M Rahman

A claims-based Markov model for Crohn's disease.

Aliment Pharmacol Ther 2010; 32: 448–458

PGI9 DEVELOPMENT OF A CLAIMS-BASED MARKOV MODEL FOR CROHN'S DISEASE

PGI9 DEVELOPMENT OF A CLAIMS-BASED MARKOV MODEL FOR CROHN’S DISEASE Malone DC,Thompson HC,Van Den Bos J, Popp J, Draaghtel K, Rahman MI University of Arizona,Tucson, AZ, USA, Centocor, Inc, Horsham, PA, USA, Milliman, Inc, Denver, CO, USA OBJECTIVE: To develop a claims-based Markov model for Crohn’s disease (CD) based on the American College of Gastroenterology (ACG) criteria. METHODS: A Markov model was developed using disease states as defined by ACG CD practice guidelines. The study sample consisted of unique individuals 18 years old from the Medstat Marketscan databases (Medicare and Commercial) with 3 years of continuous enrollment from 2000– 2005, and with an ICD-9 diagnosis code of 555.xx. Patients were classified as severe-fulminant, moderate-severe, mild-moderate, or remission based on the ACG criteria. Patient exposure was divided into six month intervals, starting on first day of exposure. For each interval, disease state was defined according to the most severe disease activity. Transition probabilities between disease states were calculated based on movement from one six month period to the next. Costs of disease states were calculated using mean per member per month (PMPM) medical claim costs, and the model was run separately for males and females due to differences in life expectancy, assuming 30 years old at start. Quality-adjusted life year (QALY) estimates were obtained from the literature. RESULTS: There were 23,419 unique individuals, with 198,497 eligible 6-month intervals. The distribution of disease states were: remission (99,584; 50.2%), mild-moderate (24,788; 12.5%), moderate-severe (56,686; 28.6%), severefulminant (17,439; 8.8%). Model results for both males and females showed that, as disease severity increased, cost per QALY also increased. Cost per QALY for mild-moderate disease in males was

PAR30 SWITCHING PATTERNS AMONG ANTI-TUMOR NECROSIS FACTORS (ANTI-TNFS) IN THE TREATMENT OF RHEUMATOID ARTHRITIS

4,310 whereas for severe-fulminant disease, it was

PAR18 A FIVE-YEAR LONGITUDINAL ANALYSIS COMPARING DOSE CHANGES FOR PATIENTS WITH RHEUMATOID ARTHRITIS TAKING INFLIXIMAB IN A MEDICARE AND IN A COMMERCIALLY INSURED POPULATION

68,538. Results were similar for females (

MD1 IMPACT OF ANTI-TUMOR NECROSIS FACTORS ON HEALTH CARE RESOURCE UTILIZATION IN PATIENTS WITH IMMUNE-MEDIATED INFLAMMATORY DISEASES

4,311 vs.

PAR12 A COST-EFFICACY ANALYSIS MODEL FOR ANTI-TNF AGENTS IN PSORIATIC ARTHRITIS

68,643). CONCLUSION: Results of this model indicate that cost of CD increases as disease severity increases. In addition, although less time is spent in the severe disease state, the cost per QALY was high, suggesting that therapies that can keep patients in other disease states may prove to be beneficial.

PMS7 A 2-YEAR EVALUATION OF INFLIXIMAB'S EFFECTIVENESS IN THE TREATMENT OF RHEUMATOID ARTHRITIS IN ACTUAL PRACTICE

we have chosen the Deviance Information Criteria (DIC). The improvement in the deviance of the model for the number of principles per patient is of 1790 and in DIC of 1760. The maps reveal geographical inequalities and zones with greater average values. CONCLUSION: The Bayesian hierarchical models have been a very useful tool for incorporating geographical information in the analysis of pharmacological prescription data. They allow mapping spatial components expressing the trend of that geographical variation.

