Acute and Continuation Treatment Adequacy With Venlafaxine Extended-Release Compared With Fluoxetin

Kristina S. Yu-Isenberg, Ph.D.; Christina L. Fontes, M.S.; George J. Wan, Ph.D.; Erika C. Geissler, R.N.; Ann S. M. Harada, Ph.D.


Pharmacotherapy. 2004;24(1) 

In This Article


This study examined treatment adequacy in depression management with venlafaxine XR compared with fluoxetine by using HEDIS performance indicators in a naturalistic setting within managed care. Among newly treated patients receiving only fluoxetine or only venlafaxine XR for depression, significant differences were noted in adequate dosage and continuous treatment days between fluoxetine and venlafaxine XR. Data on HEDIS-derived quality indicators of depression management were consistent with these findings; patients newly treated with venlafaxine XR were significantly more likely to receive adequate dosage and continuous trial of therapy during the acute treatment phase (84 days) and continuation phase (180 days) than were patients newly treated with fluoxetine.

Our study findings complement results of clinical trials that showed venlafaxine XR to have favorable remission[14,15] and similar safety and tolerability profiles as those of SSRIs.[16–18] This is the first study to our knowledge that used HEDIS-derived measures to compare adequate trial of therapy between venlafaxine XR and fluoxetine. Although not directly comparable owing to slight modification of the HEDIS measures, our findings were within the 95% confidence intervals of the national average for HEDIS published by the NCQA,[13] with higher acute and continuation treatment observed for venlafaxine XR in comparison to fluoxetine.

About one fourth of patients newly treated with fluoxetine or venlafaxine XR discontinued their therapy after their index fill, despite proven efficacy of antidepressant drugs. This finding is consistent with previous literature that shows high discontinuation rates of antidepressant therapy before the beneficial effects may be evident. In 1993, one group of authors[24] found that approximately one third of primary care patients with depression discontinued anti-depressant drugs within the first month of treatment. Given the high discontinuation rate, additional research is required to determine significant barriers to treatment adherence.

Previous studies reported significant shortcomings associated with the treatment of depression in primary care settings compared with treatment received from mental health specialists, such as patients receiving lower dosages of antidepressant therapy, reporting higher discontinuation of treatment, and receiving subtherapeutic dosages of anti-depressant therapy.[22,24–27] In contrast to previous studies, our data showed that physicians other than psychiatrists were 2.7 times (at 84 days) and 2.8 times (at 180 days) more likely to prescribe adequate dosages of fluoxetine or venlafaxine XR than were psychiatrists. Higher rates of adequate treatment observed in this study sample may reflect differences in severity of illness, which is associated with the recognition and start of treatment. Primary care patients with depression may be less severely ill than those seeking treatment from a psychiatrist. Also, psychiatrists may tend to treat more severely ill patients with higher dosages (as appropriate) of antidepressants. Thus, caution must be used in interpreting this finding. Although our study did not include baseline severity of depression, future studies are needed to examine the relationship between physician specialty and practice patterns to account for this baseline characteristic.

Finally, pharmacy benefit design was an important predictor of adequate dosing at 84 and 180 days. Although no significant difference was noted between open and managed pharmacy benefit design, patients with the buy-up (three-tier) design were 1.5 times more likely to be prescribed adequate dosage compared with those with a managed benefit. It is difficult to interpret this finding because of variations in copayments within the tiers, as well as differences within select markets. The goal of the three-tier model is to increase member choice of drugs while assigning a heavier financial burden for high-cost, nonformulary drugs. Therefore, although the three-tier design increases member access to therapy, it is difficult to ascertain the impact of pharmacy design on physician prescribing patterns, given that fluoxetine and venlafaxine XR were formulary drugs for most of the study period. Future studies are needed to further examine the impact of pharmacy benefit design, especially the more common three-tier model, on physician prescription patterns and behaviors.

As with all observational retrospective studies, biases or unknown confounders may have affected our study findings. First, we analyzed pharmacy refill claims, which reflect patients' actual behavior in filling prescriptions and not necessarily the physician's actual treatment regimen. Previous data show that patients' actual rates of use are often less than the dosages prescribed.[22] However, given that this phenomenon would have occurred for both the fluoxetine and venlafaxine XR groups, the differences observed in our study still would have been found. Second, treatment adequacy was examined regardless of the presence of a depression diagnosis, as diagnoses of depression are often underreported in our setting. This allowed greater specificity in patient inclusion, and the requirement for a minimum of two fills of the index antidepressant helped to reduce the numbers treated for conditions other than depression. However, patients included in our analysis may have included those receiving antidepressants for a myriad of depressive conditions, anxiety, smoking cessation, obsessive-compulsive disorder, chronic pain, and weight loss, in addition to those with a major depressive disorder. On a national basis, most patients (more than 75%) receiving a prescription for either fluoxetine or venlafaxine XR were doing so for depression with or without anxiety.[28] Therefore, the chance of bias of not including a diagnosis for depression in this analysis was much reduced. Third, the definition of adequate trial of therapy reflected the recommended dose and duration suggested by works from other published sources[13,22,23]; other definitions might reach different conclusions. The actual average daily dose for fluoxetine was found to be higher on average (25 ± 15 mg) compared with our definition of adequate daily dose of 20 mg ± 10%. Thus, our findings of the odds of adequate treatment during the acute and continuation phases are likely to be more conservative than if a higher dosage was required for adequacy. Fourth, we did not collect information on depression severity or other comorbid conditions that may have affected patient outcomes. Finally, our data were drawn from a large pharmacy benefit management organization with more than 75% of its membership from a large managed care organization with a predominantly white, elderly population in the Western United States. Thus, the results may not be generalizable to health care systems with a different structure, financial incentives, and demographic and clinical patient characteristics.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.