Satisfactory Adherence to Intranasal Corticosteroids is Associated - - PowerPoint PPT Presentation

Philip O. Buck1; Cheryl S. Hankin2; Linda Cox3; Amy Bronstone2; Zhaohui Wang2; Mark S. Lepore1

1Teva Pharmaceuticals, Inc., Frazer, PA; 2BioMedEcon, Moss Beach, CA; 3Nova Southeastern University School of Osteopathic Medicine, Ft Lauderdale, FL

Satisfactory Adherence to Intranasal Corticosteroids is Associated With Significantly Reduced Number and Costs of Outpatient Visits Among Patients Newly Diagnosed With Allergic Rhinitis

1

Funding for this research was provided by Teva Pharmaceuticals

Background

• Estimated global prevalence of allergic rhinitis (AR): 400 million1
• Global prevalence is on the rise
• AR is a well-documented risk factor for2,3
• Patient-related burden of AR4-6
• Economic burden of AR7
• In the U.S. alone, AR affects 30 to 60 million people and accounts for an estimated $14 billion (2012

USD) in annual direct costs.

• This is roughly equivalent to the annual direct costs of adult vision problems in the U.S.8

2

1. World Health Organization. White Book on Allergy 2011-2012 Executive Summary. By Prof. Ruby Pawankar, MD, PhD, Prof. Giorgio Walkter Canonica, MD, Prof. Stephen T. Holgate, BSc, MD, DSc, FMed Sci and Prof. Richard F. Lockey, MD.Wallace DV, Dykewicz MS, Bernstein DI, et al. J Allergy Clin Immunol. 2008;122(2 Suppl):S1-84. 2. Settipane RA. Complications of allergic rhinitis. Allergy Asthma Proc 1999;20:209-13. 3. Lack G. Pediatric allergic rhinitis and comorbid disorders. J Allergy Clin Immunol. 2001;108(suppl):S9 –S15. 4. Sundberg R, Toren K, Hoglund D, Aberg N, Brisman J. Nasal symptoms are associated with school performance in adolescents. J Adolesc Health. 2007;40:581–583. 5. Walker S, Khan-Wasti S, Fletcher M, Cullinan P, Harris J, Sheikh A. Seasonal allergic rhinitis is associated with a detrimental effect on examination performance in United Kingdom teenagers: case-control study. J Allergy Clin Immunol. 2007;120:381–387. 6. Leger D, Annesi-Maesano I, Carat F, et al. Allergic rhinitis and its consequences on quality of sleep: an unexplored area. Arch Intern Med. 2006;166:1744 –1748. 7. Soni A. Allergic rhinitis: Trends in use and expenditures, 2000 to 2005. Statistical Brief #204. Bethesda, MD: Agency for Healthcare Research and Quality; 2008. 8. Prevent Blindness America. The economic impact of vision problems: The toll of major adult eye disorders, visual impairment, and bliindness on the U.S. Economy. 2007. Accessed October 1,

• 2012. http://www.preventblindness.net/site/DocServer/Impact_of_Vision_Problems.pdf.
• 10-30% of adults
• 40% of children
• Climate change and increased air pollution
• Greatest risk within urban and polluted regions
• Asthma
• Obstructive sleep apnea
• Eustachian tube dysfunction
• Conjunctivitis
• Sinusitis
• Otitis media with effusion
• Eczema
• Pharyngitis
• Other respiratory infections
• Poor quality of life
• Learning difficulties
• Decreased appetite
• Impaired school and work

performance

• Excessive daytime fatigue and

somnolence

• Reduced participation in sports

and outdoor activities

• Compromised social interactions
• Sleep disturbance
• Depressed mood and irritability

Study Objective

• Intranasal corticosteroids (INS), first-line pharmacological treatment

for moderate to severe persistent AR,1 must be used continuously for maximum effectiveness.2

• Because of the chronic, long-term nature of persistent AR,

adherence to therapy is difficult3,4 and may result in unsatisfactory disease control, thereby increasing health care costs.

