rheumatology (ACr) Annual Scien-
tific meeting; october 24-29, 2008; San Francisco, CA.
27. Charlson me, pompei p, Ales Kl, et al. A new method of classifying
prognostic comorbidity in longitudinal studies: development and vali-
dation. J Chronic Dis. 1987:373–383.
28. Fisher eS, Bell J-e, tomek im, et al. Trends and Regional Variation in
Hip, knee, and Shoulder Replacement. Hanover, NH: Dartmouth At-
las institute for Health policy and Clinical practice; 2010.
29. morrisey mA, Sloan FA, Valvona J. medicare prospective payment and
posthospital transfers to subacute care. Med Care. 1988;26:685–697.
30. White C. rehabilitation therapy in skilled nursing facilities: effects
of medicare’s new prospective payment system. Health Aff. 2003;22:
31. Kane rl, Finch m, Chen Q, et al. post-hospital home health care for
medicare patients. Health Care Financ Rev. 1994;16:131–153.
32. liu K, Wissoker D, rimes C. Determinants and costs of post-acute
care use of medicare SNFs and HHAs. Inquiry. 1998;35:49–61.
33. Bronskill S, Normand S, mcNeil B. post-acute service use following
acute myocardial infarction in the elderly. Health Care Financ Rev.
34. Juliebo V, Krogset m, Skovlund e, et al. medical treatment predicts
mortality after hip fracture. J Gerontol A Biol Sci Med Sci.
35. Berry SD, Samelson eJ, Bordes m, et al. Survival of aged nursing
home residents with hip fracture. J Gerontol A Biol Sci Med Sci.
36. roche JJ, Wenn rt, Sahota o, et al. effect of comorbidities and post-
operative complications on mortality after hip fracture in elderly peo-
ple: prospective observational cohort study. BMJ. 2005;331:1374.
37. penrod JD, litke A, Hawkes WG, et al. the association of race, gen-
der, and comorbidity with mortality and function after hip fracture. J
Gerontol A Biol Sci Med Sci. 2008;63:867–872.
38. Graham Je, Chang pF, Berges im, et al. race/ethnicity and outcomes
following inpatient rehabilitation for hip fracture. J Gerontol A Biol
Sci Med Sci. 2008;63:860–866.
by guest on December 22, 2015