Disease Modifying Drugs as Treatment for Osteoarthritis ...



Disease Modifying Drugs as Treatment for Osteoarthritis: Technical Appendix

The purpose of this technical appendix is (1) to further described the Osteoarthritis Policy (OAPol) Model (Section A), (2) to provide further details on model input parameters, specifically quality of life, underlying medical costs, and treatment characteristics (Section B) presented in Table 1 (main body of the manuscript), as well as (3) to provide the results of additional sensitivity analyses pertinent to cost-effectiveness of disease modifying osteoarthritis drugs (DMOADs) (Section C). All costs described in this appendix were converted to USD 2010 using the consumer price index calculator1.

A. OAPol Model Overview

The OAPol Model is a state transition, agent-based Monte Carlo computer simulation model of the natural history and management of knee osteoarthritis. It simulates the clinical course of cohorts of individuals with knee osteoarthritis (OA).

The model uses a set of transition probabilities (which are derived from peer-reviewed literature or national databases) to determine which health state transitions the subject will experience annually. Health states are defined by obesity, comorbidities, severity of knee OA and knee pain.

Each year, the model records the subject’s quality of life and costs (based on the health states and treatment status); once the subject dies, the model outputs summary statistics. This cycle is repeated for all individuals in the hypothetical cohort. Overall measures such as life expectancy, quality-adjusted life expectancy (QALE), and costs are reported at the end of the simulation.

B. Model inputs

B.1 Quality of Life

Quality of life inputs used in the model are stratified by number of comorbidities, age, obesity status, and pain. The OAPol Model considers the following comorbid conditions: coronary heart disease, chronic obstructive pulmonary disease, diabetes mellitus, cancer, and musculoskeletal disorders other than OA. The methodology for deriving quality of life utilities using National Health and Nutrition Examination Survey (NHANES) 2005-8 has been published elsewhere2-4. Values have since been updated (Table 1 of main manuscript) to reflect a more precise definition of musculoskeletal disorders other than OA.

B.2 Annual Underlying Medical Costs

Annual underlying pain costs in the OAPol Model were derived by applying age- and disease-specific costs from the Centers for Medicare & Medicaid Services hierarchical condition categories (CMS-HCC) and weighting the CMS-HCC estimates according to data on prevalence of comorbid conditions from NHANES 2005-083-6.

Subjects who were in pain additionally incurred the cost of an annual pain management regimen of acetaminophen (500 mg, 3-4 times daily), tylenol with codeine (30 mg Tylenol with codeine, 3-4 doses daily), oxycodone with acetaminophen (5 mg, 3-4 doses daily), or hydrocodone (5 mg, 3-4 doses daily). Costs were derived from Red Book 2010 and weighted by utilization estimates from Medicare Current Beneficiary Survey (MCBS) 20067,8.

B.3 Regimen 1 – Non-Steroidal Anti-Inflammatory Drugs, Acetaminophen, Physical Therapy, Assistive Devices

B.3.1. Cost

The cost of this regimen incorporated the costs of: physician office visits; physical therapy; assistive devices; non-steroidal anti-inflammatory drugs (NSAIDs), including costs for associated gastroprotective agents (GPAs), complete blood count (CBC), and electrolyte tests; and acetaminophen (with costs for liver function tests). The first-year and subsequent-year costs of the pain management regimen were calculated as weighted averages using data-based estimates of the utilization of each individual treatment as weights. Costs and utilization rates for all elements of this regimen are presented in Table 1A below.