• The objective of this study was to examine the relationship between

adherence to INS and health care costs among patients with newly- diagnosed AR who initiate INS.

3 1. Soni A. Allergic rhinitis: Trends in use and expenditures, 2000 to 2005. Statistical Brief #204. Bethesda, MD: Agency for Healthcare Research and Quality; 2008. 2. Laekeman G, Simoens S, Buffels J et al. Respir Med. 2010;104:615-25. 3. Wagner S, Luskin A, Bukstein D, et al. J Allergy Clin Immunol. 2009;123:S46. 4. Bukstein D, Luskin AT, Farrar JR. Allergy Asthma Proc. 2011;32:265-71.

Methods: Study Design and Sample Selection

4

All Florida Medicaid enrollees (1997-2009) N=7,524,231

AR and age ≥12 years N=108,050 No AR or age <12 years N=7,416,181 Newly diagnosed AR N=75,337 Not newly diagnosed AR N=32,713 Ever received ≥1 INS Rx N=14,991 Never received INS Rx N=60,346 ≥1 year claims data preceding index INS fill without any INS claim N=10,794 <1 year claims data preceding index INS fill N=4,197 ≥3 years claims data following index INS fill N=3,665 <3 years claims data following index INS fill N=1,740 Index INS fill followed initial AR diagnosis N=5,405 Index INS fill preceded initial AR diagnosis N=5,389 Satisfactory INS adherence in Year 1 following index INS fill N=486 Unsatisfactory INS adherence in Year 1 following index INS fill N=3,179 DESIGN: Retrospective matched cohort study of Florida Medicaid claims data (July 1, 1997 to June 30, 2009) SELECTION CRITERIA:

• Age ≥12 years at 1st identified AR claim
• Ever received ≥1 INS Rx (per National Drug

Codes; NDC)

• Index INS fill followed (rather than

preceded) newly diagnosed AR

• Sufficient data (≥3 years following

index INS fill) for analysis Satisfactory adherence = Medication Possession Ratio ≥70% Unsatisfactory adherence = Medication Possession Ratio <70%

• “Newly diagnosed” with AR
• No AR diagnosis ≥1 year preceding 1st

identified AR claim (ICD-9 477.0, 477.8, 477.9)

• Previously naïve to INS
• ≥1 year preceding index INS fill without

any claim filed for INS

• Definition of adherence
• Medication possession ratio (MPR) = Total days supplied

365 days following 1st INS fill

• “Satisfactory” INS adherence (Satisfactory-INS): MPR ≥70%
• “Unsatisfactory” INS adherence (Unsatisfactory-INS): MPR <70%
• Satisfactory-INS patients were matched 1:3 to Unsatisfactory-INS patients on
• Age at first AR diagnosis (±6 months)
• Sex
• Race/ethnicity
• Charlson Comorbidity Index 1 year prior to initial AR diagnosis
• Comorbid atopy (asthma, atopic dermatitis, conjunctivitis) 1 year prior to index INS fill
• If an Satisfactory-INS patient had more than 3 Unsatisfactory-INS matches, we

randomly selected 3 Unsatisfactory-INS patients from eligible matches

• Because cost data are typically highly skewed, we conducted paired t-tests on

log-transformed geometric mean costs.

