QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Résumé Extracto PDF Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Articles Ahead of Print [Order Reprint] Citing Articles Citing Articles via Google Scholar Google Scholar Articles by LaFleur, J. Articles by Brixner, D. I Search for Related Content PubMed PubMed Citation Articles by LaFleur, J. Articles by Brixner, D. I

ADHERENCE

Adherence and Persistence with Single-Dosage Form Extended-Release Niacin/Lovastatin Compared with Statins Alone or in Combination with Extended-Release Niacin

Research Assistant Professor, Department of Pharmacotherapy, College of Pharmacy, University of Utah, Salt Lake City, UT

Clinton J Thompson, MStat

Research Assistant, Department of Family and Preventive Medicine, School of Public Health, University of Utah

Vijay N Joish, PhD

Research Assistant Professor, Department of Pharmacotherapy, College of Pharmacy, University of Utah

Scott L Charland, PharmD FCCP

Associate Director of Health Outcomes, Kos Pharmaceuticals, Inc., Clinical Applied Science, Medical Affairs, Cranbury, NJ; Adjunct Associate Professor, School of Pharmacy, University of Colorado, Denver, CO

Gary M Oderda, PharmD MPH

Professor, Department of Pharmacotherapy, College of Pharmacy, University of Utah

Diana I Brixner, PhD

Associate Professor and Chair, Department of Pharmacotherapy, College of Pharmacy, University of Utah

Reprints: Dr. LaFleur, Pharmacotherapy Outcomes Research Center, 421 Wakara Way, Suite #208, University of Utah, Salt Lake City, UT, 84108-3546, fax 801/581-7442, joanne.lafleur{at}pharm.utah.edu

	Abstract

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
BACKGROUND: Lipid-lowering therapies have been shown to reduce cardiovascular events and mortality; patient cooperation with therapy varies. A fixed-dose combination product, extended-release niacin/lovastatin (ERNL), has been shown to be beneficial in lipid management; however, little is known regarding patient behavior with ERNL therapy.

OBJECTIVE: To evaluate patient adherence and persistence with ERNL, statin monotherapy (SM), extended-release niacin (ERN) monotherapy, and ERN plus a statin (ERN-S).

METHODS: Prescription claims for lipid-lowering therapies were obtained from a pharmacy benefits manager between 2002 and 2003. Claims for a total of 2389 patients were analyzed for adherence and persistence, using medication possession ratios (MPRs) and proportions of days covered (PDCs). Adherence and persistence were defined, respectively, as an MPR or PDC greater than or equal to 0.80. Logistic regression was conducted to detect differences among groups. Covariates included age, gender, copay, and number of lipid-lowering therapies, a surrogate for disease severity.

RESULTS: Average MPR scores were relatively high in all groups at 0.88, 0.81, 0.89, and 0.90 for ERNL, SM, ERN, and ERN-S, respectively. The adjusted odds ratio for adherence was lowest for SM (0.69), which was statistically significant compared with ERN-S (1.43), but not ERNL (1.00) or ERN (0.74). Persistence outcomes were poor in all groups. By the fourth quarter, patients receiving ERN-S (OR 1.31) had significantly greater persistence than those receiving ERN (OR 0.41) and SM (0.61), but not those receiving ERNL (OR 1.00).

CONCLUSIONS: Managed care patients tended to be adherent to chronic lipid-lowering therapies, based on a mean MPR greater than 0.8. However, most patients failed to persist for at least 6 months.

Key Words: adherence, compliance, niacin, persistence, statins

Published Online, July 18, 2006. www.theannals.com, DOI 10.1345/aph.1G646

Prospective epidemiologic studies have consistently demonstrated that a low high-density lipoprotein cholesterol (HDL-C) level is also a strong independent risk factor for coronary artery disease.^2,6-8 Currently, fibrates and niacin are the only 2 drug classes approved by the Food and Drug Administration that significantly increase HDL-C levels and improve coronary artery disease outcomes.^2,9-12 Fibrates are associated with a 10-20% increase in HDL-C levels, while niacin is associated with a 15-35% increase.² A combination of niacin and a statin appears to significantly reduce coronary artery disease events (risk reductions of 60-90%) above the "ceiling" effect seen with statin monotherapy (SM).^13,14 However, the immediate-release niacin formulation is associated with significant adverse effects, including flushing, pruritus, rashes, and gastrointestinal distress, leading to discontinuation in 10-50% of patients in clinical trials.¹⁵

Extended-release niacin (ERN), a formulation of niacin absorbed over 8-12 hours, is available as the only approved prescription ERN product.^15,16 The ERN formulation retains the traditional lipid efficacy of immediate-release niacin, with improved tolerability and avoidance of the hepatotoxicity associated with sustained-release niacin dietary supplements sold without a prescription. In clinical trials, fewer than 6% of patients receiving ERN discontinued therapy as a result of flushing.^15,16 An ERN/lovastatin combination product (ERNL) was introduced to the market in 2001 for treatment of primary hypercholesterolemia and mixed dyslipidemias.⁹ It is the only statin/niacin combination product currently available in the US.