|Table 1A: Costs and Utilization of Regimen 1 Components |

|Treatment |Annual Cost |Utilization |First Year Cost |Subsequent Year |

| |(unweighted) | |(weighted by |Cost |

| | | |utilization rate) |(weighted by |

| | | | |utilization rate) |

|Office visits |$132 |100% |$132 |$93 |

| |(first year.) | | | |

| |$93 | | | |

| |(subs. years) | | | |

|Physical therapy |$875 |9.9% |$86 |- |

|Assistive devices |$80 |44.0% |$35 |- |

|NSAIDs+ |$1,152 |30.0% |$345 |$345 |

|Gastro-protective agents |$543 |5.8% |$32 |$32 |

|(for non-selective NSAID use) | | | | |

|CBC* and |$22 |30% |$7 |$7 |

|electrolytes lab tests | | | | |

|(for all NSAID use) | | | | |

|Acetaminophen |$20 |21.0% |$4 |$4 |

|Liver function |$15 |21.0% |$3 |$3 |

|lab test | | | | |

|(for acetaminophen use) | | | | |

|TOTAL COSTS |  |  |$643 |$483 |

|(2010 US$) | | | | |

Office Visits: It was assumed that 100% of subjects would visit a physician once annually. The cost of an office visit in the first and subsequent years on the pain management regimen was derived from Medicare reimbursement data for “new” and “established” patients9.

Physical Therapy: Using 2006 billing data for specific physical therapy procedures (including evaluation, therapeutic exercise, and gait training), we estimated the average number of bills a typical OA patient may receive on treatment7. The total cost of a one-year physical therapy regimen was calculated by multiplying the cost of each bill, derived from Medicare reimbursement data, by the number of bills per patient7,9.

Assistive Devices: The cost of assistive devices was derived as a weighted average of costs for different device types (such as canes, elbow crutches, walkers, knee braces, and orthopedic footwear) obtained from Medicare reimbursement data, using utilization rates from the literature as weights10,11.

NSAIDs: The overall annual cost of an NSAID regimen was derived as an average of NSAID costs from Red Book weighted by estimated utilization rates from MCBS7,8. To account for the lack of over-the-counter data in MCBS, we assumed that approximately 20% of subjects would utilize over-the-counter ibuprofen or naproxen.

Gastroprotective agents: Because patients who use non-selective NSAIDs are often prescribed a GPA to mitigate the effect of gastrointestinal (GI) toxicities, the cost of a GPA, weighted by the utilization of GPAs among NSAID users, was added in to the overall cost of NSAIDs7,8.

Complete blood count and electrolytes lab test: Knee OA patients who use any NSAID are likely to receive CBC and electrolytes lab tests annually. The cost of these tests, taken from Medicare data, was added to the overall cost of NSAIDs12.

Acetaminophen: The cost of acetaminophen was obtained from Red Book and weighted by its utilization rate among those with newly diagnosed OA8,13.

Liver Function Test: Patients utilizing acetaminophen are likely to undergo an annual liver function lab test; the cost of this test was taken from Medicare data12.

B.3.2. Efficacy – Pain Relief

First Year: The proportion of subjects on Regimen 1 experiencing pain relief was calculated by multiplying the average number of NSAIDs prescribed for knee OA patients (from MCBS) by the annualized probability of pain relief reported in an indomethacin trial7,14.

Subsequent Year Failure: Using the published estimates of indomethacin pain relief at 12 and 20 months, we calculated the probability that an NSAID would fail after initial efficacy using the formula below14.

PLate Failure = 1 - ( )[1 / ( t1 - t2) ]

B.3.3. Toxicity

The adverse events modeled in the pain management regimen were only applied to the proportion of subjects utilizing an NSAID (derived in Section B.3.2) and included: minor GI toxicities, myocardial infarction (MI) and thromboembolic event, or peptic ulcer bleed (major toxicities associated with non-selective NSAIDs).

Major Toxicity – Cardiovascular: The likelihood of major cardiovascular event was estimated based on a celecoxib trial that reported MI and thromboembolic events15. The annualized risk of cardiovascular event was applied to each year on treatment.

Major Toxicity – Gastrointestinal: The annualized risk of major GI toxicity was derived from literature on symptomatic GI complications associated with non-selective NSAIDs and was applied annually for subjects utilizing NSAIDs without GPAs16.

Gastroprotective Agents: The likelihood of major GI toxicity was decreased by 40% for the proportion of subjects taking NSAIDs with a GPA (derived in Section B.3.1), based on published estimates of the protective effect of misoprostol17.

Minor Toxicity – Gastrointestinal: We modeled the likelihood of minor GI toxicity as higher in the first year on the treatment than in subsequent years. We defined early risk of any minor toxicity based on a short-term naproxen trial18. Long-term risk was based on data from an 28-month study of indomethacin14.