Methods: Analytic Approach

5

X 100

Demographics After Matching

Characteristic SATISFACTORY-INS (N=343) Matched Control UNSATISFACTORY-INS (N=698) P Value Age at initial AR diagnosis, N (%) 12-17 years 29 (8.5) 65 (9.3) 0.530 18-29 years 52 (15.2) 122 (17.5) 30-39 years 68 (19.8) 154 (22.1) 40-49 years 69 (20.1) 154 (22.1) 50-59 years 53 (15.4) 97 (13.9) ≥60 years 72 (21.0) 117 (16.8) Sex, N (%) Female 279 (81.3) 596 (85.4) 0.094 Race/ethnicity, N (%) White, non-Hispanic 162 (47.2) 361 (51.7) 0.491 Black 56 (16.3) 113 (16.2) Hispanic 44 (12.8) 84 (12.0) Other 81 (23.6) 140 (20.1) Charlson Index, N (%) 0 (minimal) 228 (66.5) 497 (71.2) 0.118 ≥1 (mod/severe) 115 (33.5) 201 (28.8) Comorbidity, N (%) Asthma 38 (11.1) 53 (7.6) 0.079 Atopic dermatitis 0 (0) 0 (0) NA Conjunctivitis 5 (1.5) 7 (1.0) 0.544

6

Health care utilization in the 6, 12, and 18 months after 1st INS fill by satisfactory versus unsatisfactory adherence

Health Services Use# SATISFACTORY-INS (N=343) Matched Control UNSATISFACTORY-INS (N=698) P Value* N Mean ± SD N Mean ± SD NUMBER OF INPATIENT STAYS 6 months 5 4.4 ± 3.2 5 1.2 ± 0.4 0.151 12 months 14 3.7 ± 4.4 14 1.7 ± 1.1 0.177 18 months 23 3.7 ± 5.1 25 2.9 ± 3.6 0.606 NUMBER OF OUTPATIENT VISITS 6 months 302 10.5 ± 10.8 598 10.9 ± 10.6 0.027 12 months 325 18.8 ± 21.4 651 20.1 ± 22.8 0.005 18 months 333 27.3 ± 31.9 670 29.6 ± 37.2 0.0007 NUMBER OF PHARMACY FILLS 6 months 343 38.0 ± 25.6 698 28.8 ± 17.3 <0.0001 12 months 343 69.7 ± 49.9 698 54.8 ± 34.9 0.0007 18 months 343 101.0 ± 75.3 698 80.2 ± 52.6 <0.0001 *p-value comparisons of log-transformed geometric means.

#Based on a 2-part conditional model for each type of Health Services Use: 1) identification of each patient with any use in the Satisfactory-INS group and matched up to

3 with patients in the Unsatisfactory-INS group with any such use, followed by 2) paired t-tests of log-transformed geometric means. If no match was available in the Unsatisfactory-INS group, the patient in the Satisfactory-INS group was excluded from further analysis. −Duan, N., W.G. Manning, C.N. Morris, et al., "A Comparison of Alternative Models for the Demand for Medical Care," J Bus Econ Stat 1983;1(2): 115‐126. −Duan, N., W.G. Manning, C.N. Morris, et al., "Choosing Between the Sample Selection Model and the Multi‐Part Model," J Bus Econ Stat. 1984; 2(3):283‐289.. North Holland:Elsevier;2000. p.265-344. −Manning W.G. .Dealing with Skewed Data on Costs and Expenditures. In Jones A, editor. The Elgar Companion to Health Economics, 2nd Edition. Northhampton:Edward Elgar Publishing Inc;2012.p 439-446.

7

Health care costs in the 6, 12, and 18 months after 1st INS fill by satisfactory versus unsatisfactory adherence

Health Services Costs#

SATISFACTORY-INS (N=343) Matched Control UNSATISFACTORY-INS (N=698)

P Value* N Mean ± SD N Mean ± SD INPATIENT STAY COSTS 6 months 5 $29,736 ± $31,905 5 $3,673 ± $3,596 0.178 12 months 14 $22,317 ± $45,348 14 $4,629 ± $4,145 0.381 18 months 23 $22,265 ± $54,516 25 $11,221 ± $19,711 0.735 OUTPATIENT VISIT COSTS 6 months 302 $1,101 ± $2,108 598 $1,096 ± $1,286 0.0003 12 months 325 $1,806 ± $2,631 651 $2,003 ± $2,427 <0.0001 18 months 333 $2,622 ± $3,372 670 $2,965 ± $3,765 <0.0001 PHARMACY COSTS 6 months 343 $2,163 ± $1,844 698 $1,730 ± $1,319 0.002 12 months 343 $4,032 ± $3,627 698 $3,371 ± $2,685 0.072 18 months 343 $5,911 ± $5,356 698 $4,972 ± $4,040 0.140

p-value comparisons of log-transformed geometric means.