It is not known how patient adherence and persistence with the combination product compare with that associated with SM in a real-world setting. However, studies of real-world populations that have evaluated patient adherence or persistence with SM have consistently shown poor results.^10-12,17-19 Although no studies of single-dosage-form combination, lipid-lowering therapies have been conducted, studies in other disease states have shown greater patient cooperation with single-dosage-form combination therapy compared with 2-dosage-form polytherapy²⁰ and that the number of daily medications is inversely correlated with measures of patient cooperation.²¹

	DEFINITIONS

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
In studies of lipid-lowering therapies, the terms used to describe patient medication-taking behavior in the literature include adherence, nonadherence, persistence, discontinuation, and compliance. These terms are often confused and have not been well defined for researchers in the past. Recently, the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) proposed definitions for the terms compliance, adherence, and persistence.²² Under ISPOR's proposed definitions, compliance and adherence are equivalent terms, defined as the ratio of days over a given interval during which "a patient acts in accordance with the prescribed interval and dose of a dosing regime." The proposed definition for persistence is "the accumulation of time from initiation to discontinuation of therapy" and is reported as a time metric in days, months, or years.

	PURPOSE

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
The primary purpose of this study was to evaluate patient adherence with ERNL, SM, ERN monotherapy, and ERN plus a statin (ERN-S), using medication possession ratios (MPRs).²³ The secondary endpoint was to evaluate patient persistence, using proportions of days covered (PDCs).¹⁹

	Methods

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
DATA SOURCE
All prescription claims for lipid-lowering therapies received from July 2001 to December 2003 were obtained from a national pharmacy benefits manager, RxAmerica. The RxAmerica database contains prescription claims data for more than 4.5 million patients from 53 000 pharmacies and mail service facilities nationally. The source population included all patients who were eligible for pharmacy benefits under managed care plans continuously from July 2001 to December 2003 and who were considered new starts with ERNL, SM, ERN, or ERN-S between January 2002 and December 2003. To ensure that patients had not previously received antihyperlipidemic agents, new starts were defined as no prior therapy with the study drug within 6 months prior to index medication start date. The date of the first prescription for lipid-lowering therapy after January 1, 2002, was identified; patients with any lipid-lowering therapy in the 6 months prior to that date were excluded from the analysis. Patient identity was masked throughout the study period in accordance with the Health Insurance Portability and Accountability Act.

OUTCOME MEASURES
The primary endpoint, adherence, was measured by calculating MPRs. MPRs were defined as the total days supplied divided by the difference in days between the first fill and the last day of the last days-supplied.²³ For the ERN-S group, the mean of the MPRs for the statin and ERN were used. Adherent individuals were defined as those having an MPR of 0.8 or more. The threshold level of 0.8 was based on a secondary prevention population in Tayside, Scotland.²⁴ The secondary endpoint, persistence, was measured using PDCs.¹⁹ PDCs were calculated by summing the days supplied to each patient in each quarter following initiation of therapy and dividing by the number of days in that quarter. Persistence in each quarter was defined as a PDC of 0.8 or more.

INCLUSION/EXCLUSION CRITERIA
Patients were considered eligible for inclusion in the ERNL, ERN, and statin groups if they were new starts with the study agent. Patients were considered eligible for inclusion in the ERN-S group if they had received therapy with a statin plus ERN on overlapping days and if they were new starts with one of the drugs. Switching behavior was allowed within the statin class in the ERN-S and statin groups, meaning that endpoints were calculated for all statin prescriptions if patients started therapy with one statin and then switched to a different statin during the study period.

All groups were mutually exclusive; however, many patients were eligible for inclusion in more than one group at different times during the study period. Because it was expected that there would be many more patients in the statin group, those who were eligible for inclusion in multiple groups were assigned in the following order: (1) ERNL, (2) ERN-S, (3), ERN, and (4) SM.

