Use of a Peripheral Dexa Measurement for Osteoporosis Screening

Vol. 34, No. 3 201 Clinical Research and Methods Use of a Peripheral Dexa Measurement for Osteoporosis Screening Julienne K. Kirk, PharmD; Mindy Nichols, RD; John G. Spangler, MD Background and Objectives: Osteoporosis is a substantial cause of morbidity and mortality in the United States. While screening for this disease is important, few studies have evaluated the role of an osteoporosis screening device in a family practice setting. This study evaluated the influence of a peripheral-dexa (p-dexa) heel bone mineral density (BMD) measurement and a patient education program on changes in pharmacologic treatment for suspected osteopenia or osteoporosis by primary care physicians over a 1-year follow-up. Methods: Using a computerized database (ages 50 to 75 years), 1,927 women were identified. An invitation was mailed to register for a screening to have a p-dexa heel BMD scan done and to attend an osteoporosis prevention presentation. Medication history, age, height, weight, and risk factors for osteoporosis were collected. A follow-up chart review was carried out on women who were found to have heel BMD T-scores of <-.6 (suggested by the World Health Organization). Date of menopause onset, pharmacotherapy for osteoporosis, calcium, vitamin D, and physician intervention were assessed. Results: There were 292 women (15.2%) who self registered, obtained BMD testing, and attended an educational program. Of these women, 87 (30%) had at least one risk factor for osteoporosis, in addition to menopause. Mean BMD was.489 ±.113 gm/cm 2 (normal >.42 gm/cm 2 ). A post-screening chart review was completed in 102 women (36.6%) at greatest risk for osteoporosis based on a T-score <-.6. Following the intervention, 26 women were started on antiresorptive therapy (primarily estrogen), and three additional women had a second antiresorptive agent added to estrogen. Conclusions: P-dexa heel BMD has utility for screening patients at risk for osteoporosis. However, only 15% of invited women attended the screenings, and pharmacotherapy treatment did not significantly change after screening in the majority of women at risk for osteoporosis, based on p-dexa screening. (Fam Med 2002;34(3):201-5.) Osteoporosis is a common disorder affecting millions of Americans, predominantly postmenopausal women. 1 The estimated direct medical expenditures for osteoporotic fractures in 1995 was $13.8 million. 2 The estimated lifetime risk of fracture of the hip or wrist or clinically diagnosed vertebral fracture is about 40% in a 50-year old Caucasian woman. 3,4 Hip fractures are of particular concern because of the 300,000 hip fractures that occur each year in the United States, 20% of those lead to death within 1 year. 1,5 There are an estimated 10 million Americans with osteoporosis and an additional 18 million who have low bone mass. 1 While screening for osteoporosis among low-risk women by bone mineral density (BMD) testing has not From the Department of Family and Community Medicine, Wake Forest University. been recommended by the US Preventive Services Task Force, 6 women with multiple risk factors may warrant screening and treatment. 1,7-9 The National Osteoporosis Foundation recommends that all women under age 65 with one or more risk factors for osteoporosis (besides menopause) and all women over age 65, regardless of risk factors, be screened by BMD testing to detect this disorder. 1 BMD should be measured among these high-risk women because it may influence their decision to start antiresorptive therapy, 10-12 either hormonal (such as estrogen replacement or calcitonin) or nonhormonal (such as bisphosphonates or raloxifene) in addition to the use of calcium and vitamin D. Several techniques are available for measurement of BMD. Single-energy X ray absorptiometry of the calcaneus (heel) is an option for noninvasisve measurement of bone mass. Advantages to this technique include low cost, precision, and low amounts of radiation used. 12,13 A peripheral-dexa (p-dexa) scan may be

202 March 2002 Family Medicine an alternative for screening to detect individuals who have low bone mass. 13-15 The p-dexa determines bone mass by measuring the decrease in the strength of radiation that passes through tissue at the site being evaluated, such as the heel. While dual-energy X ray absorptiometry (dexa) is the standard for BMD testing, this method is costly and not available to all patients, especially for screening. 16 P-dexa devices are portable and are increasingly being used in traditional medical settings and even in pharmacies. 14 In addition, peripheral measurement techniques are attractive for safety reasons. 17 Screening and management strategies for the prevention and treatment of osteoporosis are greatly underused. For example, in a study of osteoporosis screening of 1,162 patients with confirmed distal fractures, only 279 (24%) underwent either diagnostic evaluation or treatment for osteoporosis. 