#Based on a 2-part conditional model for each type of Health Services Costs: 1) identification of each patient with any use in the Satisfactory-INS group

and matched up to 3 with patients in the Unsatisfactory-INS group with any use, followed by 2) paired t-tests of log-transformed geometric means. If no match was available in the Unsatisfactory-INS group, the patient in the Satisfactory-INS group was excluded from further analysis.

−Duan, N., W.G. Manning, C.N. Morris, et al., "A Comparison of Alternative Models for the Demand for Medical Care," J Bus Econ Stat 1983;1(2): 115‐126. −Duan, N., W.G. Manning, C.N. Morris, et al., "Choosing Between the Sample Selection Model and the Multi‐Part Model," J Bus Econ Stat. 1984; 2(3):283‐289.. North Holland:Elsevier;2000. p.265-344. −Manning W.G. .Dealing with Skewed Data on Costs and Expenditures. In Jones A, editor. The Elgar Companion to Health Economics, 2nd Edition. Northhampton:Edward Elgar Publishing Inc;2012.p 439-446.

8

Summary of Results

9

HEALTH SERVICES

Δ AT 6 MONTHS Δ AT 12 MONTHS Δ AT 18 MONTHS

INPATIENT STAYS $ DIFFERENCES BETWEEN GROUPS NOT SIGNIFICANT AT ANY TIME POINT OUTPATIENT VISITS $5 p=0.0003

• $197

p<0.0001

• $343

p<0.0001 PHARMACY FILLS $433 p=0.002 $ DIFFERENCES BETWEEN GROUPS NO LONGER SIGNIFICANT Note: Negative values denote cost savings (benefit) for SATISFACTORY-INS group.

Net Savings Conferred: SATISFACTORY-INS Group Minus UNSATISFACTORY-INS Group

Recap

10

• We found no differences with regard to inpatient utilization or costs by INS
• adherence. This finding is not unexpected, given that inpatient resource use

represents a catastrophic outcome that is uncommon for AR patients.

• Although patients with satisfactory adherence consistently incurred

significantly more pharmacy fills than their matched counterparts with unsatisfactory adherence, pharmacy costs did not significantly differ between groups by 12 months following INS initiation.

• Compared with matched counterparts with unsatisfactory INS adherence,

patients with satisfactory INS adherence consistently incurred significantly fewer outpatient visits across 18 months; the significant outpatient cost benefit experienced by those with satisfactory INS adherence became apparent by 12 months post INS initiation.

Limitations

11

• The retrospective nature of this study precludes definitive conclusions

regarding causality.

• Findings may not generalize to broader non-Medicaid (privately

insured) patient populations.

• Groups may have differed on variables that were not controlled for by

matching procedures.

• Claims data do not include information about type of allergy (perennial
• vs. seasonal).
• Whereas patients with perennial allergies typically require year-round

therapy, those with seasonal allergies may only need regular therapy during certain times of the year.

• The MPR as a measure of adherence may incorrectly identify some

patients with seasonal allergies as having unsatisfactory adherence to INS.

Conclusions and Discussion

• We had expected to find that AR-diagnosed patients with higher

levels of INS adherence would tend to seek more health care services and thereby incur higher health care costs.

• We were therefore surprised to find that AR-diagnosed patients

with satisfactory INS adherence incurred fewer outpatient visits and lower outpatient-related costs than matched counterparts with unsatisfactory levels of INS adherence.

• We were also surprised to find that there were no significant

differences between INS adherence groups with respect to pharmacy costs at 12 and 18 months post INS initiation.

• These findings suggest that INS adherence may confer a protective

health benefit that is reflected in significantly fewer outpatient visits and lower outpatient costs over 18 months post INS initiation.

12