The majority of the 1.8 million individuals who sustain a fracture annually in the United States have bone loss and osteoporosis1,2, yet <15% of them subsequently receive treatment for osteoporosis1. The primary reason for this lack of medical care is often reported to be a lack of understanding on the part of the treating physician regarding the role of osteoporosis as the cause of the fracture3-11.
A 2004 report by the U.S. Surgeon General focused attention on the urgent need for physicians to address the increasing burden of osteoporosis. It was noted that primary care physicians and orthopaedic surgeons rarely discussed osteoporosis with patients who have had a fracture that resulted from osteoporosis1. In response, the American Orthopaedic Association (AOA) position statement "Leadership in Orthopaedics: Taking a Stand to Own the Bone"12 acknowledged that the occurrence of a fragility fracture or a fracture in an osteoporotic adult was pathognomonic of low bone quality and identified this patient as being at high risk for subsequent fractures13,14. Orthopaedic surgeons were challenged to become champions for quality improvement in the treatment of osteoporosis. The Own the Bone project sought to improve the practices of orthopaedic surgeons with regard to the prevention of subsequent fractures in patients with a fragility fracture. The Own the Bone project changed the behavior of orthopaedic surgeons; the rate of counseling regarding osteoporosis increased to 90%, and recommendations for bone-mineral density testing were also made more frequently. A consultation letter sent to all primary care physicians participating in the study included information on osteoporosis risk factors, comorbidities, and recommendations that underscored the need for further testing and for osteoporosis therapy12. Primary care physicians are more likely to treat osteoporosis if it has been confirmed by bone-mineral density testing than if its existence is only suspected on the basis of the fracture history15.
We performed a follow-up to the Own the Bone pilot project in which we performed a prospective cohort study to compare two models of care for osteoporosis: an immediate intervention, in which the osteoporosis consultation was performed and treatment recommendations were given during the hospitalization, and a delayed intervention, in which recommendations for osteoporosis treatment were conveyed to the primary care physician. The outcomes of interest included the rate of bone-mineral density evaluation and the rate of osteoporosis treatment following a fragility fracture.
Project Participants and Organizations
Two academic hospitals that had participated in the Own the Bone pilot project12, Dartmouth-Hitchcock Medical Center (Lebanon, New Hampshire) and Northwestern Memorial Hospital (Chicago, Illinois), participated in the current follow-up study. The research protocol was approved by the institutional review boards, and all participants signed an informed consent form. Researchers attempted to enroll all patients fifty years of age or older who were hospitalized with a fragility fracture between January 1, 2007, and December 31, 2007, and who had not previously been diagnosed with or treated for osteoporosis. A fragility fracture was defined as a fracture that occurred spontaneously or in response to minimal trauma, such as a fall from a standing position or while walking16. Patients with metastatic cancer, other metabolic bone diseases, end-of-life care, end-stage renal disease, or an inability to provide consent were excluded.
Education, Evaluation, and Treatment
Patient counseling and education was provided by the orthopaedic surgeon and a nurse clinician. Participants received information regarding dietary sources of calcium, exercise, smoking cessation, calcium and vitamin-D supplements, fall prevention, home safety, bone-mineral density testing, the fact that they were at high risk for a subsequent fracture, and the ability of osteoporosis therapy to reduce that risk.
The two interventions were stratified by hospital. Patients in the first hospital received delayed osteoporosis care consisting of recommendations for osteoporosis evaluation and treatment that were communicated to the primary care physician in a consultation letter. Patients in the second hospital received immediate osteoporosis care arranged during a consultation that occurred in the hospital. An osteoporosis consultant evaluated the patient and initiated pharmacologic therapy for osteoporosis. Therapy included calcium and vitamin-D supplements as well as Food and Drug Administration (FDA)-approved medicine(s) for osteoporosis (bisphosphonates [alendronate, risedronate, or ibandronate] and selective estrogen receptor modulators [raloxifene, calcitonin, estrogen, or teriparatide]). The patient received a prescription for the pharmacologic therapy, orders for bone-mineral density testing with use of dual x-ray absorptiometry (DXA), and a recommendation for outpatient follow-up with his or her primary care physician or at the Bone Health Center17.
