- Email: [email protected]
Medical care costs of Paget's disease of bone in a privately insured population
Bone 38 (2006) 731 – 737 www.elsevier.com/locate/bone
Medical care costs of Paget's disease of bone in a privately insured population Becky A. Briesacher a,⁎, Denise Orwig b , Margaret Seton c , Mohamed Omar d , Kristijan H. Kahler d a
University of Massachusetts Medical School, Division of Geriatric Medicine, Biotech Four, Suite 315, 377 Plantation Street, Worcester, MA 01605, USA b University of Maryland School of Medicine, Baltimore, MD 21201, USA c Massachusetts General Hospital, MA 02129, USA d Novartis Pharmaceuticals Corporation, East Hanover, NJ 07936, USA Received 16 August 2005; revised 22 September 2005; accepted 11 October 2005 Available online 20 December 2005
Abstract Introduction: Medical care costs are difficult to calculate in diseases such as Paget's disease because they have low detection rates and a wide range of clinical manifestations that commonly occur in aging patient populations. Materials and methods: Using 2001–2002 MarketScan Research databases, this study linked medical claims, prescription records, and encounter data on 2.8 million active and retired employees to create a longitudinal panel with 24 months of observation. Patients with Paget's disease were identified by ICD-9 code 731.0. Matched controls (MC) were identified through an exact match procedure using gender, age, and predicted Medicare costs estimated with a risk adjuster. Diagnostic and expenditure records were extracted for the sample and prevalence rates calculated for 20 conditions with well-documented associations to Paget's disease. Comorbidities and health care costs of Paget's disease patients were compared to those of the MCs, and the differences tested using Chi-square and t tests. Results: Our study identified 244 matched pairs. The average age was 72.7 years; 50.8% were female. Significantly higher comorbidities (P b 0.05) were detected in Paget's disease patients relative to MCs for: pathological fractures (4.9% vs. 0.4%), heart murmurs (3.3% vs. 0.4%), low back pain (19.7% vs. 8.6%), spinal stenosis (16.4% vs. 9.8%), and hearing loss (13.5% vs. 5.7%), respectively. Biannual per patient outpatient costs were significantly higher in Paget's disease patients (Paget's disease $9301 vs. MC $6339, P b 0.05), especially for services associated with physician visits and diagnostic tests. Prescription costs for antiresportive agents and analgesics were also higher (Paget's disease $1115 vs. MC $507, P b 0.05). Inpatient costs (Paget's disease $16,144 vs. MC $21,480) were comparable. Conclusion: This study is the first to describe the excessive costs of Paget's disease, based on known patterns of disease expression, evaluation, and treatment. © 2005 Elsevier Inc. All rights reserved. Keywords: Reelin; Association; LD; PDT; Autism
Introduction There is no reliable information on the treatment costs of Paget's disease of bone, even though it affects 1–3% of the US ⁎ Corresponding author. Fax: +1 508 856 5024. E-mail addresses: [email protected] (B.A. Briesacher), [email protected] (D. Orwig), [email protected] (M. Seton), [email protected] (M. Omar), [email protected] (K.H. Kahler). 8756-3282/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.10.015
population aged 50 and older and is the second most common bone disorder after osteoporosis [3,29,34]. Paget's disease has eluded estimates of medical care costs because of low detection rates, unclear indicators of onset, long periods of asymptomatic presentation, and clinical manifestations that mimic other common diseases [18,24,26,34]. Diseases with these features confound standard economic analyses because of inadequate sample sizes, poor ability to distinguish incidence from prevalence cases, and no clear comparison group for calculating attributable medical costs [10,12,25]. Furthermore, patients
732
B.A. Briesacher et al. / Bone 38 (2006) 731–737
with Paget's disease have multiple comorbidities that are common in other aging populations: approximately 64% of patients with Paget's disease report having arthritis, heart problems or hypertension [13,14,16]. The clinical evidence on Paget's disease suggests that the true economic burden to individuals may be substantial. Epidemiologic surveys have documented increased risks with Paget's disease that include bone and joint pain, nerve compression syndromes, deafness, degenerative arthritis, spinal stenosis, fractures, and, in rare cases, sarcomatous degeneration of pagetic bone [20,22,27,34]. Radiographic studies have determined that the excessive bone turnover in pagetic bone creates mechanical vulnerabilities that may lead to deformity and fracture, with the most affected bone sites including the pelvis, femur, lumbar spine, tibia, skull, and humerus [8,15,18,26,28,32,34]. Disease management guidelines for Paget's disease suggest a pattern of care centered on the initial evaluation of the patient and then a wide range of medical treatments and services allocated according to the symptoms and disease activity [7,20,29]. Treatments for Paget's disease may include detection through a battery of radiological or biochemical tests, drug therapies to suppress bone turnover and relieve symptoms, surgery for patients with fractures or persistent symptoms, and life-long monitoring of bone activity. The objective of this study was to evaluate the medical care costs of Paget's disease. Our procedures offer several features that explicitly address the complexities of Paget's disease, and it is the first to provide estimates that account for comorbidities. Two studies have previously examined Paget's disease-related costs, but one derived hypothetical costs based on patterns of care outside the US health care system [19], and the other examined only the costs of bone active medications [9]. Our study used a data set with over 2 million people to identify a sufficient number of cases with clinically diagnosed Paget's disease. The study definitions came from the epidemiological and clinical literature so the costs relate to recognized manifestations of Paget's disease and known treatment modalities (e.g., radiographs, laboratory tests, bisphosphonates). Lastly, this study used risk adjustment methodology to identify a matched comparator group to minimize the influence of comorbidities. Methods This study used the 2001 and 2002 MarketScan Research databases (MEDSTAT, Ann Arbor, MI). These are secondary data sets of medical care claims, prescription drug claims, and encounter data gathered for research purposes from approximately 45 large U.S. employers, health plans, government agencies, and public organizations. Each year of the data set contains medical care information on 3 to 6 million individuals, and over 40 peer-reviewed journal articles have featured findings with these data [17,21]. The encounter files contain age, sex, geographic residence, and eligibility information. The prescription claims include the national drug codes, date of purchase, quantity, days' supply, and expenditure information. The medical claims contain payment information, diagnoses, procedure codes, and type of provider. For this analysis, we linked together two annual files to create a longitudinal panel with 24 months of inpatient, outpatient, and prescription drug utilization observations on 2.8 million people.
We identified the Paget's disease patients by examining all outpatient and inpatient claims records for primary or secondary diagnoses of Paget's disease in each year of observation (International Classification of Diseases, Ninth Revision (ICD-9) code 731.0) for a minimum of two claims over 2 years as a recommended procedure with administrative data [6]. The comparator group was identified through an exact match procedure using gender, age and comorbidity risk adjustment score. We calculated the comorbidity risk score using the Diagnostic Cost Group Hierarchical Condition Category (DCG/HCC) classification system (DxCG, Boston, MA) [5]. The DCG/HCC risk adjuster creates a single score for each person of predicted Medicare costs based on the presence of 189 medical conditions in the diagnosis fields of claims records. (The medical conditions in the risk adjuster are chosen for reliability and predictive power and do not include Paget's disease.) For our study, each Paget's disease patient was matched to an individual without Paget's disease but who has the same age, gender and the closest risk score.
Table 1 Select characteristics of study population, 2001–2002 Personal characteristics
Age, years b45 45–54 55–64 65–74 75–84 85+ Gender Male Female
Patients with PD
Matched controls
n
n
4 17 49 46 93 35
% 1.6 6.9 20.0 18.8 38.1 14.3
120 49.2 124 50.8
Potential complications of Paget's disease Musculoskeletal complications Spinal fracture 7 2.87 Vertebral compression 5 2.05 Hip fracture (in femur or 12 4.92 femoral neck) Pelvis fracture 4 1.64 Pathological fracture 12 4.92 Degenerative diseases Spinal stenosis: cervical, 40 16.39 thoracic, lumbar Low back pain 48 19.67 Osteoarthritis of the hip or 20 8.2 knee Arthritis 83 34.02 Pain in joint 91 37.3 Heart murmur 8 3.28 Neurological complications Hearing loss 33 13.52 Lumbar radiculitis 11 4.51 Nerve root compression 2 0.82 Posterior fossa 0 – compression Spinal cord compression 0 – Metabolic complications/ endocrinopathy Hyperparathyroidism 4 1.64 Hypercalcemia 4 1.64 Kidney stones 9 3.69 Hyperuricemia and gout 14 5.74 Source: 2001–2002 MarketScan.
