 |
 |
 |
 |
 |
 |
 |
|
|
|
|||||||||
|
|||||||||||
Vice President Policy Analysis, Inc. 4 Davis Court Brookline, Massachusetts 02445-7629 fax 617/232-1155 goster{at}pai2.com
Senior Analyst Policy Analysis, Inc.
Executive Director World Wide Health Economics and Outcomes Research Purdue Pharma, LP Stanford, Connecticut
Published Online, June 3, 2004. www.theannals.com, DOI 10.1345/aph.1D109b
The explicit purpose of our study was to ascertain whether there are differences in rates of therapy switching between patients who begin long-acting opioid treatment with CR oxycodone, transdermal fentanyl, or CR morphine sulfate. We examined this issue because therapy switching can be an important indicator of therapy failure. Furthermore, in numerous other therapeutic areas, significant differences in rates of therapy switching have been reported in relation to the medications that patients receive.1-3 Because therapy switching had not yet been examined in patients receiving long-acting opioids, we explored this issue. As we reported, patients who received CR morphine sulfate had consistently higher rates of therapy switching than those receiving either CR oxycodone or transdermal fentanyl. This phenomenon was observed in patients with cancer as well as those without, and after controlling for age, gender, selected comorbidities, and pretreatment healthcare charges in multivariate analyses.
Because therapy switching has been reported to be associated with elevated levels of utilization and costs in other therapeutic areas (eg, hypertension, depression, schizophrenia),1-3 we also compared the healthcare charges of patients who switched to long-acting opioid therapy with those who did not (irrespective of the agent received). Contrary to what Dr. Jaderholm suggests, we did not compare healthcare charges between agents (ie, CR oxycodone vs transdermal fentanyl vs CR morphine sulfate). Our finding, which persisted following adjustment in multivariate analysis for covariates such as age, preexisting diagnoses (eg, chronic renal failure, coronary heart disease), and pretreatment healthcare charges, is consistent with prior research—therapy switching is associated with higher costs. We also note quite explicitly that causality could not be established, commenting that "it does not necessarily follow that therapy switching is the cause of...higher charges. Causation may run in the opposite direction."4
We are in complete agreement with Dr. Jaderholm that all 3 study agents have shown efficacy in the treatment of moderate to severe malignant and nonmalignant pain. We did not examine the need for rescue therapy or rates of mortality, as these issues were outside the scope of our investigation. We also fail to see a connection between the consequences of a "broken" CR mechanism and the findings of our study. While breaking or crushing a CR delivery mechanism can result in overdosing, this applies equally to CR oxycodone and long-acting morphine products.5,6
Finally, our study was indeed funded by Purdue Pharma (as disclosed), and 3 of the authors are employed by a contract research organization. The relevance of this to questions concerning the appropriateness of study methods and the validity of findings is not apparent.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
 |
 |
 |
 |
