Lead Investigator: Adrian Wagg, University of Alberta
Title of Proposal Research: A pooled analysis of the efficacy and tolerability of tolterodine for the treatment of Overactive Bladder (OAB) in patients over the age of 65 years of age
Vivli Data Request: 5436
Funding Source: None
Potential Conflicts of Interest: Dr. Wagg or his institution has received monies for any of research, speaker honoraria, or consultancy from Astellas Pharma, Essity Health and Hygiene AB, Pierre Fabre Medicaments, Pfizer Corp. His wife is an employee of Pfizer Canada in an unrelated business area.
Summary of the Proposed Research:
Overactive bladder (OAB) is a common bladder condition which leads to a troublesome and urgent need to visit the washroom and often leads to accidental leakage of urine. Although not a normal part of ageing, incontinence is more common in older people, and two of every three people with OAB are over the age of 65. OAB causes much embarrassment and distress, and leads to an increased risk of falls and depression. OAB is treatable, usually by managing fluid intake and by a method called bladder retraining. However, many people do not respond to this and require treatment with medicines. There are a number of these available but there is limited evidence of their effect on older people. This study will look at the effect of one medicine called tolterodine. The study will examine all of the available evidence from clinical trials which involved both older and younger patients treated with tolterodine and dummy medicines and pull it together to provide prescribing clinical staff with the best evidence for treatment and safety of tolterodine in older people. We will use a technique called pooled analysis to combine the results of several trials into one larger trial, which will allow us to make better and more accurate estimates of how well this medicine works in older people than the individual trials alone do.
Statistical Analysis Plan
The following analysis plan shall be used.
Estimating study-specific effects
For estimating the study-specific risk estimates a fixed effect method, such as logistic regression, can be used. Other fixed effects methods are also used for pooled analyses. In these methods, we assume that the true effect is same for each study and the estimates are different due to random sampling variability and the pooled effect is the weighted average of individual studies’ effects with the inverse variance as the weight.
Examining homogeneity of effects
The assumption of homogenicity is required for fixed effects methods. There are three main approaches for examining the homogeneity of the effects. The first approach is to do the test of statistical heterogeneity, like Cochran’s chi-squared test or the Q-test. Due to lack of statistical power, if the test fails, it does not necessarily mean that the studies are homogeneous. The second approach is to inspect visually to see whether the results differ only randomly. The third approach is to assess methodologic differences between studies. This can be done by dividing studies into groups and reexamining the differences between subgroups.
Estimating pooled effects
If the study-specific effects are homogeneous, then a fixed effects method would be appropriate method for estimating the summary effect in pooled analysis. If heterogeneity is found across the studies, random or mixed effect models are the appropriate methods to estimate the summary effects.
Explaining any heterogeneity
In this analysis, we will distinguish between the heterogeneity that is from the true differences in risk and the heterogeneity that is from the study designs, data collection methods and study quality. So, we should explain any heterogeneity in the study-specific effects of the pooled analysis. All aspects of the study designs, samples, methods, and quality can be considered as explanations. One approach for explaining heterogeneity is to include covariates representing features of the study and data collection methods in random effect model. Another method is to use stratified analyses.
Sensitivity analysis
Finally, a sensitivity analysis shall be performed to assess the robustness of the pooled analysis, i.e. whether the same results are obtained with different methods, and whether the quality of the studies and publication biases have influenced the summary estimates.
Requested Studies:
Tolterodine and Tamsulosin for Treatment of Men With Lower Urinary Tract Symptoms and Overactive Bladder
Data Contributor: Pfizer Inc.
Study ID: NTC00147654
Effects of tolterodine ER on patient-reported outcomes in sexually active women with overactive bladder and urgency urinary incontinence
Data Contributor: Pfizer Inc.
Study ID: NCT00143481
Pharmacia study: A6121087 Protocol ID: 98-TOCR-007
Data Contributor: Pfizer Inc.
Study ID: A6121087 – 98-TOCR-007
Summary of Results:
A Pooled Analysis of Tolterodine in Older Adults: Closing report
OAB is a highly prevalent, chronic and distressing condition defined by the International continence Society Urinary urgency, usually accompanied by in-creased daytime frequency and/or nocturia, with urinary incontinence (OAB-wet) or without (OAB-dry), in the absence of urinary tract infection or other detectable disease OAB symptoms affect different aspects of health-related quality of life such as social interactions, work productivity and relations with family members. The primary pharmacological treatments for OAB are the antimuscarinic agents, but there has until relatively recently, been a dearth of data concerning patients over the 65 and 75 years of age. Data on tolterodine relating to older people relate to a pooled analysis of data from a single clinical trial. The publication of the LUTS-FORTA guidelines began to give some weight to the evidence for OAB treatment in older people. LUTS-FORTA looked at evidence for the benefits and harms of oral therapies for OAB in older people with at least 2 concomitant diseases and ranked them according to the validated FORTA system.
This project comprises a pooled analysis of trials assessing tolterodine vs. placebo for OAB, with a specific subgroup analysis by age.
We identified three trials which included adults with OAB over the age of 65 and compared tolterodine to placebo.
Of these only two had raw data available through the Vivli Platform These data were obtained and statistical analysis was attempted.
Unfortunately, the quality of the data provided was extremely poor with large amounts of unavailable or redacted data. We had many problems in finding the variables we were looking for, as no useful information regarding variable labels had been provided, forcing us to give best guesses for some. We also realized that some of the data such as the participants’ comorbidities, baseline characteristics, and a number of potential outcome variables were not in the tables provided or replaced with XXXXX. We were therefore unable to describe the populations.