Relative rates (PD/ MC)
% 1.6 6.9 20.0 18.8 38.1 14.3
– – – – – –
120 49.2 124 50.8
– –
4 17 49 46 93 35
4 1 8
1.64 0.41 3.28
1.75 (0.5–5.90) 5.00 (0.59–42.49) 1.50 (0.62–3.60)
2 1
0.82 2.00 (0.3 7–10.82) 0.41 12.00 a (1.57–91.58)
24
9.84
1.67 a (1.04–2.68)
21 10
8.61 4.1
2.28 a (1.04–2.68) 2.00 (1.41–3.69)
47 19.26 60 24.59 1 0.41 14 6 0 0 1
3 2 5 14
1.77 a (0.96–4.18) 1.52 a (1.15–1.99) 8.00 a (1.01–63.48)
5.74 2.36 a (2.39–5.39) 2.46 1.83 (0.69–4.88) – – – – 0.41 –
1.23 0.82 2.05 5.74
1.33 (0.30–5.89) 2.00 (0.37–10.82) 1.80 (0.61–5.29) 1.00 (0.49–2.05)
B.A. Briesacher et al. / Bone 38 (2006) 731–737 The analysis extracted all diagnostic and expenditure records for the matched sample and calculated the prevalence of 20 conditions with document links to Paget's disease and their related costs over the 2-year period. The conditions were identified from previous examinations of the literature detailing the clinical manifestations of Paget's disease [2,11,30] and operationalized as ICD-9 codes, Current Procedural Terminology (CPT) codes, and standardized service category codes in the database. (See Appendix A for a list of codes.) Complications selected in this study included fracture, low back pain and degenerative arthritis, acquired deformity, and neurological compression syndromes. We elected to exclude metastases and osteosarcomatous degeneration of pagetic bone as the sample size was too small to reliably detect rare disorders. To calculate the treatment costs, we identified relevant categories of medical services from published reviews of diagnosing and monitoring Paget's disease [3,4,31,33]. In several cases, we also included routine procedures such as X-rays of the skull, distal femur, and proximal tibia as other prevalent sites for monitoring the disease [29]. Cost items were excluded if they related to treatments for “pagetoid” lesions or “Paget's Disease of the Breast”, which we identified through ICD-9 codes and free text searches. We also limited the prescription drug costs to include only medications commonly used to treat Paget's disease or Paget's disease symptoms [30,34]. These included any use of antiresorptive agents (risedronate sodium, alendronate sodium, calcitonin, tiludronate disodium, and etidronate disodium) and analgesics. After identifying the itemized cost categories potentially related to Paget's disease, we summed all out-of-pocket and third party costs for three major types of costs: prescription drugs, outpatient services and inpatient services. For the statistical analysis, we compared the prevalence rates of conditions clinically related to Paget's disease complications and the health care costs between the two groups and tested the differences using Chi-squares and t tests at the P b 0.05 level.
Results We identified 244 individuals treated for Paget's disease and 244 matched controls (MC). Table 1 shows the distribution of age and gender in our study population and the prevalence of complicating conditions potentially associated with Paget's disease. Approximately 71.2% of the study population are over the age of 65, and 50.8% are female. Rates of complications potentially related to Paget's disease ranged from 37.3% to 0% in the Paget's disease group and 24.6% to 0% in the MC group. Overall, in the Paget's disease sample, approximately 62.3% experienced degenerative diseases, 16.8% neurological complications, 11.5% metabolic complications, and 9.0% musculoskeletal complications (summary data not shown). For specific conditions, Paget's
733
disease patients had significantly higher rates than MCs of: fractures, especially pathological (4.9% vs. 0.4%, P b 0.05); heart murmurs (3.3% vs. 0.4%, P b 0.05); hearing loss (13.5% vs. 5.7%, P b 0.05); spinal stenosis (16.4% vs. 9.8%, P b 0.05); and lower back pain (19.7% vs. 8.6%, P b 0.05). The analysis detected no significant differences in metabolic complications as defined in this study. Fig. 1 compares the treatment costs of inpatient care, outpatient services, and prescription drugs for Paget's disease patients and MCs during 2001 to 2002. The bars show that treatment costs were similar for both groups: the 244 Paget's disease patients spent $3.9 million compared to the 244 MCs who spent $3.3 million. However, the allocation of those expenditures is quite different: most of the medical care spending for Paget's disease patients occurred in the outpatient setting (58.0%) followed by inpatient setting (36.3%) and prescription drugs (5.6%). In comparison, most of the health care expenditures for MC patients were for inpatient services (51.0%) followed by outpatient (46.5%) and lastly prescription drugs (2.3%). The differences are statistically significant for outpatient costs (Paget's disease: $2.2 million, MC: 1.5 million, P b 0.05) and prescription drugs (Paget's disease: $220,831, MC: $79,082, P b 0.05). Inpatient costs were lower for Paget's disease patients than MCs ($1.4 million vs. $1.6 million, NS), although the difference is not statistically different. Tables 2, 3, and 4 provide details on the types of medical service represented in Fig. 1, both in terms of the proportion of patients using the services and the average service costs. Table 2 shows that Paget's disease patients incur higher outpatient costs than MCs primarily because they use these services more often rather than pay higher prices per service. Paget's disease patients were more likely than MCs to receive care from nearly every category of outpatient care, and most of the differences (19 of 27 service types) were statistically significant. For instance, over 82.7% of the Paget's disease group had an outpatient hospital visit compared to 73.7% of the MCs; over 77.9% of Paget's disease patients had outpatient surgery compared to 68.0% of MCs; and 51.2% of Paget's disease patients saw a radiologist compared to 44.7% of MCs. Paget's disease patients were 12 times more likely to undergo diagnostic testing: 29.1% of Paget's disease patients underwent imaging/nuclear medicine
Fig. 1. Comparison of medical care costs, 2001–2002 (n = 488).