The adherence and persistence endpoints had different inclusion criteria. Since, by definition, MPR requires 2 or more fills of the study medication, patients with one fill of that drug were excluded from the MPR analysis. PDCs may be calculated for patients who only received a single fill of each study agent. However, PDCs were calculated only for patients whose index dates were at least 90 days prior to the end of the observation period since the patient had to be observable for at least one quarter to calculate a PDC. Patients starting therapy fewer than 90 days prior to the end of the observation period were excluded from the PDC analysis, and PDCs in each quarter were calculated only for the proportion of patients who were observable for a sufficient duration.

STATISTICAL ANALYSES
All tests were conducted at an level of 0.05 using Intercooled Stata 8.0.²⁵ Two-sided t-tests were used for continuous variables, and ² tests were used for categorical variables. Bonferroni corrections were made to account for multiple comparisons.

For the MPR endpoint, multiple linear regression analysis was used to compare mean MPRs between groups. Logistic regression analysis was also used to determine whether there were any differences in the probability of being categorized as adherent for each comparison. For the PDC endpoint, logistic regression was used to determine the differences in the probability of being categorized as persistent among the groups in the third and fourth quarter.

For both endpoints, several covariates were adjusted for, including age as a continuous variable and gender as a dichotomous variable. In addition, because the difference in days between the index dates and the end of the observation period varied substantially among groups, the number of days each patient was observed was also adjusted for as a continuous variable. Two additional categorical variables were also included to adjust for copay and for the number of lipid-lowering therapies prescribed to each patient. The categories for number of lipid-lowering agents included 1, 2, or more than 2 different classes; this variable was considered a surrogate for disease severity.

The copay variable was calculated by averaging all out-of-pocket expenditures over the time intervals between the index date and the dates of each comparison. The variable for the number of classes of lipid-lowering therapies was calculated by summing the number of agents received over the same intervals. The comparison date for the MPR analysis was the last day of treatment with the study drug; for the PDC analysis, comparisons were made at the end of each quarter for which a PDC was calculated. For the PDC analysis, if patients did not fill a prescription in a subsequent interval, the last observation was carried forward to subsequent intervals since a prior copay was thought to have bearing on subsequent filling behavior. All copays were normalized for a 30 day supply, and the copays were categorized as $0-10.00, $10.01-20.00, and more than $20.

	Results

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
PATIENTS
A total of 33 295 patients received lipid-lowering medication; of these, 6651 were considered new starts on the study drugs between January 2002 and December 2003. All of the patients who were new starts with ERNL, ERN, and ERN-S were included in the source population. A total of 5325 SM patients were identified. To avoid overpowering the study, a random sample of 1000 statin patients were included in the source population. In the ERNL, SM, ERN, and ERN-S groups, 29.5%, 93.2%, 50.4%, and 6.8%, respectively, received no therapy prior to the index date. Baseline demographics of the source population and the subsets used in the MPR and PDC analysis are shown in Table 1.

Gender, mean age, number of days observed, copay, and number of classes of medications tried differed substantially among the groups. Patients in the ERNL and ERN-S groups tended to be slightly older, and the groups tended to have a higher proportion of males compared with the ERN and SM groups. Patients in the ERNL group tended to be observed for fewer days compared with patients in the other groups. Patient copays were higher for the ERN-S group compared with the other groups. As would be expected, the average number of classes of lipid-lowering medications was higher for the combination therapy groups compared with the SM and ERN groups. The median per-day costs to the third party for the study drug were $1.15, $1.76, $0.73, and $3.43 for ERNL, SM, ERN, and ERN-S, respectively.

Of the 2389 patients in the source population, 88 ERNL, 216 SM, 248 ERN, and 145 ERN-S patients filled only a single prescription for the study drug and consequently were excluded from the MPR analysis. A total of 46 ERNL, 121 SM, and 96 ERN patients had index dates fewer than 90 days prior to the end of the observation period and were excluded from the PDC analysis.

ADHERENCE
Adherence outcomes are summarized in Table 2. Average MPR scores were relatively high in all groups. The differences were statistically significant for SM compared with ERN-S (p = 0.016) and ERN (p = 0.007) and for ERN-S compared with ERNL (p = 0.033) when adjusting for covariates. The proportions of patients who were categorized as adherent (MPR 0.80) were also relatively high in all groups with 72.5%, 63.9%, 65.9%, and 75.8% meeting the 80% threshold in the ERNL, SM, ERN, and ERN-S groups, respectively. The adjusted odds ratio for adherence was lowest for SM, which was statistically significant compared with ERN-S, but not with ERNL or ERN.

