Underuse of osteoporosis medications in elderly patients with fractures☆
Article Outline
Fractures related to osteoporosis result in substantial morbidity, mortality, and costs (1). Patients with a prior fracture are two to five times more likely to have future fractures than are persons without fractures (2); in these patients, pharmacologic intervention reduces spine and hip fractures by 40% to 60% (3). Nevertheless, the rates of osteoporosis treatment in patients with previous fracture have been relatively low 4, 5, 6. Most of these studies involved limited samples, with only one or a few hospitals or health care systems, and none assessed the longitudinal trends in pharmacologic treatments before and after fracture. We analyzed osteoporosis treatment trends in a large cohort of older adults in Pennsylvania.
Methods
Eligible subjects were drawn from enrollees of the Pennsylvania Pharmaceutical Assistance Contract for the Elderly (PACE), a pharmaceutical benefits program funded by the state for lower-income Medicare beneficiaries. Persons enrolled can purchase medications for a small copayment; all medications for osteoporosis are available without limitations. Persons were included in the study if they had been enrolled in PACE and Medicare for 2 consecutive years from 1994 to 2000, and had a hip or wrist fracture confirmed by inpatient and outpatient diagnosis and procedure codes from Medicare. Prescription data were examined to determine whether patients received a medication for osteoporosis before or after a fracture.
Data analysis
We first examined the characteristics of patients who had a hip or wrist fracture. The proportion of persons who filled at least one prescription for an osteoporosis medication, including alendronate, calcitonin, estrogen replacement therapy, etidronate, raloxifene, or risedronate, in the 6 months before or after the fracture was calculated for each year. Data on calcium and vitamin D supplements were limited and were thus excluded. Patients with a fracture in more than 1 year were considered in each year they sustained a fracture. To ensure 6 months of follow-up, only fractures that occurred before July 1, 2000, were included. We then defined the use of osteoporosis medications after a fracture, adjusting for potential predictors of postfracture osteoporosis medication use (prefracture osteoporosis medication use; age; sex; race; comorbid conditions; prior fractures; and nursing home residence, number of physician visits, and number of medications used in the prior 12 months) using the LSMEANS option in SAS (version 8.2; Cary, North Carolina). Finally, we constructed multivariable logistic regression models to identify patient characteristics associated with postfracture treatment. Data were collapsed across years and for type of fracture. Indicator variables for year and type of fracture were included in the fully adjusted models. Separate models were constructed with and without persons who used osteoporosis medications before the fracture.
Results
Of the 392,255 eligible persons, 21,192 comprised the study group, of whom 10,279 (49%) had a prior fracture. During the 6-year study period, there were 17,325 hip and 11,836 distal forearm fractures. The mean (± SD) age was 82 ± 7 years, 19,075 (90%) were female, and 20,550 (97%) were white. Fourteen percent of patients (n = 2967) lived in a nursing home at some point during the year before the fracture, and 58% (n = 12,274) had a hospitalization during that year. Patients had a mean of 9 ± 7 physician visits and filled prescriptions for 9 ± 5 medications in the year before the fracture.
In 1995, only 345 patients (6%) filled a prescription for an osteoporosis medication in the 6 months after a fracture, compared with 898 (22%) in 2000 (P for trend <0.001). Use of osteoporosis medications increased steadily during this period and was similar for patients sustaining a hip or wrist fracture (Figure). The increase in the use of osteoporosis medications during the 6 months before a fracture versus the 6 months after the fracture was small, averaging 3% during the study. In 2000, 2% of persons filling a prescription after a fracture received hormone replacement therapy.
Figure.
Trends in osteoporosis medication use in persons over 65 years old who had a hip (A) or wrist (B) fracture. The number of hip fractures by year was as follows: n = 3359 in 1995; n = 3286 in 1996; n = 2958 in 1997; n = 2707 in 1998; n = 2579 in 1999; and n = 2436 in 2000. The number of wrist fractures by year was as follows: n = 2388 in 1995; n = 2171 in 1996; n = 2025 in 1997; n = 1851 in 1998; n = 1781 in 1999; and n = 1620 in 2000. P for trend <0.001. White bars refer to the 6 months before a fracture and black bars refer to the 6 months after a fracture.
Younger patients, women, and whites were more likely to use a medication for osteoporosis (Table). Other factors associated with filling a prescription included fewer comorbid conditions, a prior fracture, use of oral glucocorticoids, and later study year. There was no difference in the likelihood of starting a medication between patients with hip fractures and those with wrist fractures. Use of an osteoporosis medication before a fracture was the strongest predictor of postfracture medication use (odds ratio = 21; 95% confidence interval: 19 to 24).