734
B.A. Briesacher et al. / Bone 38 (2006) 731–737
Table 2 Average outpatient costs and use of services, 2001–2002 Patients with Paget's disease Matched controls (n = 244) (n = 244) Proportion Mean using services costs (%) ($) Total outpatient 100.0 Outpatient 82.8 a hospital visits Physician visits IM/Family 52.0 medicine Radiology 51.2 a Physical 18.9 therapy Cardiology 19.7 Podiatry 21.3 a Chiropractor 8.6 Endocrinology 9.0 a Rheumatology 7.8 a Other 97.5 a unspecified Outpatient 77.9 a surgery Orthopedic 17.6 surgery Osteopathic 8.6 medicine Diagnostic testing Imaging/ 29.1 a Nuclear medicine X-rays Femur 7.0 a Knee 20.1 a Skull 7.0 a Pelvis 20.9 a Abdomen 8.2 Spine 23.0 a (lumbosacral) Other 89.8 a diagnostic Radiology Laboratory tests Alkaline 17.2 a phosphatase Cross links 7.0 a Calcium 6.1 a screen in blood Other 74.2 a laboratory Diagnostic 54.1 a cardiology Durable 21.3 medical equipment
SE
Proportion Mean using services costs (%) ($)
9301 a 4382
891 100.0 792 73.8
1009 a
116
454 942
SE
6339 3273
578 446
52.0
733
80
57 169
44.7 18.9
440 867
83 125
635 297 a 532 403 a 465 1422 a
112 51 160 103 101 118
18.0 14.8 10.7 0.4 1.6 93.0
876 195 632 141 488 885
234 27 159 – 231 70
1362
160
68.0
1203
171
564
116
16.4
430
87
362
64
4.9
1435
848
232
35
2.5
253
124
a
significantly higher for the following physician visits: IM = internal medicine/family medicine ($1009 vs. $733, P b 0.05), podiatry ($297 vs. $195, P b 0.1), endocrinology ($403 vs. $141, P b 0.05), and other unspecified ($1422 vs. $885, P b 0.05). Average outpatient diagnostic costs were significantly higher for: X-rays, femur, or knee ($74–$58 vs. $36–$34, P b 0.05); other diagnostic radiology ($704 vs. $446, P b 0.05); and alkaline phosphatase tests ($44 vs. $10, P b 0.05). Table 3 shows a similar pattern of Paget's disease patients paying more for prescription drugs than MCs because of higher use of these medications: over 20% of the Paget's disease group took antiresorptive agents compared to less than 5% of MCs. Most notably, Paget's disease patients more often received nonsteroidal anti-inflammatory drugs (46.7% vs. 38.1%, P b 0.05), salicylates (3.7% vs. 0.8%, P b 0.05), and opiate agonists (50.4% vs. 43.4%, P b 0.05) than MCs. Average prescription drug costs were significantly higher for only risedronate sodium ($1038 vs. $300, P b 0.05) and etidronate disodium ($392 vs. $68, P b 0.05). The average analgesic medication costs were similar for both groups. Table 4 compares details of the inpatient spending for Paget's disease patients relative to that for MCs. For all types of inpatient services, the Paget's disease and MC groups were as likely to use the care, and both groups paid similar average costs except in two cases. For instance, Paget's disease patients were as likely to be hospitalized as the MCs (30.7% vs. 32.4%), and their average inpatient hospitalization costs Table 3 Average prescription drug costs and use, 2001–2002
–
74 58 a 51 a 53 33 76
31 8 13 9 5 10
0.4 9.8 0.4 5.3 11.9 10.7
34 36 78 42 44 67
704 a
72
82.0
446
60
44 a
12
2.9
10
2
141 a 15
70 5
0.0 2.0
0 8
634
120
61.1
577
147
448
82
46.7
450
81
1136
299
16.4
1042
313
5 – 9 8 12
– 3
Source: 2001–2002 MarketScan. a P b 0.05.
procedures compared to 2.5% of MCs. In contrast, Paget's disease patients paid higher average prices than MCs for only 10 types of outpatient care. Average outpatient costs were
Patients with Paget's disease (n = 244)
Matched controls (n = 244)
Proportion Mean using Rx (%) costs ($)
Proportion Mean using Rx (%) costs ($)
Total 81.1 a prescription drug Antiresorptive agents Risedronate 13.9 a sodium Alendronate 21.7 a sodium Calcitonin 9.8 a (salmon) Tiludronate 0.4 disodium Etidronate 1.6 a disodium Analgesics Antipyretic, 46.7 a NSAID Antipyretic, 3.7 a salicylates Antipyretic/ 9.4 a NEC Opiate 50.4 a agonists
1115 a
SE
$165 63.9
SE
507
$67
1038 a
153
1.2
300
124
637
69
4.9
514
79
755
225
3.7
423
105
0.0
0
–
0.4
68
–
93 38.1
539
68
2378
Source: 2001–2002 MarketScan. a P b 0.05.
392 a
659
– 157
225
123
0.8
605
531
221
94
3.3
103
52
216 43.4
151
56
386
B.A. Briesacher et al. / Bone 38 (2006) 731–737 Table 4 Average inpatient costs and use of services, 2001–2002 Patients with Paget's disease Matched controls (n = 244) (n = 244)
Total inpatient Inpatient hospitalization Surgery Physician services Cardiology Radiology Other unspecified Diagnostic laboratory Diagnostic radiology Diagnostic cardiology Orthopedic surgery Hip replacement Knee replacement
Proportion using services (%)
Mean costs ($)
SE
Proportion using services (%)
36.1 30.7
16,144 3404 32.4 15,781 3526 29.1
Mean costs ($)
SE
21,480 4324 19,890 3814
18.9
2013
438 18.9
1868
206
5.3 10.7 27.9
1348 352 775
503 5.3 203 11.1 100 22.1
406 601 930
194 266 176
8.2
1064
595
20.1 12.3
4.1
1769 1006
166
62 17.6
1721 1404
563
196 10.7
298
113 282
2.9
934 a 585
3.7
2345
1.6 0.8
1750 a 88 2658 898
0.4 0.8
1955 – 2318 784
Source: 2001–2002 MarketScan. a P b 0.05.
were lower ($15,781 vs. $19,890, NS), although the difference was not statistically significant. Average inpatient costs were significantly lower (P b 0.05) for Paget's diseases than MCs for orthopedic surgery ($934 vs. $2345) and hip replacement ($1750 vs. $1955). Other average inpatient costs were similar. Discussion To our knowledge, this study is the first to consider a wide range of medical care costs for Paget's disease including inpatient, outpatient, and drug therapies, based on known patterns of disease expression, evaluation and treatment. This study is also the first to provide cost estimates from the actual payments made by the patients and their employer-based insurance health plans and to use risk adjustment methodology and a matched control group in assessing excess costs. In concurrence with other epidemiological surveys, our study detected significant disease burden due to the known complications of Paget's disease. For instance, the Paget's disease patients in our study were at increased risk of suffering fractures compared to MCs (RR 2.0), [1,23]. Low back pain was also significant in the Paget's disease cohort with 20% having this diagnosis and a relative rate of 2.28 compared to MCs. Specifically, patients with Paget's disease were more often identified as suffering from spinal stenosis (increased risk of 67%) and arthritis at any site compared to controls (RR = 1.77). The consistent finding of these
735
comorbidity rates that are in accordance with the clinical literature helps in establishing the credibility of our approach. In context of the complication rates, we found that the 244 Paget's disease patients in this study spent $588,280 more than the 244 MCs on medical care between 2001 and 2002, which is a 17.7% increase. Most of the difference came from outpatient care services, where patients with Paget's disease spent 46.7% more ($722,742) than the MCs, primarily because of higher rates of physician visits, surgery, and diagnostic testing. During the same period, Paget's disease patients also spent $141,749 more than MCs on antiresorptive agents and analgesics, which was also a significant difference. Lastly, patients with Paget's disease spent $276,211 less than MCs on inpatient care, although this difference is not significant. Harder to interpret was the increased use of some general medical services as in physician visits in the outpatient setting. Clearly, the use of specialists in rheumatology and endocrinology reached significance in this cohort of Paget's disease patients compared with matched controls, as did podiatry and radiology services. However, the increased use of unspecified physician services (97.5% vs. 93.0%) and outpatient surgery (77.9% vs. 68.0%) is not interpretable. These kinds of discrepancies highlight both the strengths and weaknesses of this study. The application of knowledge about the clinical expression and treatment of Paget's disease helped in isolating some of the most relevant cost categories, but it could not elucidate why use was so high of physician services without explicit details. Although assessing appropriateness of treatment was not an objective of this study, it is interesting to note that less than a third of our cases with Paget's disease received any antiresportive therapy but over 70% took analgesics. Our analysis has several clear limitations. The first is the low ascertainment of individuals with Paget's disease. In general, low sensitivity but high specificity is a common trait of research with administrative data, and this is especially true for conditions like Paget's disease that are commonly undetected [6]. Radiographic studies estimate that only 7% of patients with the disorder reach a clinical diagnosis, and this likely explains our low detection rate [34]. Our case selection criterion of at least 2 claims with Paget's disease diagnoses over 2 years minimized “rule out” individuals who were tested for but did not have the disease, but it decreased statistical power. We identified 1021 individuals with Paget's disease in 2001 and 1416 in 2002, but only 244 in both years. A related issue is the cross-sectional study design of the study which prevented us from distinguishing between prevalent and incident Paget's disease. Based on the observed patterns of diagnostic testing and antiresorptive prescriptions, perhaps a third of our cases were newly diagnosed, and they may have distinct costs from the others. Another limitation was our focus on conditions and categories of medical services with documented links to Paget's disease. This prevented the identification of health care costs and comorbidities that have not yet been recognized. Lastly, it
736
B.A. Briesacher et al. / Bone 38 (2006) 731–737
is also important to note that these results may not be generalizable to other patient groups, especially those who do not have employer-sponsored health insurance and prescription drug coverage. Conclusion This analysis provides new insight into the disease burden of Paget's disease and the economic consequences. By including a broad range of medical care utilization costs and using a matched control group, we were able to describe the excess burden of Paget's disease in a patient population with private insurance. Our study found evidence of significant disease burden with respect to known complications, even after accounting for comorbidities. These findings suggest a high economic burden for Paget's disease patients, especially for outpatient services, diagnostic testing, analgesics, and antiresportive treatments. Future research should assess the appropriateness of this care, as well as the longitudinal impact of therapies that may slow the progression of Paget's disease and prevent the associated complications and excess medical care costs. Acknowledgments Funding source: this study was funded in part by an unrestricted research grant from Novartis Pharmaceuticals Corporation, East Hanover, NJ. Appendix List of ICD-9 and CPT codes used for identifying complications potentially related to Paget's disease: fractures: spine, no cord injury (excl. fragility) 805.0–805.9; spine with cord injury 806.0–806.9; vertebral compression 733.13; other vertebral 805.xx, 806.xx; thoracic 806.2, 806.3; lumbar 806.4, 806.5; pelvis 808.0–808.9; femur 820.0–820.9; other parts of femur 821.0–821.3; pathological fracture: 733.10–733.19; osteoarthritis of hip or knee: 715.15, 715.25, 715.35, 715.95; hip replacement: 27130–27132; knee replacement: 27440– 27445, 27447; hearing Loss: 389.xx; heart murmur: 785.2; kidney stones: 592.0; spinal cord complications: lumbar radiculitis 724.4; spinal stenosis 724.00; cervical 723.0– 723.1; lumbar 724.02; thoracic 724.01, 724.1; posterior fossa compression: 348.4; spinal cord compression: 336.9; nerve root compression: 724.9; generic low back pain: 724.2; pain in joint: 719.4; hyperparathyroid disease: 252.0 hyperuricemia: 790.6; gout: 274.9; arthritis: 274.0; hypercalcemia 275.42. References [1] Ahmed F, Janes GR, Baron R, Latts LM. Preferred provider organization claims showed high predictive value but missed substantial proportion of adults with high-risk conditions. J Clin Epidemiol 2005;58:624–8. [2] Altman RD. Musculoskeletal manifestations of Paget's disease of bone. Arthritis Rheum 1980;23:1121–7.
