To the Editor:
In light of the recently published JUPITER trial, which found that patients with elevated high-sensitivity C-reactive protein levels but without hyperlipidemia can reduce their risk of cardiovascular outcomes by nearly half with rosuvastatin treatment,1 the study by Robertson and colleagues is germane and we commend the authors for their attempt to add to the debate on medication burden as a predictor of medication persistence.2 The authors conclude that medication burden does not reduce persistence with newly added lipid-lowering (LL) medications.
The results of this study should be interpreted and acted upon with caution. First, the authors exclude all patients who filled only 1 LL drug prescription during the study period without justification for this decision. Patients who do not fill more than 1 LL drug prescription may be those who are least persistent. Therefore, the study cohort consists of only those patients who already show some persistence with their LL therapy by having filled at least 2 prescriptions.
Another concern is the definition of medication burden. For a medication to count toward a patient’s cumulative medication burden, the patient must have filled prescriptions for that medication totaling at least a 90-day supply. Generally, this translates to filling at least 3 prescriptions. Thus, medication burden is defined in terms of persistence, so a patient’s persistence is likely a better predictor of medication burden rather than the reverse, which the authors were attempting to determine.
The authors find that patients who are most persistent on chronic medications continue to be persistent on a new LL drug. Among such a select group, these findings are hardly surprising, yet the authors fail to acknowledge the questionable generalizability of such results. For the clinician treating patients in the general population, the addition of an LL drug is much less straightforward. Unlike researchers with automated databases, who can view LL treatment history retrospectively, clinicians do not have the benefit of knowing which patients will fill at least 2 LL drug prescriptions at the time of treatment initiation. Furthermore, to determine persistence with other chronic medications, most clinicians can only rely on patients’ self-reports, which can be inaccurate. Therefore, we implore clinicians to use great caution in acting upon the authors’ advice that they “need not be concerned” with medication burden when adding LL therapy. Quite the contrary, the debate about the relationship between medication burden and persistence with new medications remains unresolved among the general patient population.
Jennifer M. Polinski, MPH, MS
Joshua J. Gagne, PharmD, MS
Division of Pharmacoepidemiology and Pharmacoeconomics
Brigham and Women’s Hospital
Department of Epidemiology
Harvard School of Public Health
Funding: None reported.
Author Disclosure: The authors (JMP, JJG) report no relationship or financial
interest with any entity that would pose a conflict of interest with the
subject matter in this letter.
Address correspondence to: Jennifer M. Polinksi, MPH, MS, Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women’s Hospital, 1620 Tremont St, Ste 3030, Boston, MA 02120. E-mail: firstname.lastname@example.org.
1. Ridker PM, Danielson E, Fonseca FA, et al; JUPITER Study Group. Rosuvastatin to prevent vascular events in men and women with elevated C-reactive protein. N Engl J Med. 2008;359(21):2195-2207.
2. Robertson TA, Cooke CE, Wang J, Shaya FT, Lee HY. Effect of medication burden on persistent use of lipid-lowering drugs among patients with hypertension. Am J Manag Care. 2008;14(11):710-716
Thank you to Drs. Polinski and Gagne for their comments on our article in the November 2008 issue of the Journal.1
Polinski and Gagne noted that patients who did not fill more than 1 lipid-lowering (LL) drug may be less persistent. Although we agree that these patients may have been less persistent depending on days’ supply of the first prescription, we must also point out that these patients would have the highest medication possession ratio of 1. Our study excluded 326 patients with 1 LL drug filled throughout the study period. This exclusion only affected group 2, since those in group 1 by definition had to have at least 2 prescription claims with 2 different regimens. We ran a post hoc analysis including the 326 patients in group 2 and found that medication burden was still positively associated with persistence.
We would like to clarify our definition of medication burden, which is defined as the number of unique drug products. The managed care organization allows a 90-day supply for those drugs that they define as chronic drug therapies. Any prescription for a chronic medication was counted in our study, regardless of the days’ supply as long as it overlapped with the index date.
Our study may not apply to all populations. However, the clinician is often faced with adding additional therapies to existing regimens. In patients with hypertension with an average medication burden of 2.9 chronic medications, adding LL therapy did not affect persistence.
Teisha A. Robertson, PharmD, MBA
Catherine E. Cooke, PharmD, BCPS
Jingshu J. Wang, PhD
Fadia T. Shaya, PhD, MPH
School of Pharmacy
University of Maryland
Helen Y. Lee, PharmD, MBA
CareFirst BlueCross BlueShield
Funding Source: None reported.
Author Disclosure: The authors (TAR, JW, FTS, HYL) report no relationship
or financial interest with any entity that would pose a conflict of interest with the subject matter of this article. Dr Cooke is a former employee of Pfizer, Inc, and reports owning stock in that company.
Address correspondence to: Teisha A. Robertson, PharmD, MBA, PO Box 6893, Largo, MD 20774. E-mail: email@example.com.
1. Robertson TA, Cooke CE, Wang J, Shaya FT, Lee HY. Effect of medication burden on persistent use of lipid-lowering drugs among patients with hypertension. Am J Manag Care. 2008;14(11):710-716.