19 The authors also found that only 16.2% of patients filled a prescription for osteoporosis medication after a fracture. 18 Some research indicates that direct disclosure of BMD to women and their general practitioner leads to increased knowledge of BMD status, without increasing anxiety. 20 The current study evaluated the influence of p-dexa (PIXI machine made by the Lunar Corporation) on treatment status before and after a patient-focused education program. Women eligible for pharmacologic intervention were identified by p-dexa scanning, and we then measured the rates at which these women were started on osteoporosis medication. Methods Setting The study was carried out in a large family practice teaching program consisting of approximately 20 faculty physicians, 30 family practice residents, and 3 physician assistants. There are also nutritionists, nurses, a pharmacist, and a variety of other health care professionals who work in this program providing patient care. The screening took place in the patient education area of the outpatient clinic. The breakdown of payor mix in this practice includes Medicare (15%), managed care (63%), commercial insurance (9%), Medicaid (8%), self pay (4%), and workers compensation (1%). There were approximately 60,000 patient visits annually to this family practice center in 1999 when this chart review was completed, and there were approximately 20,000 total patients registered in the practice. Study Procedures Approval was obtained by the Institutional Review Board at Wake Forest University Baptist Medical Center for the chart review. The study was conducted between May and October of 1999. Subjects We identified potential subjects for this study from a database of female patients in our family practice who were ages 50 to 70. We excluded those whose records already listed an International Classification of Diseases, Ninth Edition-Clinically Modified (ICD-9-CM) code for osteoporosis. A total of 1,927 women were identified. These women were recruited by a mailed invitation that asked the women to register for an osteoporosis BMD screening, using the p-dexa, and for an education program on bone health done simultaneously with the p-dexa scanning. Data Collected Participants reported risk factors for osteoporosis on a 10-item questionnaire prior to having the BMD test. Women were told that this questionnaire represented common risk factors associated with osteoporosis. Data collected from patients included age, race, height, weight, smoking history (past and current), alcohol consumption, activity level, calcium intake, patient and family fracture history, and medication history. Other risk factors for osteoporosis collected from the patient included a history of back pain, kyphosis, height loss, and early menopause (defined as age at menopause of less than 45 years). After the BMD screening and after completing the questionnaire, participants attended a 45-minute patientfocused presentation on nutrition, exercise, and medication options to prevent and restore bone loss. In addition, the primary care providers in this practice were offered a lunch conference that described the project and the implications of p-dexa data. Announcements and information were also made at resident and faculty meetings about this project. BMD Measurements We measured BMD on all subjects with the p-dexa scan. Guidelines for measurement of BMD are based on T-scores, defined as the standard deviation from the mean of healthy young adults. A T-score represents the mean peak of bone mass. 18 The heel has a slower bone loss rate than other sites in the body, such as the hip, spine, or forearm. This means that the T-score used from other skeletal areas may underestimate BMD loss if the same standards are used to measure the heel. 14,18 The World Health Organization (WHO) T-score for hip, spine, and forearm is defined as normal at greater than -1, low bone density (osteopenia) at a reading between -1 and -2.5, and osteoporosis at a T-score less than - 2.5. 21 The WHO equivalent for heel BMD includes >-.6 for a normal T-score, -.6 to -1.6 for osteopenia, and less than -1.6 for osteoporosis. 18,21 The participant s primary care provider received a p-dexa heel BMD report that contained T-score and the date of the mea-