[3] Altman RD, Bloch DA, Hochberg MC, Murphy WA. Prevalence of pelvic Paget's disease of bone in the United States. J Bone Miner Res 2000; 15:461–5. [4] Ankrom MA, Shapiro JR. Paget's disease of bone (osteitis deformans). J Am Geriatr Soc 1998;46:1025–33. [5] Ash AS, Ellis RP, Pope GC, Ayanian JZ, Bates DW, Burstin H, et al. Using diagnoses to describe populations and predict costs. Health Care Financ Rev 2000;21:7–28. [6] Borzecki AM, Wong AT, Hickey EC, Ash AS, Berlowitz DR. Identifying hypertension-related comorbidities from administrative data: what's the optimal approach? Am J Med Qual 2004;19:201–6. [7] Delmas PD, Meunier PJ. The management of Paget's disease of bone. N Engl J Med 1997;336:558–66. [8] Dove J. Complete fractures of the femur in Paget's disease of bone. J Bone Jt Surg, Br 1980;62-B:12–7. [9] Drake WM, Kendler DL, Brown JP. Consensus statement on the modern therapy of Paget's disease of bone from a Western Osteoporosis Alliance symposium. Biannual Foothills Meeting on Osteoporosis, Calgary, Alberta, Canada, September 9–10, 2000. Clin Ther 2001; 23:620–6. [10] Etzioni R, Riley GF, Ramsey SD, Brown M. Measuring costs: administrative claims data, clinical trials, and beyond. Med Care 2002;40: III63–72. [11] Franck WA, Bress NM, Singer FR, Krane SM. Rheumatic manifestations of Paget's disease of bone. Am J Med 1974;56:592–603. [12] Gabriel SE, Tosteson AN, Leibson CL, Crowson CS, Pond GR, Hammond CS, et al. Direct medical costs attributable to osteoporotic fractures. Osteoporos Int 2002;13:323–30. [13] Gold DT, Boisture J, Shipp KM, Pieper CF, Lyles KW. Paget's disease of bone and quality of life. J Bone Miner Res 1996;11:1897–904. [14] Gold DT, Burchett BM, Shipp KM, Pieper CF, Lyles KW. Factors associated with self-rated health in patients with Paget's disease of bone. J Bone Miner Res 1999;14(Suppl 2):99–102. [15] Grundy M. Fractures of the femur in Paget's disease of bone. Their etiology and treatment. J Bone Jt Surg, Br 1970;52:252–63. [16] Hamdy RC, Moore S, LeRoy J. Clinical presentation of Paget's disease of the bone in older patients. South Med J 1993;86:1097–100. [17] Hu XH, Markson LE, Lipton RB, Stewart WF, Berger ML. Burden of migraine in the United States: disability and economic costs. Arch Intern Med 1999;159:813–8. [18] Keen RW. The current status of Paget's disease of the bone. Hosp Med 2003;64:230–2. [19] Lafuma A, Fagnani F, Meunier PJ. An economic evaluation of tiludronic acid treatment in Paget's disease of bone. Pharmacoeconomics 1997;12:460–74. [20] Langston AL, Ralston SH. Management of Paget's disease of bone. Rheumatology (Oxford) 2004;43:955–9. [21] Mark TL, Coffey RM. What drove private health insurance spending on mental health and substance abuse care, 1992–1999? Health Aff (Millwood) 2003;22:165–72. [22] Melton III LJ, Tiegs RD, Atkinson EJ, O'Fallon WM. Fracture risk among patients with Paget's disease: a population-based cohort study. J Bone Miner Res 2000;15:2123–8. [23] Menzin J, Lang K, Friedman M, Neumann P, Cummings JL. The economic cost of Alzheimer's disease and related dementias to the California Medicaid program (“Medi-Cal”) in 1995. Am J Geriatr Psychiatry 1999; 7:300–8. [24] Mirra JM, Brien EW, Tehranzadeh J. Paget's disease of bone: review with emphasis on radiologic features: Part II. Skelet Radiol 1995;24:173–84. [25] Newton KM, Wagner EH, Ramsey SD, McCulloch D, Evans R, Sandhu N, et al. The use of automated data to identify complications and comorbidities of diabetes: a validation study. J Clin Epidemiol 1999;52:199–207. [26] Ooi CG, Fraser WD. Paget's disease of bone. Postgrad Med J 1997;73: 69–74. [27] Saifuddin A, Hassan A. Paget's disease of the spine: unusual features and complications. Clin Radiol 2003;58:102–11. [28] Schneider D, Hofmann MT, Peterson JA. Diagnosis and treatment of Paget's disease of bone. Am Fam Phys 2002;65:2069–72.
B.A. Briesacher et al. / Bone 38 (2006) 731–737 [29] Selby PL, Davie MW, Ralston SH, Stone MD. Guidelines on the management of Paget's disease of bone. Bone 2002;31: 366–73. [30] Seton M, Choi HK, Hansen MF, Sebaldt RJ, Cooper C. Analysis of environmental factors in familial versus sporadic Paget's disease of bone—The New England Registry for Paget's Disease of Bone. J Bone Miner Res 2003;18:1519–24. [31] Seton M, Krane SM. Paget's disease of bone. Curr Ther Endocrinol Metab 1997;6:568–72.
737
[32] Tiegs RD. Paget's disease of bone: indications for treatment and goals of therapy. Clin Ther 1997;19:1309–29 [discussion 1523–1524]. [33] Tiegs RD, Lohse CM, Wollan PC, Melton LJ. Long-term trends in the incidence of Paget's disease of bone. Bone 2000;27: 423–7. [34] van Staa TP, Selby P, Leufkens HG, Lyles K, Sprafka JM, Cooper C. Incidence and natural history of Paget's disease of bone in England and Wales. J Bone Miner Res 2002;17:465–71.
