Time filter

Source Type

McGettigan P.,William Harvey Research Institute | Henry D.,Institute for Clinical Evaluative science | Henry D.,University of Toronto | Henry D.,University of Newcastle
PLoS Medicine | Year: 2013

Background: Certain non-steroidal anti-inflammatory drugs (NSAIDs) (e.g., rofecoxib [Vioxx]) increase the risk of heart attack and stroke and should be avoided in patients at high risk of cardiovascular events. Rates of cardiovascular disease are high and rising in many low- and middle-income countries. We studied the extent to which evidence on cardiovascular risk with NSAIDs has translated into guidance and sales in 15 countries. Methods and Findings: Data on the relative risk (RR) of cardiovascular events with individual NSAIDs were derived from meta-analyses of randomised trials and controlled observational studies. Listing of individual NSAIDs on Essential Medicines Lists (EMLs) was obtained from the World Health Organization. NSAID sales or prescription data for 15 low-, middle-, and high-income countries were obtained from Intercontinental Medical Statistics Health (IMS Health) or national prescription pricing audit (in the case of England and Canada). Three drugs (rofecoxib, diclofenac, etoricoxib) ranked consistently highest in terms of cardiovascular risk compared with nonuse. Naproxen was associated with a low risk. Diclofenac was listed on 74 national EMLs, naproxen on just 27. Rofecoxib use was not documented in any country. Diclofenac and etoricoxib accounted for one-third of total NSAID usage across the 15 countries (median 33.2%, range 14.7-58.7%). This proportion did not vary between low- and high-income countries. Diclofenac was by far the most commonly used NSAID, with a market share close to that of the next three most popular drugs combined. Naproxen had an average market share of less than 10%. Conclusions: Listing of NSAIDs on national EMLs should take account of cardiovascular risk, with preference given to low risk drugs. Diclofenac has a risk very similar to rofecoxib, which was withdrawn from worldwide markets owing to cardiovascular toxicity. Diclofenac should be removed from EMLs. Please see later in the article for the Editors' Summary. © 2013 McGettigan, Henry. Source

Austin P.C.,Institute for Clinical Evaluative science | Austin P.C.,University of Toronto | Austin P.C.,Sunnybrook Research Institute
Statistics in Medicine | Year: 2014

Propensity-score matching is increasingly being used to reduce the confounding that can occur in observational studies examining the effects of treatments or interventions on outcomes. We used Monte Carlo simulations to examine the following algorithms for forming matched pairs of treated and untreated subjects: optimal matching, greedy nearest neighbor matching without replacement, and greedy nearest neighbor matching without replacement within specified caliper widths. For each of the latter two algorithms, we examined four different sub-algorithms defined by the order in which treated subjects were selected for matching to an untreated subject: lowest to highest propensity score, highest to lowest propensity score, best match first, and random order. We also examined matching with replacement. We found that (i) nearest neighbor matching induced the same balance in baseline covariates as did optimal matching; (ii) when at least some of the covariates were continuous, caliper matching tended to induce balance on baseline covariates that was at least as good as the other algorithms; (iii) caliper matching tended to result in estimates of treatment effect with less bias compared with optimal and nearest neighbor matching; (iv) optimal and nearest neighbor matching resulted in estimates of treatment effect with negligibly less variability than did caliper matching; (v) caliper matching had amongst the best performance when assessed using mean squared error; (vi) the order in which treated subjects were selected for matching had at most a modest effect on estimation; and (vii) matching with replacement did not have superior performance compared with caliper matching without replacement. © 2013 The Authors. Statistics in Medicine published by John Wiley & Sons, Ltd. Source

Austin P.C.,Institute for Clinical Evaluative science | Austin P.C.,University of Toronto
American Journal of Epidemiology | Year: 2010