B.4. Regimen 2 – Corticosteroid Injections

B.4.1. Cost

Annual costs for corticosteroid injections included one office visit and the cost of injections. The derivation of office visit costs for established patients was described in Section B.3.19. In order to derive the cost of injections, the average cost of corticosteroid injections from Medicare data was multiplied by the average number of corticosteroid injections used by patients with OA7,9.

B.4.2. Efficacy – Pain Relief

Efficacy for injections was derived by calculating the percent of individuals expected to achieve at least a ten-point drop in WOMAC pain (defined as a clinically significant decrease in pain) after one year19. The rate of failure in subsequent years was then calculated using the formula described in Section B.3.219.

B.4.3. Toxicity

The likelihoods of sepsis (a major toxicity) and skin flares (a minor toxicity) attributable to corticosteroid use were derived from a review on corticosteroid characteristics20.

B.5. Regimens 3 & 4 – Primary & Revision Total Knee Replacement

B.5.1. Costs

Total costs for primary and revision total knee replacement (TKR) included the costs of the procedure (derived using Medicare physician fee schedules), cost of the hospital stay (derived using Medicare data), and rehabilitation costs (derived from published literature)1,9,21,22.

Follow-up costs for TKR included the cost of regular physicians visits, and x-ray were derived from Medicare data9,23. The frequency of follow-up visits was based on a calculated average from a national survey of physicians’ recommendations24.

B.5.2. Efficacy

Early pain efficacy and late pain failure were derived from a detailed study of Medicare beneficiaries who underwent primary/revision TKR in 200025. First-year pain relief rates were calculated by subtracting percent of patients whose knee pain was not resolved by primary and revision TKR. Failure in subsequent years was calculated using the formula described in Section B.3.1.

B.5.3. Toxicity

Both medical and surgical adverse events (AEs) and their related mortalities were considered for primary and revision TKR. Post-operative medical AEs in the model included myocardial infarction, pneumonia leading to hospitalization, and pulmonary embolus; surgical AEs included prosthetic joint infection leading to prosthesis removal and rehabilitation. Prosthetic joint infection had a higher likelihood of occurring in revision than primary TKR. The likelihoods for these complications were derived from peer-reviewed TKR studies26,27.

B.6 Disease-Modifying Osteoarthritis Drugs

B.6.1 Toxicity Characteristics

The toxicity of DMOADs was modeled using data for NSAIDs. Each toxicity had associated likelihood, cost, and quality of life. For DMOADs two types of major toxicity were considered: cardiovascular and gastrointestinal. One general minor toxicity was considered.

Major Toxicity, Cardiovascular: The mortality and costs associated with cardiovascular toxicity were derived from HCUP data on patients admitted to the hospital for acute MI28. The utility associated with major cardiovascular toxicity was estimated using unadjusted European Quality of Life-5 Dimensions utility values for acute MI29.

Major Toxicity, Gastrointestinal: Mortality and costs due to a major GI toxicity were estimated based using HCUP data on patients admitted to the hospital for GI hemorrhage28. It was estimated that persons experiencing a major GI toxicity would experience 1 month of peptic ulcer bleed utility, 5 months of dyspepsia utility, and the remainder of the year in average health. Utilities for peptic ulcer bleed and dyspepsia were derived from the literature; average health utility was determined using NHANES data3,4,30.

Major Toxicity, Likelihoods: In the OAPol model, after subjects are determined to experience a major toxicity, the model next determines what type of major toxicity the subject will experience. The likelihoods of cardiovascular and GI toxicities among those who experience a major toxicity were derived from published literature on NSAID toxicity risk15,16.

Minor Toxicity: The characteristics of minor toxicity on DMOADs were derived from published data on minor GI toxicities due to NSAIDs. Cost was calculated as an average of treatment costs GPA, endoscopy to rule out severe complications) weighted by the likelihood of treatment utilization31. It was estimated that individuals with minor GI toxicity would experience one month with a dyspepsia utility and 11 months in average health30.