Medical care costs of Paget's disease of bone in a privately insured population Becky A. Briesacher a,⁎, Denise Orwig b , Margaret Seton c , Mohamed Omar d , Kristijan H. Kahler d a
University of Massachusetts Medical School, Division of Geriatric Medicine, Biotech Four, Suite 315, 377 Plantation Street, Worcester, MA 01605, USA b University of Maryland School of Medicine, Baltimore, MD 21201, USA c Massachusetts General Hospital, MA 02129, USA d Novartis Pharmaceuticals Corporation, East Hanover, NJ 07936, USA Received 16 August 2005; revised 22 September 2005; accepted 11 October 2005 Available online 20 December 2005
Abstract Introduction: Medical care costs are difficult to calculate in diseases such as Paget's disease because they have low detection rates and a wide range of clinical manifestations that commonly occur in aging patient populations. Materials and methods: Using 2001–2002 MarketScan Research databases, this study linked medical claims, prescription records, and encounter data on 2.8 million active and retired employees to create a longitudinal panel with 24 months of observation. Patients with Paget's disease were identified by ICD-9 code 731.0. Matched controls (MC) were identified through an exact match procedure using gender, age, and predicted Medicare costs estimated with a risk adjuster. Diagnostic and expenditure records were extracted for the sample and prevalence rates calculated for 20 conditions with well-documented associations to Paget's disease. Comorbidities and health care costs of Paget's disease patients were compared to those of the MCs, and the differences tested using Chi-square and t tests. Results: Our study identified 244 matched pairs. The average age was 72.7 years; 50.8% were female. Significantly higher comorbidities (P b 0.05) were detected in Paget's disease patients relative to MCs for: pathological fractures (4.9% vs. 0.4%), heart murmurs (3.3% vs. 0.4%), low back pain (19.7% vs. 8.6%), spinal stenosis (16.4% vs. 9.8%), and hearing loss (13.5% vs. 5.7%), respectively. Biannual per patient outpatient costs were significantly higher in Paget's disease patients (Paget's disease $9301 vs. MC $6339, P b 0.05), especially for services associated with physician visits and diagnostic tests. Prescription costs for antiresportive agents and analgesics were also higher (Paget's disease $1115 vs. MC $507, P b 0.05). Inpatient costs (Paget's disease $16,144 vs. MC $21,480) were comparable. Conclusion: This study is the first to describe the excessive costs of Paget's disease, based on known patterns of disease expression, evaluation, and treatment. © 2005 Elsevier Inc. All rights reserved. Keywords: Reelin; Association; LD; PDT; Autism
Introduction There is no reliable information on the treatment costs of Paget's disease of bone, even though it affects 1–3% of the US ⁎ Corresponding author. Fax: +1 508 856 5024. E-mail addresses: [email protected] (B.A. Briesacher), [email protected] (D. Orwig), [email protected] (M. Seton), [email protected] (M. Omar), [email protected] (K.H. Kahler). 8756-3282/$ - see front matter © 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2005.10.015
population aged 50 and older and is the second most common bone disorder after osteoporosis [3,29,34]. Paget's disease has eluded estimates of medical care costs because of low detection rates, unclear indicators of onset, long periods of asymptomatic presentation, and clinical manifestations that mimic other common diseases [18,24,26,34]. Diseases with these features confound standard economic analyses because of inadequate sample sizes, poor ability to distinguish incidence from prevalence cases, and no clear comparison group for calculating attributable medical costs [10,12,25]. Furthermore, patients
732
B.A. Briesacher et al. / Bone 38 (2006) 731–737
with Paget's disease have multiple comorbidities that are common in other aging populations: approximately 64% of patients with Paget's disease report having arthritis, heart problems or hypertension [13,14,16]. The clinical evidence on Paget's disease suggests that the true economic burden to individuals may be substantial. Epidemiologic surveys have documented increased risks with Paget's disease that include bone and joint pain, nerve compression syndromes, deafness, degenerative arthritis, spinal stenosis, fractures, and, in rare cases, sarcomatous degeneration of pagetic bone [20,22,27,34]. Radiographic studies have determined that the excessive bone turnover in pagetic bone creates mechanical vulnerabilities that may lead to deformity and fracture, with the most affected bone sites including the pelvis, femur, lumbar spine, tibia, skull, and humerus [8,15,18,26,28,32,34]. Disease management guidelines for Paget's disease suggest a pattern of care centered on the initial evaluation of the patient and then a wide range of medical treatments and services allocated according to the symptoms and disease activity [7,20,29]. Treatments for Paget's disease may include detection through a battery of radiological or biochemical tests, drug therapies to suppress bone turnover and relieve symptoms, surgery for patients with fractures or persistent symptoms, and life-long monitoring of bone activity. The objective of this study was to evaluate the medical care costs of Paget's disease. Our procedures offer several features that explicitly address the complexities of Paget's disease, and it is the first to provide estimates that account for comorbidities. Two studies have previously examined Paget's disease-related costs, but one derived hypothetical costs based on patterns of care outside the US health care system [19], and the other examined only the costs of bone active medications [9]. Our study used a data set with over 2 million people to identify a sufficient number of cases with clinically diagnosed Paget's disease. The study definitions came from the epidemiological and clinical literature so the costs relate to recognized manifestations of Paget's disease and known treatment modalities (e.g., radiographs, laboratory tests, bisphosphonates). Lastly, this study used risk adjustment methodology to identify a matched comparator group to minimize the influence of comorbidities. Methods This study used the 2001 and 2002 MarketScan Research databases (MEDSTAT, Ann Arbor, MI). These are secondary data sets of medical care claims, prescription drug claims, and encounter data gathered for research purposes from approximately 45 large U.S. employers, health plans, government agencies, and public organizations. Each year of the data set contains medical care information on 3 to 6 million individuals, and over 40 peer-reviewed journal articles have featured findings with these data [17,21]. The encounter files contain age, sex, geographic residence, and eligibility information. The prescription claims include the national drug codes, date of purchase, quantity, days' supply, and expenditure information. The medical claims contain payment information, diagnoses, procedure codes, and type of provider. For this analysis, we linked together two annual files to create a longitudinal panel with 24 months of inpatient, outpatient, and prescription drug utilization observations on 2.8 million people.
We identified the Paget's disease patients by examining all outpatient and inpatient claims records for primary or secondary diagnoses of Paget's disease in each year of observation (International Classification of Diseases, Ninth Revision (ICD-9) code 731.0) for a minimum of two claims over 2 years as a recommended procedure with administrative data [6]. The comparator group was identified through an exact match procedure using gender, age and comorbidity risk adjustment score. We calculated the comorbidity risk score using the Diagnostic Cost Group Hierarchical Condition Category (DCG/HCC) classification system (DxCG, Boston, MA) [5]. The DCG/HCC risk adjuster creates a single score for each person of predicted Medicare costs based on the presence of 189 medical conditions in the diagnosis fields of claims records. (The medical conditions in the risk adjuster are chosen for reliability and predictive power and do not include Paget's disease.) For our study, each Paget's disease patient was matched to an individual without Paget's disease but who has the same age, gender and the closest risk score.
Table 1 Select characteristics of study population, 2001–2002 Personal characteristics
Age, years b45 45–54 55–64 65–74 75–84 85+ Gender Male Female
Patients with PD
Matched controls
n
n
4 17 49 46 93 35
% 1.6 6.9 20.0 18.8 38.1 14.3
120 49.2 124 50.8
Potential complications of Paget's disease Musculoskeletal complications Spinal fracture 7 2.87 Vertebral compression 5 2.05 Hip fracture (in femur or 12 4.92 femoral neck) Pelvis fracture 4 1.64 Pathological fracture 12 4.92 Degenerative diseases Spinal stenosis: cervical, 40 16.39 thoracic, lumbar Low back pain 48 19.67 Osteoarthritis of the hip or 20 8.2 knee Arthritis 83 34.02 Pain in joint 91 37.3 Heart murmur 8 3.28 Neurological complications Hearing loss 33 13.52 Lumbar radiculitis 11 4.51 Nerve root compression 2 0.82 Posterior fossa 0 – compression Spinal cord compression 0 – Metabolic complications/ endocrinopathy Hyperparathyroidism 4 1.64 Hypercalcemia 4 1.64 Kidney stones 9 3.69 Hyperuricemia and gout 14 5.74 Source: 2001–2002 MarketScan.