In the delayed-care group, the consultation letter containing clinical guidelines regarding osteoporosis treatment and recommendations for bone-mineral density testing and osteoporosis therapy was sent to the primary care physician through the hospital's integrated electronic medical-record system. In the immediate-care group, a similar consultation letter discussing clinical guidelines for osteoporosis treatment and recommending that the patient undergo the ordered bone-mineral density test and adhere to the prescribed osteoporosis treatment was sent to the patient's primary care physician.
Follow-up and Analysis
Participants were interviewed six months after the fracture regarding the follow-up care that they had received from their primary care physician, including whether they had undergone bone-mineral density testing and whether they were receiving treatment for osteoporosis. Participants were asked whether they were taking calcium, vitamin D, and an FDA-approved medication for osteoporosis. The participant's medical and pharmacy records were reviewed to obtain information on recent office visits, medication use, and the performance of bone-mineral density testing. If the osteoporosis treatment recommendations were not being followed, the medical records were examined for any possible reasons that had been noted by the physician.
We also calculated the FRAX score of the participants in this study18. The FRAX tool (http://www.shef.ac.uk/FRAX/) was developed by the World Health Organization to evaluate the ten-year risk of hip fracture and other major osteoporotic fractures in an individual patient. It is based on a model that integrates the risks associated with clinical factors as well as with the bone-mineral density at the femoral neck. FRAX models have been developed from studies of population-based cohorts from Europe, North America, Asia, and Australia18.
Demographic variables were studied with use of descriptive statistics. Dichotomous outcome measures included follow-up with the primary care physician (yes/no), bone-mineral density testing (yes/no), and osteoporosis treatment (yes/no). Data were analyzed with use of the Fisher exact test and chi-square analysis.
One hundred and seventy-six participants were enrolled in the study; seventy-five were in the immediate-care group and 101 were in the delayed-care group. A total of 115 participants (65%) were successfully contacted for the six-month follow-up, including forty-nine (65%) in the immediate-care group and sixty-six (65%) in the delayed-care group. The mean age (and standard deviation) was 73 ± 10 years (range, fifty to ninety-nine years), and 80% of the participants were women (Table I). Eighty-seven percent of the participants surveyed (forty of forty-nine in the immediate-care group and sixty of sixty-six in the delayed-care group) had seen their primary care physician following hospital discharge. Seventy-nine percent of the participants surveyed (thirty-nine in the immediate-care group and fifty-two in the delayed-care group; p = 0.2) had discussed osteoporosis with their primary care physician. Bone-mineral density testing had been performed more frequently in the immediate-care group (92% [forty-five of forty-nine]; 95% confidence interval [CI], 85% to 100%) than in the delayed-care group (76% [fifty of sixty-six]; 95% CI, 69% to 82%) (p < 0.05). The fractures in the immediate-care group were located in the hip in thirty-two patients (43%), in the distal portion of the femur in seventeen (23%), in the tibia in six (8%), in the humerus in thirteen (17%), and in the wrist in seven (9%). The fractures in the delayed-care group were located in the tibia in fifty-three patients (52%), in the humerus in eleven (11%), and in the wrist in thirty-seven (37%). The sixty-one participants (35%) who were lost to follow-up had a mean age of 78 ± 8 years; 92% were female and most required nursing home placement at the time of hospital discharge.
Calcium, Vitamin D, and Pharmacologic Therapy
The majority of participants surveyed were consuming calcium and vitamin-D supplements six months after the fracture, including forty-two (86%) of forty-nine in the immediate-care group and forty-five (90%) of sixty-six in the delayed-care group. The rate of calcium and vitamin-D usage had increased significantly from the baseline rate in both groups (p = 0.03) (Fig. 1). Osteoporosis therapy (calcium, vitamin D, and an FDA-approved medication for osteoporosis) had been initiated in forty-one (84%; 95% CI, 79% to 93%) of the forty-nine surveyed patients in the immediate-care group by the time of hospital discharge when indicated and in no patients in the delayed-care group.