C. Supplemental incremental Cost-effectiveness Data for DMOADs

Technical Appendix Figures 1A and 2A present the results of additional two-way sensitivity analyses of DMOAD cost, pain relief, suspended progression, failure to sustain pain relief, failure to sustain suspended progression, and major toxicity.

Technical Appendix Figure 1A: Two-way sensitivity of DMOAD pain relief, suspended progression failure, major toxicity, suspended progression, and costs.

|Suspended |

|Progression Failure |

|(%, Subs. Yr.) |

| |Dominated |

| | |

|  |ICER > $150,000/QALY |

|  |$100,000 < ICER ≤ $150,000/QALY |

|  |$50,000 < ICER ≤ $100,000/QALY |

|  |ICER ≤ $50,000/QALY |

| | |

Technical Appendix Figure 2A: Two-way sensitivity of DMOAD suspended progression failure, cost, and pain relief failure

|Suspended Progression Failure |

| |Dominated |

| | |

|  |ICER > $150,000/QALY |

|  |$100,000 < ICER ≤ $150,000/QALY |

|  |$50,000 < ICER ≤ $100,000/QALY |

|  |ICER ≤ $50,000/QALY |

| | |

D. REFERENCES

1. Consumer Price Index Inflation Calculator. 2010; . Accessed August 4, 2012.

2. Losina E, Walensky RP, Reichmann WM, Holt HL, Gerlovin H, Solomon DH, et al. Impact of Obesity and Knee Osteoarthritis on Morbidity and Mortality in Older Americans. Annals of Internal Medicine. February 15, 2011 2011;154(4):217-226.

3. 2005-2006 National Health and Nutrition Examination Survey (NHANES) Data. National Center for Health Statistics (NCHS), U.S. Department of Health and Human Services; 2006. . Accessed August 4, 2012.

4. 2007-2008 National Health and Nutrition Examination Survey (NHANES) Data. National Center for Health Statistics (NCHS), U.S. Department of Health and Human Services; 2008. . Accessed August 4, 2012.

5. Pope GC, Kautter J, Ellis RP, Ash AS, Ayanian JZ, Lezzoni LI, et al. Risk adjustment of Medicare capitation payments using the CMS-HCC model. Health care financing review. Summer 2004;25(4):119-141.

6. Annual Estimates of the Hispanic, White, and Black Resident Populations by Sex and Age for the United States: April 1, 2000 to July 1, 2009 (NC-EST2009-04-HISP/WANH/BA). Population Division, U.S. Census Bureau; 2010. . Accessed August 17, 2012.

7. Medicare Current Beneficiary Survey: Centers for Medicare & Medicaid Services; 2006.

8. Red Book: Pharmacy's Fundamental Reference: 2010 Edition. Montvale, NJ: Thomson Healthcare; 2010.

9. Medicare Fee Schedules. Centers for Medicare & Medicaid Services; 2010. . Accessed August 4, 2012.

10. Van der Esch M, Heijmans M, Dekker J. Factors contributing to possession and use of walking aids among persons with rheumatoid arthritis and osteoarthritis. Arthritis Care & Research. 2003;49(6):838-842.

11. Medicare Durable Medical Equipment, Prosthetics/Orthotics, and Supplies Fee Schedule, 2010. Centers for Medicare & Medicaid Services. . Accessed August 4, 2012.

12. Medicare Clinical Diagnostic Laboratory Fee Schedule 2010. Centers for Medicare & Medicaid Services; 2010. . Accessed August 4, 2012.

13. Grindrod KA, Marra CA, Colley L, Cibere J, Tsuyuki RT, Esdaile JM, et al. After patients are diagnosed with knee osteoarthritis, what do they do? Arthritis Care & Research. 2010;62(4):510-515.

14. Scott DL, Berry H, Capell H, Coppock J, Daymond T, Doyle DV, et al. The long-term effects of non-steroidal anti-inflammatory drugs in osteoarthritis of the knee: a randomized placebo-controlled trial. Rheumatology. October 1, 2000 2000;39(10):1095-1101.