Relative rates (PD/ MC)
% 1.6 6.9 20.0 18.8 38.1 14.3
– – – – – –
120 49.2 124 50.8
– –
4 17 49 46 93 35
4 1 8
1.64 0.41 3.28
1.75 (0.5–5.90) 5.00 (0.59–42.49) 1.50 (0.62–3.60)
2 1
0.82 2.00 (0.3 7–10.82) 0.41 12.00 a (1.57–91.58)
24
9.84
1.67 a (1.04–2.68)
21 10
8.61 4.1
2.28 a (1.04–2.68) 2.00 (1.41–3.69)
47 19.26 60 24.59 1 0.41 14 6 0 0 1
3 2 5 14
1.77 a (0.96–4.18) 1.52 a (1.15–1.99) 8.00 a (1.01–63.48)
5.74 2.36 a (2.39–5.39) 2.46 1.83 (0.69–4.88) – – – – 0.41 –
1.23 0.82 2.05 5.74
1.33 (0.30–5.89) 2.00 (0.37–10.82) 1.80 (0.61–5.29) 1.00 (0.49–2.05)
B.A. Briesacher et al. / Bone 38 (2006) 731–737 The analysis extracted all diagnostic and expenditure records for the matched sample and calculated the prevalence of 20 conditions with document links to Paget's disease and their related costs over the 2-year period. The conditions were identified from previous examinations of the literature detailing the clinical manifestations of Paget's disease [2,11,30] and operationalized as ICD-9 codes, Current Procedural Terminology (CPT) codes, and standardized service category codes in the database. (See Appendix A for a list of codes.) Complications selected in this study included fracture, low back pain and degenerative arthritis, acquired deformity, and neurological compression syndromes. We elected to exclude metastases and osteosarcomatous degeneration of pagetic bone as the sample size was too small to reliably detect rare disorders. To calculate the treatment costs, we identified relevant categories of medical services from published reviews of diagnosing and monitoring Paget's disease [3,4,31,33]. In several cases, we also included routine procedures such as X-rays of the skull, distal femur, and proximal tibia as other prevalent sites for monitoring the disease [29]. Cost items were excluded if they related to treatments for “pagetoid” lesions or “Paget's Disease of the Breast”, which we identified through ICD-9 codes and free text searches. We also limited the prescription drug costs to include only medications commonly used to treat Paget's disease or Paget's disease symptoms [30,34]. These included any use of antiresorptive agents (risedronate sodium, alendronate sodium, calcitonin, tiludronate disodium, and etidronate disodium) and analgesics. After identifying the itemized cost categories potentially related to Paget's disease, we summed all out-of-pocket and third party costs for three major types of costs: prescription drugs, outpatient services and inpatient services. For the statistical analysis, we compared the prevalence rates of conditions clinically related to Paget's disease complications and the health care costs between the two groups and tested the differences using Chi-squares and t tests at the P b 0.05 level.
Results We identified 244 individuals treated for Paget's disease and 244 matched controls (MC). Table 1 shows the distribution of age and gender in our study population and the prevalence of complicating conditions potentially associated with Paget's disease. Approximately 71.2% of the study population are over the age of 65, and 50.8% are female. Rates of complications potentially related to Paget's disease ranged from 37.3% to 0% in the Paget's disease group and 24.6% to 0% in the MC group. Overall, in the Paget's disease sample, approximately 62.3% experienced degenerative diseases, 16.8% neurological complications, 11.5% metabolic complications, and 9.0% musculoskeletal complications (summary data not shown). For specific conditions, Paget's
733
disease patients had significantly higher rates than MCs of: fractures, especially pathological (4.9% vs. 0.4%, P b 0.05); heart murmurs (3.3% vs. 0.4%, P b 0.05); hearing loss (13.5% vs. 5.7%, P b 0.05); spinal stenosis (16.4% vs. 9.8%, P b 0.05); and lower back pain (19.7% vs. 8.6%, P b 0.05). The analysis detected no significant differences in metabolic complications as defined in this study. Fig. 1 compares the treatment costs of inpatient care, outpatient services, and prescription drugs for Paget's disease patients and MCs during 2001 to 2002. The bars show that treatment costs were similar for both groups: the 244 Paget's disease patients spent $3.9 million compared to the 244 MCs who spent $3.3 million. However, the allocation of those expenditures is quite different: most of the medical care spending for Paget's disease patients occurred in the outpatient setting (58.0%) followed by inpatient setting (36.3%) and prescription drugs (5.6%). In comparison, most of the health care expenditures for MC patients were for inpatient services (51.0%) followed by outpatient (46.5%) and lastly prescription drugs (2.3%). The differences are statistically significant for outpatient costs (Paget's disease: $2.2 million, MC: 1.5 million, P b 0.05) and prescription drugs (Paget's disease: $220,831, MC: $79,082, P b 0.05). Inpatient costs were lower for Paget's disease patients than MCs ($1.4 million vs. $1.6 million, NS), although the difference is not statistically different. Tables 2, 3, and 4 provide details on the types of medical service represented in Fig. 1, both in terms of the proportion of patients using the services and the average service costs. Table 2 shows that Paget's disease patients incur higher outpatient costs than MCs primarily because they use these services more often rather than pay higher prices per service. Paget's disease patients were more likely than MCs to receive care from nearly every category of outpatient care, and most of the differences (19 of 27 service types) were statistically significant. For instance, over 82.7% of the Paget's disease group had an outpatient hospital visit compared to 73.7% of the MCs; over 77.9% of Paget's disease patients had outpatient surgery compared to 68.0% of MCs; and 51.2% of Paget's disease patients saw a radiologist compared to 44.7% of MCs. Paget's disease patients were 12 times more likely to undergo diagnostic testing: 29.1% of Paget's disease patients underwent imaging/nuclear medicine
Fig. 1. Comparison of medical care costs, 2001–2002 (n = 488).