The osteoporosis treatment rate at the six-month follow-up was substantially greater than the baseline rate of 0% at the time of hospital admission (p < 0.001) in both groups. Thus, both immediate and delayed care resulted in an increase in the rate of osteoporosis treatment following a fragility fracture. The proportion of individuals who were receiving osteoporosis therapy was greater in the immediate-care group (67%; 95% CI, 52% to 85%) than in the delayed-care group (30%; 95% CI, 18% to 43%) (p = 0.001). Although most patients had undergone bone-mineral density testing, a gap in the medical care provided by the primary care physician had occurred between the bone-mineral density testing and the initiation of osteoporosis therapy (Fig. 2). The probability that the physician would initiate therapy was not affected by the age of the patient (p = 0.47) or by the location of the fragility fracture (p = 0.25). It should be noted that some of the observed difference between the groups could have resulted from the difference between the two study centers rather than from a difference between the interventions, if the staff at one of the centers had received substantially more education regarding osteoporosis than the staff at the other center.
Predicted Ten-Year Fracture Risk and the Effect of Age
A seventy-year-old white female patient (height, 1.6 m; weight, 52 kg) with a current fracture and a balance disorder resulting in a history of falls has a 34% risk of a major fracture, including a 14% risk of a hip fracture, within ten years. An eighty-four-year-old woman with similar risk factors would have a 47% risk of a major fracture and a 27% risk of a hip fracture within ten years. Both of these individuals would therefore qualify for pharmacologic intervention. However, a fifty-five-year-old woman with a recent wrist fracture resulting from a fall would have only a 12% risk of a major fracture and a 1.8% risk of a hip fracture within ten years, which would not meet the criteria for pharmacologic therapy. Recent National Osteoporosis Foundation guidelines for the U.S. recommend osteoporosis treatment for individuals with a vertebral or hip fragility fracture19.
Reasons for Nonadherence to Recommendations
Osteoporosis therapy was not initiated in two individuals in the delayed-care group whose survival was expected to be limited and who required end-of-life care. In addition, it is noteworthy that ten individuals in this group declined medical care for osteoporosis because of the high cost of the medication and concern regarding the possible side effects of the medication. Patients in the immediate-care group received prescriptions and bone-mineral density testing orders during the osteoporosis consultation scheduled during their hospital stay. Ten patients in this group declined to participate in the consultation; six of these patients commented that their primary care physician would address osteoporosis with them if it was truly an important issue. Eight additional patients had discontinued osteoporosis therapy by the time of the six-month follow-up; three attributed the discontinuation to financial constraints, and five to the development of side effects.
Our study assessed the effectiveness of immediate and delayed osteoporosis care following a fragility fracture. Both interventions resulted in a high rate of bone-mineral density testing as well as an increase in the rate of osteoporosis treatment. However, the treatment rate was higher in the immediate-care group than in the delayed-care group six months following the fracture. We hypothesize that the lower treatment rate in the delayed-care group may have been due to failure of the primary care physician to initiate appropriate osteoporosis therapy after receiving bone-mineral density results. The identification of this gap in medical care between bone-mineral density testing and osteoporosis treatment will allow further targeted quality-improvement interventions.
The rate of osteoporosis treatment of patients with a fragility fracture increases when an orthopaedic surgeon intervenes. The osteoporosis treatment rates in our study were equal to or greater than those resulting from information technology-based interventions20. Communication with the primary care physician was effective, as our treatment rates were higher than those in previous reports on interventions to increase awareness of osteoporosis by primary care physicians following a wrist fracture (30% compared with 21%)21. Efforts by the orthopaedic community to raise awareness of the presence of osteoporosis in patients with a fragility fracture can positively impact the care of older adults12. Multidisciplinary consultations involving the collaboration of orthopaedic surgeons, medical specialists, and primary care physicians have been reported17,22-25. Osteoporosis therapy following a fragility fracture is recommended in all clinical guidelines of which we are aware1,2,13,26-28. The 2010 Osteoporosis Prevention, Diagnosis, and Therapy Consensus Conference recommended therapy in patients with osteoporosis who have a ten-year fracture risk of >20% overall or a >3% risk of a major fracture such as a hip fracture29.