PSS1 CO-MORBIDITIES INCREASE IN THE YEAR FOLLOWING A DIAGNOSIS OF PSORIASIS

time before switching were examined. Descriptive and ChiSquare statistical analyses were conducted to determine if differences existed among the three cohorts. RESULTS: A total of 1242 patients were analyzed: 490 (39.4%) infliximab plus MTX; 607 (48.9%) etanercept plus MTX; and 145 (11.7%) adalimumab plus MTX. Over two-thirds of the patients were female and the mean age was 50.0 years. The Charlson Co-morbidity Index and disease staging were similar among the three cohorts. During the 12 months follow-up, 39 patients (7.9%) in the infliximab plus MTX cohort switched compared to 72 patients (11.9%) in the etanercept group and 23 patients (15.9%) in the adalimumab group. Chi-Square analyses indicated the differences were statistically significant (p < 0.05) as compared to the infliximab plus MTX cohort. The infliximab group had an average time of 195.9 days before switching, compared to 183.1 days in the etanercept group, and 165.3 in the adalimumab group; this was not statistically significant. CONCLUSION: The rate of switching and time before switching are important measures of the effectiveness of RA treatment in real world practice. This study found that infliximab plus MTX is associated with a longer time before switching and a significantly lower switching rate, as compared to the other anti-TNFs. Further studies are needed to evaluate the impact of switching on clinical and economic outcomes.

PMS45 A TWO-YEAR LONGITUDINAL STUDY OF SWITCHING PATTERNS AMONG ANTI-TUMOR NECROSIS FACTORS IN THE TREATMENT OF RHEUMATOID ARTHRITIS

prescription drug benefit program for the elderly instituted at the end of 2005 in the United States. METHODS: We implemented retrospective analyses of pharmacy claims of beneficiaries aged 67–79 years from 2005 to 2006, from a large pharmacy chain in the United States. Subjects aged 61–63 were used a control group in a differences-in-differences approach to account for trends not related to Part D. The final sample represented approximately 2.4 million unique beneficiaries aged 67–79. The main outcomes are: 1) Changes in proportion of total days of therapy dispensed as generics, and 2) changes in prescription utilization for each therapeutic class. RESULTS: Prescription drug use by these beneficiaries increased by 11% from 2005 to 2006. After adjustment for secular trends and other potential confounders, utilization of each therapeutic class was similar in 2005 and 2006. Small increases in drug utilization occurred for several drug classes, ranging from 0.66 pill days (0.46%) for users of nonsteroidal anti-inflammatories (NSAIDs) to 4.64 pill days (1.78%) for users of angiotensin-converting enzyme (ACE) inhibitors. Decreases occurred for anti-diabetic drugs (–2.06 pill days, –0.58%), betablockers (–1.24, –0.49%), and benzodiazepines (–5.96 pill days, –3.57%). Overall, beneficiaries were slightly less likely to fill prescriptions for generic drugs vs. brand-name drugs in 2006 compared to 2005 (OR 0.98, 95% CI 0.97–0.98). CONCLUSION: Small increases in prescription drug utilization occurred across numerous drug classes for these Medicare seniors following the implementation of the Medicare Part D Prescription Benefit, while overall market share by drug class did not change significantly. Further analyses are needed to explore the degree to which these changes reflect moral hazard versus beneficial expansions of coverage.

PMS31 TWO-YEAR LONGITUDINAL STUDY OF PERSISTENCE TO ANTI-TUMOR NECROSIS FACTOR TREATMENT AMONG RHEUMATOID ARTHRITIS PATIENTS

model were obtained from published clinical trials and were complemented with Mexican expert opinion surveys. Effectiveness measure was the number of patients with articular pain controlled without adverse events (peptic ulcers, gastrointestinal bleeding, and others). The analysis was conducted from the healthcare payer’s perspective. Resource use and costs were obtained from hospital records and Mexican official databases. Threshold and probabilistic sensitivity analysis was performed and acceptability curves were constructed. RESULTS: The model indicates that the use of celecoxib could lead to the avoidance of a significant number of adverse events associated to NSAIDs and acetaminophen. Celecoxib showed on the six-months period similar (p = 0.52) expected costs per patient (US