Propensity-score matching is increasingly being used to estimate the effects of treatments using observational data. In many-to-one (M:1) matching on the propensity score, M untreated subjects are matched to each treated subject using the propensity score. The authors used Monte Carlo simulations to examine the effect of the choice of M on the statistical performance of matched estimators. They considered matching 1-5 untreated subjects to each treated subject using both nearest-neighbor matching and caliper matching in 96 different scenarios. Increasing the number of untreated subjects matched to each treated subject tended to increase the bias in the estimated treatment effect; conversely, increasing the number of untreated subjects matched to each treated subject decreased the sampling variability of the estimated treatment effect. Using nearest-neighbor matching, the mean squared error of the estimated treatment effect was minimized in 67.7% of the scenarios when 1:1 matching was used. Using nearest-neighbor matching or caliper matching, the mean squared error was minimized in approximately 84% of the scenarios when, at most, 2 untreated subjects were matched to each treated subject. The authors recommend that, in most settings, researchers match either 1 or 2 untreated subjects to each treated subject when using propensity-score matching. © The Author 2010. Source

Holloway K.A.,World Health Organization | Henry D.,Institute for Clinical Evaluative science | Henry D.,University of Toronto
PLoS Medicine | Year: 2014

Suboptimal medicine use is a global public health problem. For 35 years the World Health Organization (WHO) has promoted essential medicines policies to improve quality use of medicines (QUM), but evidence of their effectiveness is lacking, and uptake by countries remains low. Our objective was to determine whether WHO essential medicines policies are associated with better QUM.We compared results from independently conducted medicines use surveys in countries that did versus did not report implementation of WHO essential medicines policies. We extracted survey data on ten validated QUM indicators and 36 self-reported policy implementation variables from WHO databases for 2002–2008. We calculated the average difference (as percent) for the QUM indicators between countries reporting versus not reporting implementation of specific policies. Policies associated with positive effects were included in a regression of a composite QUM score on total numbers of implemented policies. Data were available for 56 countries. Twenty-seven policies were associated with better use of at least two percentage points. Eighteen policies were associated with significantly better use (unadjusted p<0.05), of which four were associated with positive differences of 10% or more: undergraduate training of doctors in standard treatment guidelines, undergraduate training of nurses in standard treatment guidelines, the ministry of health having a unit promoting rational use of medicines, and provision of essential medicines free at point of care to all patients. In regression analyses national wealth was positively associated with the composite QUM score and the number of policies reported as being implemented in that country. There was a positive correlation between the number of policies (out of the 27 policies with an effect size of 2% or more) that countries reported implementing and the composite QUM score (r = 0.39, 95% CI 0.14 to 0.59, p = 0.003). This correlation weakened but remained significant after inclusion of national wealth in multiple linear regression analyses. Multiple policies were more strongly associated with the QUM score in the 28 countries with gross national income per capita below the median value (US$2,333) (r = 0.43, 95% CI 0.06 to 0.69, p = 0.023) than in the 28 countries with values above the median (r = 0.22, 95% CI −0.15 to 0.56, p = 0.261). The main limitations of the study are the reliance on self-report of policy implementation and measures of medicine use from small surveys. While the data can be used to explore the association of essential medicines policies with medicine use, they cannot be used to compare or benchmark individual country performance.WHO essential medicines policies are associated with improved QUM, particularly in low-income countries.Please see later in the article for the Editors' Summary. © 2014 Holloway, Henry. Source

Hutchison B.,McMaster University | Glazier R.,Institute for Clinical Evaluative science
Health Affairs | Year: 2013

Primary care in Ontario, Canada, has undergone a series of reforms designed to improve access to care, patient and provider satisfaction, care quality, and health system efficiency and sustainability. We highlight key features of the reforms, which included patient enrollment with a primary care provider; funding for interprofessional primary care organizations; and physician reimbursement based on varying blends of fee-for-service, capitation, and pay-for-performance. With nearly 75 percent of Ontario's population now enrolled in these new models, total payments to primary care physicians increased by 32 percent between 2006 and 2010, and the proportion of Ontario primary care physicians who reported overall satisfaction with the practice of medicine rose from 76 percent in 2009 to 84 percent in 2012. However, primary care in Ontario also faces challenges. There is no meaningful performance measurement system that tracks the impact of these innovations, for example. A better system of risk adjustment is also needed in capitated plans so that groups have the incentive to take on high-need patients. Ongoing investment in these models is required despite fiscal constraints. We recommend a clearly articulated policy road map to continue the transformation. © 2013 Project HOPE-The People-to-People Health Foundation, Inc. Source

Discover hidden collaborations