734
B.A. Briesacher et al. / Bone 38 (2006) 731–737
Table 2 Average outpatient costs and use of services, 2001–2002 Patients with Paget's disease Matched controls (n = 244) (n = 244) Proportion Mean using services costs (%) ($) Total outpatient 100.0 Outpatient 82.8 a hospital visits Physician visits IM/Family 52.0 medicine Radiology 51.2 a Physical 18.9 therapy Cardiology 19.7 Podiatry 21.3 a Chiropractor 8.6 Endocrinology 9.0 a Rheumatology 7.8 a Other 97.5 a unspecified Outpatient 77.9 a surgery Orthopedic 17.6 surgery Osteopathic 8.6 medicine Diagnostic testing Imaging/ 29.1 a Nuclear medicine X-rays Femur 7.0 a Knee 20.1 a Skull 7.0 a Pelvis 20.9 a Abdomen 8.2 Spine 23.0 a (lumbosacral) Other 89.8 a diagnostic Radiology Laboratory tests Alkaline 17.2 a phosphatase Cross links 7.0 a Calcium 6.1 a screen in blood Other 74.2 a laboratory Diagnostic 54.1 a cardiology Durable 21.3 medical equipment
SE
Proportion Mean using services costs (%) ($)
9301 a 4382
891 100.0 792 73.8
1009 a
116
454 942
SE
6339 3273
578 446
52.0
733
80
57 169
44.7 18.9
440 867
83 125
635 297 a 532 403 a 465 1422 a
112 51 160 103 101 118
18.0 14.8 10.7 0.4 1.6 93.0
876 195 632 141 488 885
234 27 159 – 231 70
1362
160
68.0
1203
171
564
116
16.4
430
87
362
64
4.9
1435
848
232
35
2.5
253
124
a
significantly higher for the following physician visits: IM = internal medicine/family medicine ($1009 vs. $733, P b 0.05), podiatry ($297 vs. $195, P b 0.1), endocrinology ($403 vs. $141, P b 0.05), and other unspecified ($1422 vs. $885, P b 0.05). Average outpatient diagnostic costs were significantly higher for: X-rays, femur, or knee ($74–$58 vs. $36–$34, P b 0.05); other diagnostic radiology ($704 vs. $446, P b 0.05); and alkaline phosphatase tests ($44 vs. $10, P b 0.05). Table 3 shows a similar pattern of Paget's disease patients paying more for prescription drugs than MCs because of higher use of these medications: over 20% of the Paget's disease group took antiresorptive agents compared to less than 5% of MCs. Most notably, Paget's disease patients more often received nonsteroidal anti-inflammatory drugs (46.7% vs. 38.1%, P b 0.05), salicylates (3.7% vs. 0.8%, P b 0.05), and opiate agonists (50.4% vs. 43.4%, P b 0.05) than MCs. Average prescription drug costs were significantly higher for only risedronate sodium ($1038 vs. $300, P b 0.05) and etidronate disodium ($392 vs. $68, P b 0.05). The average analgesic medication costs were similar for both groups. Table 4 compares details of the inpatient spending for Paget's disease patients relative to that for MCs. For all types of inpatient services, the Paget's disease and MC groups were as likely to use the care, and both groups paid similar average costs except in two cases. For instance, Paget's disease patients were as likely to be hospitalized as the MCs (30.7% vs. 32.4%), and their average inpatient hospitalization costs Table 3 Average prescription drug costs and use, 2001–2002
–
74 58 a 51 a 53 33 76
31 8 13 9 5 10
0.4 9.8 0.4 5.3 11.9 10.7
34 36 78 42 44 67
704 a
72
82.0
446
60
44 a
12
2.9
10
2
141 a 15
70 5
0.0 2.0
0 8
634
120
61.1
577
147
448
82
46.7
450
81
1136
299
16.4
1042
313
5 – 9 8 12
– 3
Source: 2001–2002 MarketScan. a P b 0.05.
procedures compared to 2.5% of MCs. In contrast, Paget's disease patients paid higher average prices than MCs for only 10 types of outpatient care. Average outpatient costs were
Patients with Paget's disease (n = 244)
Matched controls (n = 244)
Proportion Mean using Rx (%) costs ($)
Proportion Mean using Rx (%) costs ($)
Total 81.1 a prescription drug Antiresorptive agents Risedronate 13.9 a sodium Alendronate 21.7 a sodium Calcitonin 9.8 a (salmon) Tiludronate 0.4 disodium Etidronate 1.6 a disodium Analgesics Antipyretic, 46.7 a NSAID Antipyretic, 3.7 a salicylates Antipyretic/ 9.4 a NEC Opiate 50.4 a agonists
1115 a
SE
$165 63.9
SE
507
$67
1038 a
153
1.2
300
124
637
69
4.9
514
79
755
225
3.7
423
105
0.0
0
–
0.4
68
–
93 38.1
539
68
2378
Source: 2001–2002 MarketScan. a P b 0.05.
392 a
659
– 157
225
123
0.8
605
531
221
94
3.3
103
52
216 43.4
151
56
386
B.A. Briesacher et al. / Bone 38 (2006) 731–737 Table 4 Average inpatient costs and use of services, 2001–2002 Patients with Paget's disease Matched controls (n = 244) (n = 244)
Total inpatient Inpatient hospitalization Surgery Physician services Cardiology Radiology Other unspecified Diagnostic laboratory Diagnostic radiology Diagnostic cardiology Orthopedic surgery Hip replacement Knee replacement
Proportion using services (%)
Mean costs ($)
SE
Proportion using services (%)
36.1 30.7
16,144 3404 32.4 15,781 3526 29.1
Mean costs ($)
SE
21,480 4324 19,890 3814
18.9
2013
438 18.9
1868
206
5.3 10.7 27.9
1348 352 775
503 5.3 203 11.1 100 22.1
406 601 930
194 266 176
8.2
1064
595
20.1 12.3
4.1
1769 1006
166
62 17.6
1721 1404
563
196 10.7
298
113 282
2.9
934 a 585
3.7
2345
1.6 0.8
1750 a 88 2658 898
0.4 0.8
1955 – 2318 784
Source: 2001–2002 MarketScan. a P b 0.05.
were lower ($15,781 vs. $19,890, NS), although the difference was not statistically significant. Average inpatient costs were significantly lower (P b 0.05) for Paget's diseases than MCs for orthopedic surgery ($934 vs. $2345) and hip replacement ($1750 vs. $1955). Other average inpatient costs were similar. Discussion To our knowledge, this study is the first to consider a wide range of medical care costs for Paget's disease including inpatient, outpatient, and drug therapies, based on known patterns of disease expression, evaluation and treatment. This study is also the first to provide cost estimates from the actual payments made by the patients and their employer-based insurance health plans and to use risk adjustment methodology and a matched control group in assessing excess costs. In concurrence with other epidemiological surveys, our study detected significant disease burden due to the known complications of Paget's disease. For instance, the Paget's disease patients in our study were at increased risk of suffering fractures compared to MCs (RR 2.0), [1,23]. Low back pain was also significant in the Paget's disease cohort with 20% having this diagnosis and a relative rate of 2.28 compared to MCs. Specifically, patients with Paget's disease were more often identified as suffering from spinal stenosis (increased risk of 67%) and arthritis at any site compared to controls (RR = 1.77). The consistent finding of these
735
comorbidity rates that are in accordance with the clinical literature helps in establishing the credibility of our approach. In context of the complication rates, we found that the 244 Paget's disease patients in this study spent $588,280 more than the 244 MCs on medical care between 2001 and 2002, which is a 17.7% increase. Most of the difference came from outpatient care services, where patients with Paget's disease spent 46.7% more ($722,742) than the MCs, primarily because of higher rates of physician visits, surgery, and diagnostic testing. During the same period, Paget's disease patients also spent $141,749 more than MCs on antiresorptive agents and analgesics, which was also a significant difference. Lastly, patients with Paget's disease spent $276,211 less than MCs on inpatient care, although this difference is not significant. Harder to interpret was the increased use of some general medical services as in physician visits in the outpatient setting. Clearly, the use of specialists in rheumatology and endocrinology reached significance in this cohort of Paget's disease patients compared with matched controls, as did podiatry and radiology services. However, the increased use of unspecified physician services (97.5% vs. 93.0%) and outpatient surgery (77.9% vs. 68.0%) is not interpretable. These kinds of discrepancies highlight both the strengths and weaknesses of this study. The application of knowledge about the clinical expression and treatment of Paget's disease helped in isolating some of the most relevant cost categories, but it could not elucidate why use was so high of physician services without explicit details. Although assessing appropriateness of treatment was not an objective of this study, it is interesting to note that less than a third of our cases with Paget's disease received any antiresportive therapy but over 70% took analgesics. Our analysis has several clear limitations. The first is the low ascertainment of individuals with Paget's disease. In general, low sensitivity but high specificity is a common trait of research with administrative data, and this is especially true for conditions like Paget's disease that are commonly undetected [6]. Radiographic studies estimate that only 7% of patients with the disorder reach a clinical diagnosis, and this likely explains our low detection rate [34]. Our case selection criterion of at least 2 claims with Paget's disease diagnoses over 2 years minimized “rule out” individuals who were tested for but did not have the disease, but it decreased statistical power. We identified 1021 individuals with Paget's disease in 2001 and 1416 in 2002, but only 244 in both years. A related issue is the cross-sectional study design of the study which prevented us from distinguishing between prevalent and incident Paget's disease. Based on the observed patterns of diagnostic testing and antiresorptive prescriptions, perhaps a third of our cases were newly diagnosed, and they may have distinct costs from the others. Another limitation was our focus on conditions and categories of medical services with documented links to Paget's disease. This prevented the identification of health care costs and comorbidities that have not yet been recognized. Lastly, it
736
B.A. Briesacher et al. / Bone 38 (2006) 731–737
is also important to note that these results may not be generalizable to other patient groups, especially those who do not have employer-sponsored health insurance and prescription drug coverage. Conclusion This analysis provides new insight into the disease burden of Paget's disease and the economic consequences. By including a broad range of medical care utilization costs and using a matched control group, we were able to describe the excess burden of Paget's disease in a patient population with private insurance. Our study found evidence of significant disease burden with respect to known complications, even after accounting for comorbidities. These findings suggest a high economic burden for Paget's disease patients, especially for outpatient services, diagnostic testing, analgesics, and antiresportive treatments. Future research should assess the appropriateness of this care, as well as the longitudinal impact of therapies that may slow the progression of Paget's disease and prevent the associated complications and excess medical care costs. Acknowledgments Funding source: this study was funded in part by an unrestricted research grant from Novartis Pharmaceuticals Corporation, East Hanover, NJ. Appendix List of ICD-9 and CPT codes used for identifying complications potentially related to Paget's disease: fractures: spine, no cord injury (excl. fragility) 805.0–805.9; spine with cord injury 806.0–806.9; vertebral compression 733.13; other vertebral 805.xx, 806.xx; thoracic 806.2, 806.3; lumbar 806.4, 806.5; pelvis 808.0–808.9; femur 820.0–820.9; other parts of femur 821.0–821.3; pathological fracture: 733.10–733.19; osteoarthritis of hip or knee: 715.15, 715.25, 715.35, 715.95; hip replacement: 27130–27132; knee replacement: 27440– 27445, 27447; hearing Loss: 389.xx; heart murmur: 785.2; kidney stones: 592.0; spinal cord complications: lumbar radiculitis 724.4; spinal stenosis 724.00; cervical 723.0– 723.1; lumbar 724.02; thoracic 724.01, 724.1; posterior fossa compression: 348.4; spinal cord compression: 336.9; nerve root compression: 724.9; generic low back pain: 724.2; pain in joint: 719.4; hyperparathyroid disease: 252.0 hyperuricemia: 790.6; gout: 274.9; arthritis: 274.0; hypercalcemia 275.42. References [1] Ahmed F, Janes GR, Baron R, Latts LM. Preferred provider organization claims showed high predictive value but missed substantial proportion of adults with high-risk conditions. J Clin Epidemiol 2005;58:624–8. [2] Altman RD. Musculoskeletal manifestations of Paget's disease of bone. Arthritis Rheum 1980;23:1121–7.