Table. Factors Associated with the Use of a Medication for Osteoporosis after a Hip or Wrist Fracture
| Predictor Variable | Adjusted Odds Ratio* (95% Confidence Interval) |
|---|---|
| Age | |
| 65–74 years | 1.48 (1.35–1.66) |
| 75–84 | 1.42 (1.29–1.54) |
| ≥85 | 1.0 |
| Female sex | 5.6 (4.4–7.2) |
| White race | 1.8 (1.4–2.4) |
| Comorbid conditions | |
| None | 1.54 (1.37–1.72) |
| One | 1.34 (1.19–1.47) |
| Two or more | 1.0 |
| Prior fracture | 1.22 (1.10–1.33) |
| Oral glucocorticoid use | 1.31 (1.14–1.48) |
| Study year (per year) | 1.24 (1.08–1.43) |
* Also adjusted for index fracture site; recent hospitalization; and nursing home residence, number of physician visits, and number of different medications used in the year before fracture. |
Discussion
We examined trends in pharmacologic treatment for a large cohort of older patients with a hip or distal forearm fracture. Although medication use increased steadily during the study period, by 2000 approximately 4 of 5 patients who sustained a fracture still did not fill a prescription for any osteoporosis medication in the 6 months after the fracture. The small (3%) increase in medication prescribing between the periods before and after the fracture suggests that physicians may not recognize fractures as sentinel events requiring treatment.
Prior studies of osteoporosis management patterns have documented undertreatment, but none have systematically compared medication use before and after fracture over an extended study period. For example, in a study of 1162 older women with distal radial fracture (4), 23% received prescription treatment for osteoporosis, but there was no increase in treatment during the study period from 1994 to 1997. A study of older adults hospitalized for hip fracture found that 4% of patients received osteoporosis medication on admission, whereas 5% received medication on discharge (5). A review of the medical records of 343 women who sustained a distal forearm fracture during 1993 to 1997 in Olmsted County, Minnesota, revealed that only 17% of women received advice to begin or continue treatment for osteoporosis (6). Similarly, investigators from Canada surveyed 108 patients who sustained a fragility-type fracture and found that 19% reported taking a bisphosphonate or hormone replacement therapy during the subsequent year (7).
In contrast with these previous studies, we examined pharmacy records from before and after the index fracture, which allowed us to appreciate the small increment in osteoporosis medication prescribing after a fracture. Although the proportion of persons treated increased over time, the increase between before and after the fracture was consistent. We also studied a very large cohort of patients, suggesting that undertreatment of osteoporosis was not a local issue. Moreover, all patients had access to prescription medications. Finally, we were able to examine treatment trends over time.
This study has several limitations. We did not have information on over-the-counter vitamin D and calcium supplementation use. Although these agents are important in osteoporosis treatment, most guidelines for postfracture treatment emphasize prescription-strength antiresorptive medications 8, 9. We also did not have information on other fracture prevention measures that may have been instituted after the index fracture, such as home safety evaluations and lower extremity strengthening. The study database contains information only on filled prescriptions; thus, prescriptions never filled by patients were not included. In addition, patients were relatively old and of low-to-moderate income status. We did not include persons with spine or other types of fracture because of concerns about inaccurate coding. It seems unlikely, however, that the results would differ if we included patients with other fractures.
These data suggest that osteoporosis is poorly treated after a new fracture, and that certain groups of patients, such as men, older persons, nonwhites, and those with more comorbid conditions, are less likely to receive treatments. Why such subgroups are receiving less care is not known, but treatment for all patients after an initial fracture is important because repeat fractures are relatively common (2).
Acknowledgements
The authors thank Tom Snedden and the staff of the Pennsylvania Pharmaceutical Assistance Contract for the Elderly for help with many aspects of the project.
References
- . Healthy People 2010 (Understanding and Improving Health). 2nd ed. Washington, DC: U.S. Government Printing Office; 2000;
- . Patients with prior fractures have an increased risk of future fractures (a summary of the literature and statistical synthesis). J Bone Miner Res. 2000;15:721–739
- . Treatment of postmenopausal osteoporosis. JAMA. 2001;285:1415–1418
- . Treatment of osteoporosis (are physicians missing an opportunity?). J Bone Joint Surg. 2000;82A:1063–1070
- . Failure to diagnose and treat osteoporosis in elderly patients hospitalized with hip fracture. Am J Med. 2000;109:326–328
- Osteoporosis intervention following distal forearm fractures. A missed opportunity?. Arch Intern Med. 2002;162:421–426
- . Investigation and treatment of osteoporosis in patients with fragility fractures. CMAJ. 2000;163:819–822
- . Osteoporosis (Review of the Evidence for Prevention, Diagnosis, and Treatment and Cost-effectiveness Analysis). Washington, DC: The Foundation; 1998;
- . Guidance on the Prevention and Management of Corticosteroid Induced Osteoporosis. Bath, United Kingdom: National Osteoporosis Society; 1998;
☆ Supported by grants AR48616, DK02759, AR47782, and AR02123 from the National Institutes of Health, Bethesda, Maryland; and the Arthritis Foundation, Atlanta, Georgia.
PII: S0002-9343(03)00357-7
doi:10.1016/S0002-9343(03)00357-7
© 2003 Excerpta Medica Inc. All rights reserved.