15. Solomon SD, McMurray JJV, Pfeffer MA, Wittes J, Fowler R, Finn P, et al. Cardiovascular Risk Associated with Celecoxib in a Clinical Trial for Colorectal Adenoma Prevention. New England Journal of Medicine. 2005;352(11):1071-1080.

16. Goldstein JL, Silverstein FE, Agrawal NM, Hubbard RC, Kaiser J, Maurath CJ, et al. Reduced risk of upper gastrointestinal ulcer complications with celecoxib, a novel COX-2 inhibitor. Am J Gastroenterol. 2000;95(7):1681-1690.

17. Silverstein FE, Faich G, Goldstein JL, Simon LS, Pincus T, Whelton A, et al. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. JAMA. 2000;284:1247 - 1255.

18. Bensen WG, Fiechtner JJ, McMillen JI, Zhao WW, Yu SS, Woods EM, et al. Treatment of osteoarthritis with celecoxib, a cyclooxygenase-2 inhibitor: a randomized controlled trial. Mayo Clinic Proceedings. November 1, 1999 1999;74(11):1095-1105.

19. Raynauld JP, Buckland-Wright C, Ward R, Choquette D, Haraoui B, Martel-Pelletier J, et al. Safety and efficacy of long-term intraarticular steroid injections in osteoarthritis of the knee: a randomized, double-blind, placebo-controlled trial. Arthritis Rheum. Feb 2003;48(2):370-377.

20. Ayral X. Injections in the treatment of osteoarthritis. Best Pract Res Clin Rheumatol. 2001;15(4):609-626.

21. Buntin MB, Garten AD, Paddock S, Saliba D, Totten M, Escarce JJ. How much is postacute care use affected by its availability? Health Serv Res. Apr 2005;40(2):413-434.

22. Medicare Hospital Inpatient Prospective Payment System. Centers for Medicare & Medicaid Services; 2010. . Accessed August 4, 2012.

23. Medicare Hospital Outpatient Prospective Payment System. Centers for Medicare & Medicaid Services; 2010. . Accessed August 4, 2012.

24. Teeny SM, York SC, Mesko JW, Rea RE. Long-term follow-up care recommendations after total hip and knee arthroplasty: Results of the american association of hip and knee surgeons' member survey. The Journal of Arthroplasty. 2003;18(8):954-962.

25. Katz JN, Mahomed NN, Baron JA, Barrett JA, Fossel AH, Creel AH, et al. Association of hospital and surgeon procedure volume with patient-centered outcomes of total knee replacement in a population-based cohort of patients age 65 years and older. Arthritis Rheum. Feb 2007;56(2):568-574.

26. Paxton EW, Namba RS, Maletis GB, Khatod M, Yue EJ, Davies M, et al. A prospective study of 80,000 total joint and 5000 anterior cruciate ligament reconstruction procedures in a community-based registry in the United States. J Bone Joint Surg Am. Dec 2010;92 Suppl 2:117-132.

27. Katz JN, Barrett J, Mahomed NN, Baron JA, Wright RJ, Losina E. Association between hospital and surgeon procedure volume and the outcomes of total knee replacement. J Bone Joint Surg Am. Sep 2004;86-A(9):1909-1916.

28. Healthcare Cost and Utilization Project (HCUP). Nationwide Inpatient Sample (NIS). Agency for Healthcare Research and Quality; 2008. . Accessed August 4, 2012.

29. Sullivan PW, Ghushchyan V. Preference-Based EQ-5D index scores for chronic conditions in the United States. Med Decis Making. Jul-Aug 2006;26(4):410-420.

30. Jansen JP, Pellissier J, Choy EH, Ostor A, Nash JT, Bacon P, et al. Economic evaluation of etoricoxib versus non-selective NSAIDs in the treatment of ankylosing spondylitis in the UK. Curr Med Res Opin. Dec 2007;23(12):3069-3078.

31. Kamath CC, Kremers HM, Vanness DJ, O'Fallon WM, Cabanela RL, Gabriel SE. The Cost-Effectiveness of Acetaminophen, NSAIDs, and Selective COX-2 Inhibitors in the Treatment of Symptomatic Knee Osteoarthritis. Value Health. 2003;6(2):144-157.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download