[3] Altman RD, Bloch DA, Hochberg MC, Murphy WA. Prevalence of pelvic Paget's disease of bone in the United States. J Bone Miner Res 2000; 15:461–5. [4] Ankrom MA, Shapiro JR. Paget's disease of bone (osteitis deformans). J Am Geriatr Soc 1998;46:1025–33. [5] Ash AS, Ellis RP, Pope GC, Ayanian JZ, Bates DW, Burstin H, et al. Using diagnoses to describe populations and predict costs. Health Care Financ Rev 2000;21:7–28. [6] Borzecki AM, Wong AT, Hickey EC, Ash AS, Berlowitz DR. Identifying hypertension-related comorbidities from administrative data: what's the optimal approach? Am J Med Qual 2004;19:201–6. [7] Delmas PD, Meunier PJ. The management of Paget's disease of bone. N Engl J Med 1997;336:558–66. [8] Dove J. Complete fractures of the femur in Paget's disease of bone. J Bone Jt Surg, Br 1980;62-B:12–7. [9] Drake WM, Kendler DL, Brown JP. Consensus statement on the modern therapy of Paget's disease of bone from a Western Osteoporosis Alliance symposium. Biannual Foothills Meeting on Osteoporosis, Calgary, Alberta, Canada, September 9–10, 2000. Clin Ther 2001; 23:620–6. [10] Etzioni R, Riley GF, Ramsey SD, Brown M. Measuring costs: administrative claims data, clinical trials, and beyond. Med Care 2002;40: III63–72. [11] Franck WA, Bress NM, Singer FR, Krane SM. Rheumatic manifestations of Paget's disease of bone. Am J Med 1974;56:592–603. [12] Gabriel SE, Tosteson AN, Leibson CL, Crowson CS, Pond GR, Hammond CS, et al. Direct medical costs attributable to osteoporotic fractures. Osteoporos Int 2002;13:323–30. [13] Gold DT, Boisture J, Shipp KM, Pieper CF, Lyles KW. Paget's disease of bone and quality of life. J Bone Miner Res 1996;11:1897–904. [14] Gold DT, Burchett BM, Shipp KM, Pieper CF, Lyles KW. Factors associated with self-rated health in patients with Paget's disease of bone. J Bone Miner Res 1999;14(Suppl 2):99–102. [15] Grundy M. Fractures of the femur in Paget's disease of bone. Their etiology and treatment. J Bone Jt Surg, Br 1970;52:252–63. [16] Hamdy RC, Moore S, LeRoy J. Clinical presentation of Paget's disease of the bone in older patients. South Med J 1993;86:1097–100. [17] Hu XH, Markson LE, Lipton RB, Stewart WF, Berger ML. Burden of migraine in the United States: disability and economic costs. Arch Intern Med 1999;159:813–8. [18] Keen RW. The current status of Paget's disease of the bone. Hosp Med 2003;64:230–2. [19] Lafuma A, Fagnani F, Meunier PJ. An economic evaluation of tiludronic acid treatment in Paget's disease of bone. Pharmacoeconomics 1997;12:460–74. [20] Langston AL, Ralston SH. Management of Paget's disease of bone. Rheumatology (Oxford) 2004;43:955–9. [21] Mark TL, Coffey RM. What drove private health insurance spending on mental health and substance abuse care, 1992–1999? Health Aff (Millwood) 2003;22:165–72. [22] Melton III LJ, Tiegs RD, Atkinson EJ, O'Fallon WM. Fracture risk among patients with Paget's disease: a population-based cohort study. J Bone Miner Res 2000;15:2123–8. [23] Menzin J, Lang K, Friedman M, Neumann P, Cummings JL. The economic cost of Alzheimer's disease and related dementias to the California Medicaid program (“Medi-Cal”) in 1995. Am J Geriatr Psychiatry 1999; 7:300–8. [24] Mirra JM, Brien EW, Tehranzadeh J. Paget's disease of bone: review with emphasis on radiologic features: Part II. Skelet Radiol 1995;24:173–84. [25] Newton KM, Wagner EH, Ramsey SD, McCulloch D, Evans R, Sandhu N, et al. The use of automated data to identify complications and comorbidities of diabetes: a validation study. J Clin Epidemiol 1999;52:199–207. [26] Ooi CG, Fraser WD. Paget's disease of bone. Postgrad Med J 1997;73: 69–74. [27] Saifuddin A, Hassan A. Paget's disease of the spine: unusual features and complications. Clin Radiol 2003;58:102–11. [28] Schneider D, Hofmann MT, Peterson JA. Diagnosis and treatment of Paget's disease of bone. Am Fam Phys 2002;65:2069–72.
B.A. Briesacher et al. / Bone 38 (2006) 731–737 [29] Selby PL, Davie MW, Ralston SH, Stone MD. Guidelines on the management of Paget's disease of bone. Bone 2002;31: 366–73. [30] Seton M, Choi HK, Hansen MF, Sebaldt RJ, Cooper C. Analysis of environmental factors in familial versus sporadic Paget's disease of bone—The New England Registry for Paget's Disease of Bone. J Bone Miner Res 2003;18:1519–24. [31] Seton M, Krane SM. Paget's disease of bone. Curr Ther Endocrinol Metab 1997;6:568–72.
737
[32] Tiegs RD. Paget's disease of bone: indications for treatment and goals of therapy. Clin Ther 1997;19:1309–29 [discussion 1523–1524]. [33] Tiegs RD, Lohse CM, Wollan PC, Melton LJ. Long-term trends in the incidence of Paget's disease of bone. Bone 2000;27: 423–7. [34] van Staa TP, Selby P, Leufkens HG, Lyles K, Sprafka JM, Cooper C. Incidence and natural history of Paget's disease of bone in England and Wales. J Bone Miner Res 2002;17:465–71.