Clinical Research and Methods surement, and this information was placed in the medical record. Study Variables For those women with a p-dexa T-score < -.6, a postscreening chart review was performed 6 months after screening. This gave patients at least 6 months to have contact with their physician to review screening results that were abnormal. Medical charts and medication profiles were specifically evaluated for past or current treatment of osteoporosis. Corroborating documentation for initiation and/or continued use of pharmacotherapy was found in the primary care provider written notes. Past medical history obtained included BMD measurements, hysterectomy, estrogen deficiency, hyperparathyroidism, and height loss. Medical record flow sheets containing demographic data, drugs, preventive lifestyle information, and office visits were reviewed. Table 1 Vol. 34, No. 3 Presence of Selected Risk Factors and Test Results Among Study Subjects* Historical Risk Factors Percent Postmenopausal 96.6 Inactivity 67.1 Hysterectomy 66.7 Lack of calcium supplement 38.0 Cigarette smoking 33.6 Alcohol use 21.9 Hypothyroidism 15.8 Kyphosis 9.6 Fractures 4.1 Early menopause 1.0 Family history of osteoporosis 0 Positive Screening Tests p-dexa Osteopenia 22.9 p-dexa Osteoporosis 11.9 203 Data Analysis Descriptive statistics were used to describe general characteristics and presence of risk factors for women attending the program. For women at highest risk for low BMD (T-score <-.6), chi-square (for categorical data) and ANOVA (for continuous data) tests were used to evaluate correlates related to change in treatment status before and after the program. Backward stepwise logistic regression analysis was also performed to document prediction of antiresorptive therapy after program presentation, using the Wald statistic, which has a chisquare distribution, for inclusion within the model. The significance level was set at.05. Results Subjects A total of 292 women (15.2%) attended the BMD screening and the education session. Of the women who participated in the BMD screening, 262 were Caucasian, 27 were African-American, 2 were Hispanic, and 1 was Asian. The mean age of participants was 59.7 years, mean height was 64.4 inches, and average weight was 73.2 kgs, with a mean body mass index of 27.4. Selected risk factors for women who volunteered for the program are outlined in Table 1. They reported an average of three osteoporosis risk factors. Overall, 24 women (8.1%) had one risk factor other than menopause, 32 (10.8%) had two risk factors, and 31 (10.5%) had three risk factors. BMD Measurement Mean BMD from the p-dexa PIXI machine was.489 +.113 gm/cm 2 (normal >.42 gm/cm 2 ). There were 102 (35%) women with T-scores indicating osteopenia or osteoporosis (T-score < -.6). Characteristics of these women by treatment status before and after the screening program are shown in Table 2. * n = 292 High-risk women were defined as those women with a positive BMD screening. Twenty-seven high-risk women (26.5%) had documentation of a p-dexa result in their medical chart at an office visit with their primary care provider. There was a central dexa scan performed in 14 women who had p-dexa heel BMD results indicative of osteopenia (T-score -.6 to -1.6) and in four women with osteoporosis (T-score < -1.6). The results of these central dexa T-scores were consistent with the diagnosis of osteopenia or osteoporosis in all of these cases. There were four women with a central BMD done within the last 2 years, and none of these women had a change in their pharmacotherapy regimen. Antiresorptive Therapy Among the 162 women with low BMD on p-dexa scans, 33 (32.4%) never received antiresorptive therapy either before or after the screening program, 40 (39.2%) were already taking antiresorptive therapy, and 29 (28.4%) were started on a new prescription for antiresorptive therapy after the program. The only statistically significant correlate of treatment status among these women was kyphosis, with five out of six women with this condition started on a new prescription for antiresorptive therapy. However, in multivariate analysis using logistic regression, we did not find any characteristics or risk factors that predicted which of the participants would be started on treatment. Among the 40 women already on antiresorptive therapy, 95% were receiving estrogen. For those women started on a new antiresorptive medication, 10.3% were started on alendronate, 82.8% were started on estrogen, and the remainder were begun on nasal calcitonin

204 March 2002 Family Medicine or raloxifene (Table 3). Of the 102 women evaluated on post-intervention chart review, 60.8% (62/102) were taking estrogen, and 24 were newly started on estrogen after the p-dexa BMD was obtained. Discussion In this study, we evaluated change in treatment status among women found to be osteopenic during a p-dexa screening program. Overall, 292 women registered to attend the educational program and obtained a heel p-dexa BMD. This was a small participant rate (15.2%), considering we mailed invitations to 1,927 women. Of the 292 women, 102 were found to have low BMD. Only 27 of these women, however, had documented follow-up with their primary care clinician. Overall, therefore, the program did not result in many women coming or seeking follow-up after the program to have therapy changed The generalizability of our results are limited