PERSISTENCE
Persistence outcomes are summarized in Table 3. Persistence decreased substantially after 6 months. Fewer than 25% of patients persisted through the third quarter and fewer than 20% of patients persisted through the fourth quarter in any group. Persistence at the end of the third quarter was substantially lower for patients in the ERN group compared with patients in all other groups (p < 0.05). However, by the fourth quarter, persistence in the SM group had fallen substantially and was no longer significantly higher compared with ERN.

	Discussion

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
We hypothesized that ERNL would provide benefits to measures of patient cooperation with lipid-lowering therapies; in fact, we found that adherence and persistence with ERNL were similar to those with ERN-S and SM and different from those with ERN in the population studied. The mean MPR score, a proxy for adherence, was higher for ERNL (0.88) compared with statins (0.81), but this difference was not statistically significant when adjusted for covariates. In addition, the mean MPR score of all groups was above the threshold of 0.8,²⁶ suggesting that any differences between ERNL and other lipid-lowering therapies was not clinically significant. In addition, the difference in the probability of patients being adherent for ERNL was not significant compared with any of the other 4 therapies.

Two previous studies using administrative claims data have shown adherence estimates of about 60% after 5 years¹¹ and 40% after 10 years.¹⁹ The higher adherence estimates we found may be due to 2 factors. First, we only looked at patients for a 2 year period compared with the 5 and 10 year periods used in the previous studies. Second, and probably more important, those investigations used the number of days in the entire observation interval as the denominator in their adherence ratios—from the time the patient began therapy until the end of the observation period. We used the classic MPR denominator, which was the difference in days between the first and last fill. Other studies that used the difference in days between the first and last fill as the denominator found adherence ratios of 74%¹⁰ and 80%,¹² similar to those that we found.

Contrary to what we expected based on other studies of 1- versus 2-dosage-form combination therapies,²⁰ our persistence analyses showed that patients were not more likely to persist with ERNL therapy than with any other treatment except ERN. This finding was unexpected and most likely an artifact of the methodology that allowed switching behavior with the statins in both the ERN-S and SM groups but not the ERNL group. However, we selected this methodology to bias our study conservatively toward the null hypothesis since switching between statins in clinical practice occurs commonly and is frequently appropriate.

The differences in measures of patient cooperation with therapy among the 4 groups were very small. Additionally, although adherence rates would suggest that patients are compliant with these therapies, an evaluation of persistence rates indicates that, although patients take these therapies consistently as long as they continue to get refills, most patients fail to persist with therapy prematurely. This suggests that reporting MPR alone is an insufficient measure of patient medication-taking behavior for studies of medications intended to be chronic therapies. A measure of persistence is also necessary.

	Limitations

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
This study of lipid utilization in a PBM was associated with several limitations. First, the analysis was conducted using a prescription claims database; thus, we did not have clinical response variables. Patients may be more likely to persist with therapy if it is working for them. However, the clinical efficacy of each of these agents has been established previously. Second, drug-related adverse events are not indicated in the database, so it is impossible to know whether the drugs were discontinued as a result of adverse events. Finally, the database only indicates that a drug was dispensed—not that it was consumed. However, it is reasonable to assume that, if a patient continues to refill a prescription, continued adherence is intended.

	Conclusions

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 
The results of our study show that the managed care patients evaluated in our analysis tended to be adherent with chronic lipid-lowering therapies, based on a mean MPR greater than 0.8. However, most patients failed to persist for more than 6 months.

	Footnotes

 
This project was funded in part by an unrestricted educational grant from Kos Pharmaceuticals, Inc.

	References

Top Abstract DEFINITIONS PURPOSE Methods Results Discussion Limitations Conclusions References

 