Although orthopaedists consider a hip or vertebral fragility fracture to be "highly indicative" of low bone quality12, primary care physicians may not fully appreciate the import of such a fracture in the absence of a bone-mineral density test confirming the diagnosis of osteoporosis. Thus, the primary care physician is more likely to treat osteoporosis on the basis of bone-mineral density test results than on the basis of the occurrence of a fragility fracture15. Unfortunately, medical decision-making becomes more complex as the primary care physician adds an additional procedure, which may lead to an increase in medical errors or omissions28,30,31. Furthermore, a fragility fracture may occur in a patient with a bone-mineral density T-score above the osteoporosis cutoff value of —2.532. In the Rotterdam epidemiologic study, 56% of women and 79% of men with a nonvertebral fracture had a T-score of ≥—2.533. The Study of Osteoporotic Fractures Research Group found that 54% of women with a nonvertebral fracture had a T-score of ≥—2.032. Primary care physicians may be uncertain of the correct treatment in a patient with an apparent fragility fracture but without osteoporosis34. Few primary care physicians are versed in the use of the FRAX calculator, which would provide guidance regarding a patient's suitability for pharmacologic therapy. Advances in the treatment of osteoporosis after a fragility fracture35 have not been highlighted sufficiently in the medical literature, and educational interventions with primary care physicians have proven only modestly effective20,36,37. A lower rate of adherence to osteoporosis treatment guidelines has been observed in association with advanced patient age, male sex, black race, and the presence of comorbidities38,39. Some primary care physicians may also have concerns regarding the effect of bisphosphonate therapy on the healing of the initial fracture, although the benefits of osteoporosis therapy following a fragility fracture seem to outweigh this concern40-42.
The importance of identifying patients with a fracture as being at high risk for a subsequent fracture34 has been highlighted by numerous specialty organizations such as the American Academy of Orthopaedic Surgeons26, the American Orthopaedic Association12, the National Osteoporosis Foundation19, the American Association of Clinical Endocrinologists41, the U.S. Preventive Services Task Force27, and the Office of the U.S. Surgeon General1. Current quality measures such as the National Committee of Quality Assurance (NCQA) HEDIS (Healthcare Effectiveness Data and Information Set) measure also track the appropriate use of osteoporosis prevention43 The Physician Quality Reporting Indicators (PQRI) used by the Centers for Medicare & Medicaid Services (CMS) "Pay for Performance Measures"44 and the quality measures used by the Joint Commission will include similar tracking of adherence to osteoporosis treatment guidelines by hospitals45.
Patients are often hesitant to adhere to osteoporosis treatment recommendations because of their low awareness of osteoporosis46. This was noted in the current study, in which 10% to 15% of the patients in each group declined treatment and up to an additional 16% discontinued medication within six months following the fracture. The results of prior studies have shown that up to 19% of women discontinue osteoporosis medication during the first six months, citing the side effects and cost of the medication or a lack of motivation47,48. In addition, as many as 18% of patients who do receive osteoporosis medication do not adhere to all of the requirements for its proper administration49.
Our study has limitations. First, one-third of the patients were lost to follow-up, and this group tended to be older than the remaining patients and be likely to have greater functional and/or cognitive impairment requiring nursing home care as well as a higher mortality rate50-52. Therefore, a bias may have been introduced by surveying a group that was relatively younger and healthier than the initial cohort. Second, the study compares contrasting strategies implemented at two different centers. Thus, differences resulting from the strategies could not be distinguished from differences resulting from the location, including differences in the institutional culture of osteoporosis awareness, the institutional culture of osteoporosis care, and patient awareness of osteoporosis. Third, our study was conducted at academic institutions, and physician practices were integrated, which allowed the researchers to review patient medical records; thus, our findings may not be generalizable to community hospitals. Fourth, we could not verify that the primary care physician received the letter from the orthopaedist; however, the 75% rate of counseling and bone-mineral density testing was high, suggesting that the message from the orthopaedist was reaching the appropriate person, as has been demonstrated in prior studies20. Fifth, the distribution of fractures differed between the two cohorts, which could have introduced a bias by affecting the follow-up treatment that the patients received. Finally, we did not assess the level of comorbidity, which may have played a role in the primary care physician's decision to withhold treatment in some cases.
In conclusion, both immediate and delayed-care strategies resulted in an increase in the rate of osteoporosis treatment. However, immediate care for osteoporosis, involving a consultation and the prescription of osteoporosis therapy at the time of hospitalization for a fragility fracture, resulted in a higher osteoporosis treatment rate six months after the fracture compared with delayed care. Barriers to osteoporosis care by outpatient physicians merit further research.