since we did not evaluate the eligible women who did not participate in the screening program. Further, limitations on the use of a p-dexa measurement include an underestimation of bone loss, since the heel has a slower bone loss rate than other sites in the body. Thus, the p-dexa can be advocated only as an instrument for screening. Use of a p-dexa measurement for monitoring effectiveness of treatment for osteoporosis is not recommended. This disadvantage, however, is offset by the advantages of p-dexa for screening, which include low X ray dose (limited to use in the heel or forearm), no clothing restrictions, Table 3 Women for Whom a Change Was Made in Treatment* TREATMENT AFTER TREATMENT PRIOR INTERVENTION Women With Osteopenia (p-dexa -.6 to -1.6) No pharmacotherapy (n=15) Addition of estrogen No pharmacotherapy (n=1) No pharmacotherapy (n=1) Addition of raloxifene Addition of calcitonin Women With Osteoporosis (p-dexa < -1.6) No pharmacotherapy (n=9) Addition of estrogen * n=29 Table 2 Characteristics of 102 Women With Low BMD (T Score < -.60), by Treatment Status TREATMENT STATUS Never* Prior* New* Characteristics (n=33) (n=40) (n=29) P Value** Age 61.5 + 6.6 61.0 + 6.8 62.2 + 5.7 ns T-score -1.52 +.1-1.50 +.78-1.61 +.58 ns BMD.38 +.04.38 +.06.37 +.05 ns Height 63.6 + 2.3 63.9 + 2.1 64.0 + 2.3 ns Weight 147.8 + 27.0 142.0 + 28.8 142.1 + 18.8 ns Caucasian 30 (90.9) 39 (97.5) 28 (96.5) ns Smoker 4 (12.1) 4 (10.0) 6 (20.7) ns Postmenopausal 26 (78.8) 31 (77.5) 28 (96.5).07 Kyphosis 1 (3.0) 0 (0) 5 (17.2).008 Hysterectomy 11 (33.3) 17 (42.5) 10 (34.5) ns Fracture 4 (12.1) 4 (10.0) 2 (6.9) ns Family history 4 (12.1) 3 (7.5) 6 (20.7) ns Early menopause 9 (27.3) 12 (30.0) 8 (27.6) ns Calcium supplementation 14 (42.2) 20 (50.0) 12 (41.4) ns Alcohol use 7 (21.2) 12 (30.0) 6 (20.7) ns * Never refers to no antiresorptive treatment before or after heel densitometry screen, prior refers to women already on treatment before heel densitometry, and new refers to women started on therapy after heel densitometry. Values in columns are number of subjects (percent) for categorical data (mean + SD) for continuous data. **P values were derived from chi-square for categorical data on ANOVA for continuous data with the groups (never, prior, and new treatment status) compared to each other. ns not significant and immediate results (5 seconds). Other data support the accuracy and precision of peripheral machines. 22 In this study, the consistency between the results of the p-dexa and central dexa scan for the 18 women who underwent both procedures is encouraging. The data from our study indicate that notifying the primary care provider by written study results (p-dexa BMD data) did not effectively result in a change or institution of pharmacotherapy treatment. Some providers may not believe that a p-dexa result is valid or that this information warranted further follow-up. Or this may have been due to practice procedures in our office. For example, physicians may not have seen the report in the patient s chart, depending on where it was filed. Our intention could have been more successful if a structured, mandatory education component detailing the p-dexa and osteoporosis was included along with practice management procedures to assure that physicians saw the p-dexa results. For example, approaches to change our systems to facilitate better patient follow-up and care could have included reminders to patients to discuss the p-dexa information with their provider at an office visit. This would have also allowed an opportunity to encourage patient referral by the primary care provider.

Clinical Research and Methods Vol. 34, No. 3 205 The results of the 102 women on follow-up chart review were surprising because many of the risk factors for osteoporosis, such as early menopause, fracture history, and smoking, did not correlate with treatment status. Indeed, we did not find our patient characteristics or risk factors in multivariate analysis that predicted which of these osteopenic women would go on to receive antiresorptive medication. However, this finding should be interpreted cautiously considering our sample size (n=102). In bivariate analysis (Table 2), only kyphosis was associated with new use of antiresorptive therapy (P<.008). A larger sample size may have highlighted these and other correlates as significant predictors of receiving antiresorptive therapy. Unfortunately, only a small percentage of high-risk women had a follow-up visit that was documented in the patient s chart with their primary care provider (27 out of 102) over the course of the 1-year chart review. The limitation to this finding is that a long enough period of time may have not elapsed from the initial BMD results, patients could have changed physician, or patients could have died. Conclusions In summary, the implementation and delivery of a program to screen for