1. American Heart Association. Heart disease and stroke statistics: 2004 Update. Previous update 2003. www.americanheart.org/downloadable/heart/1040391091015HDSStats_04.pdf (accessed 2004 Sept 19).
2. National Cholesterol Education Panel (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). Third Report of the National Cholesterol Education Panel (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report.Circulation 2002;106:3143-421.[Free Full Text]
3. Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7-22.[CrossRef][Medline]
4. Cannon C, Braunwald E, McCabe C, et al. Comparison of intensive and moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004;350:1495-504.[Abstract/Free Full Text]
5. de Lemos J, Blazing M, Wiviott SD, et al. Early intensive vs a delayed conservative simvastatin strategy in patients with acute coronary syndromes: phase Z of the A to Z trial. JAMA 2004;292:1307-16.[Abstract/Free Full Text]
6. Assman G, Schulte H, von Eckardstein A, Huang Y. High-density lipoprotein cholesterol as a predictor of coronary heart disease risk: the PROCAM experience and pathophysiological implications for reverse cholesterol transport. Atherosclerosis 1996;124(suppl):S11 -20.[CrossRef][Medline]
7. Gordon T, Castelli W, Hjortland M, Kannel W, Dawber T. High density lipoprotein as a protective factor against coronary heart disease. The Framingham study. Am J Med 1977;62:707-14.[CrossRef][Medline]
8. Despres J, Lemieux G, Dagenais B, Lamarche B. HDL-cholesterol as a marker of coronary heart disease risk: the Quebec cardiovascular study.Atherosclerosis 2000;153:263-72.[CrossRef][Medline]
9. US Food and Drug Administration Center for Drug Evaluation and Research. Approval letter for Advicor (niacin extended-release lovastatin) tablets. www.fda.gov/cder/foi/nda/2001/21-249_Advicor.htm (accessed 2003 Jul 18).
10. Sung JCY, Nichol MB, Venturini F, Bailey KL, McCombs JS, Cody M. Factors affecting patient compliance with antihyperlipidemic medications in an HMO population. Am J Manag Care 1998;4:1421-30.[Medline]
11. Avorn J, Monette J, Lacour A, et al. Persistence of use of lipid-lowering medications: a cross-national study. JAMA 1998;279:1458-62.[Abstract/Free Full Text]
12. Ellis JJ, Erickson SR, Stevenson JG, Bernstein SJ, Stiles RA, Fendrick AM. Suboptimal statin adherence and discontinuation in primary and secondary prevention populations. J Gen Intern Med 2004;19:638-45.[CrossRef][Medline]
13. Brown G, Albers J, Fisher L, et al. Regression of coronary artery disease as a result of intensive lipid-lowering therapy in men with high levels of apolipoprotein B. N Engl J Med 1990;323:1289-98.[Abstract]
14. Brown B, Zhao X, Chait A, et al. Simvastatin and niacin, antioxidant vitamins, or the combination for the prevention of coronary disease. N Engl J Med 2001;345:1583-92.[Abstract/Free Full Text]
15. McKenney J. New perspectives on the use of niacin in the treatment of lipid disorders. Arch Intern Med 2004;164:697-705.[Abstract/Free Full Text]
16. Package insert. Advicor (niacin extended-release and lovastatin tablets). Miami, FL: KOS Pharmaceuticals, 2001.
17. Andrade SE, Walker AM, Gottlieb LK, et al. Discontinuation of antihyperlipidemic drugs—do rates reported in clinical trials reflect rates in primary care settings? N Engl J Med 1995;332:1125-31.[Abstract/Free Full Text]
18. Jackevicius CA, Marndani M, Tu JV. Adherence with statin therapy in elderly patients with and without acute coronary syndromes.JAMA 2002;288:462-7.[Abstract/Free Full Text]
19. Benner JS, Glynn RJ, Mogun H, Neuman PJ, Weinstein MC, Avorn J. Long-term persistence in use of statin therapy in elderly patients.JAMA 2002;288:455-61.[Abstract/Free Full Text]
20. Dezii C. A retrospective study of persistence with single-pill combination therapy vs concurrent two-pill therapy in patients with hypertension. Manag Care 2000;9(9 suppl):2 -6.[Medline]
21. Eisen SA, Miller DK, Woodward RS, et al. The effect of prescribed daily dose frequency on patient medication compliance. Arch Intern Med 1990;150:1881-4.[Abstract]
22. International Society for Pharmacoeconomics and Outcomes Research.ISPOR Medication Compliance Special Interest Group . www.ispororg/sigs/medicationasp (accessed 2004 Dec 16).
23. Blanford L, Dans P, Ober J, Wheelock C. Analyzing variations in medication compliance related to individual drug, drug class, and prescribing physician. J Manag Care Pharm 1999;5:47-51.
24. Wei L, Wang J, Thompson P, Wong S, Struthers AD, MacDonald TM. Adherence to statin treatment and readmission of patients after myocardial infarction: a six year follow-up study. Heart 2002;88:229-33.[Abstract/Free Full Text]
25. Intercooled Stata 8.0 for Windows. College Station, TX: Stata Corporation, 2005.
26. Perwien AR, Hall J, Swensen A, Swindle R. Stimulant treatment patterns and compliance in children and adults with newly treated attentiondeficit/hyperactivity disorder. J Manag Care Pharm 2004;10:122-9.[Medline]

This Article Abstract Résumé Extracto PDF Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Articles Ahead of Print [Order Reprint] Citing Articles Citing Articles via Google Scholar Google Scholar Articles by LaFleur, J. Articles by Brixner, D. I Search for Related Content PubMed PubMed Citation Articles by LaFleur, J. Articles by Brixner, D. I