osteoporosis and educate patients about the risks and treatment of osteoporosis was accomplished in this study. Unfortunately, the program did not result in a major change in the percentage of women with low BMD receiving anti-osteoporosis treatments. Still, we believe that while there are some limitations to the use of a p-dexa BMD scan, the use of such a device for screening women at risk should be considered. Increased education and improved practice management interventions could enhance osteoporosis diagnosis and treatments. Acknowledgment: We thank Novartis for the use of the p-dexa machine (PIXI, Lunar Corporation) that was implemented in this study. Corresponding Author: Address correspondence to Dr Kirk, Wake Forest University, Department of Family and Community Medicine, Medical Center Boulevard, Winston-Salem, NC 27157-1084. 336-716-9043. Fax: 336-716- 9126. jkirk@wfubmc.edu. REFERENCES 1. National Osteoporosis Foundation. Physician s guide to prevention and treatment of osteoporosis. Hillsborough, NJ: Excerpta Medica, Inc, 1998. 2. Ray NF, Chan JK, Thamer M, Melton LJ. Medical expenditures for the treatment of osteoporotic fractures in the United States in 1995: report from the National Osteoporosis Foundation. J Bone Miner Res 1997; 12:24-35. 3. Melton LJ III, Chrischilles EA, Cooper C, Lane AW, Riggs BL. Perspective: how many women have osteoporosis? J Bone Miner Res 1992;7:1005-10. 4. Ross PD. Osteoporosis frequency, consequences, and risk factors. Arch Intern Med 1996;156:1399-1411. 5. Consensus development conference: diagnosis, prophylaxis, and treatment of osteoporosis. Am J Med 1993;94:646-50. 6. US Preventive Services Task Force. Guide to clinical preventive services, second edition. Baltimore: Williams and Wilkins, 1996. 7. Cummings SR, Nevitt MC, Browner WS, et al. Risk factors for hip fracture in white women. N Engl J Med 1995;332:767-73. 8. American Association of Clinical Endocrinologists Osteoporosis Task Force. AACE clinical practice guidelines for the prevention and treatment of postmenopausal osteoporosis. Endocrine Practice 1996;2:155-71. 9. Kanis JA, Delmas P, Burckhardt P, Cooper C, Torgerson D. Guidelines for diagnosis and management of osteoporosis. Osteoporos Int 1997;7:390-406. 10. American College of Obstetrics and Gynecology. Hormone replacement therapy. Technical bulletin no. 166. Washington, DC: American College of Obstetrics and Gynecology, 1992. 11. Siris E, Miller P, Barrett-Conner E, Abbott T, Sherwood L, Berger M. Design of NORA, the National Osteoporosis Risk Assessment Program: a longitudinal US registry of postmenopausal women. Osteoporos Int 1998;8(suppl 1):S62-S69. 12. Eastell R. Treatment of postmenopausal osteoporosis. N Engl J Med 1998;338:736-46. 13. Blake GM, Patel R, Fogelman I. Peripheral or axial bone density measurements? Journal of Clinical Densitometry 1998;1:55-63. 14. Miller PD, Bonnick SL, Johnston CC, Kleerekoper M, Lindsay RL, Sherwood LM, et al. The challenges of peripheral bone density testing. Which patients need additional central density measurements? Journal of Clinical Densitometry 1998;1:211-7. 15. Rackoff PJ, Rosen CJ. Peripheral bone mass measurements. Journal of Clinical Densitometry 1998;1:287-94. 16. Blake GM, Patel R, Fogelman I. Peripheral or axial bone density measurements? Journal of Clinical Densitometry 1998;1:55-63. 17. Glher C, Jergas M, Hans D. Peripheral measurement techniques for the assessment of osteoporosis. Semin Nucl Med 1997;28:229-47. 18. Faulkner KG. Bone densitometry: choosing the proper skeletal site to measure. Journal of Clinical Densitometry 1997;1:279-85. 19. Freedman KB, Kaplan FS, Bilker WB, Strom BL, Lowe RA. Treatment of osteoporosis: are physicians missing an opportunity? J Bone Joint Surg 2000;82:1063-70. 20. Campbell MK, Torgerson DJ, Thomas RE, McClure JD, Reid DM. Direct disclosure of bone density results to patients: effect on knowledge of osteoporosis risk and anxiety level. Osteoporos Int 1998:8:584-90. 21. World Health Organization. Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: report of WHO Study Group. Technical report series 843.Geneva, Switzerland: World Health Organization, 1994. 22. Rackoff PJ, Rosen CJ. Peripheral bone mass measurement, current and future perspective on quantitative ultrasound and peripheral dexa. Journal of Clinical Densitomertry 1998;3:287-94. 23. Davis D. Does CME work? An analysis of the effect of educational activities on physician performance or health care outcomes. Int J Psychiatry Med 1998;29(1):21-39. 24. Walsh JME, McPhee SJ. A systems model of clinical preventive care: an analysis of factors influencing patient and physician. Health Educ Q 1992;19(2):157-75. 25. South-Paul JE. Osteoporosis: part I. Evaluation and assessment. Am Fam Physician 2001;63:897-908. 26. South-Paul JE. Osteoporosis: part II. Nonpharmacologic and pharmacologic treatment. Am Fam Physician 2001;63:1121-8.