Patented drug prices and clinical trials in 31 OECD countries 2017: implications for Canada’s PMPRB
Brett J Skinner, Ph.D.
ABSTRACT
The federal government is implementing changes to the rules used by the Patented Medicine Prices Review Board (PMRPB) to regulate the prices of new medicines sold in Canada. The regulatory changes are intended to further depress the prices of patented drugs. In its Regulatory Impact Analysis Statement, the PMPRB stated: “It is not anticipated that these amendments would generate adverse impacts on industry employment or investment in the Canadian economy.” This study examined whether there is any empirical link between price and industry investment in clinical R&D. The analysis tested for statistical correlations between the geographic distribution of industry-funded clinical trials across 31 OECD countries and variation in drug price levels, controlling for differences in GDP and market size. A multi-variable regression analysis suggested that a lower price ceiling resulting from the PMPRB regulatory changes will likely cause a substantial decline in the number of industry-funded clinical trials in Canada. Analysis of other data suggest clinical trial activity could already be declining over time and the trend is linked to a simultaneous decline in the Canadian price level for patented medicines relative to competing markets in the 7 reference countries currently used by the PMPRB.
Publication Date: August 2019
CITATION: Skinner, Brett J. Patented drug prices and clinical trials in 31 OECD countries 2017: implications for Canada’s PMPRB. Canadian Health Policy, August 2019. Canadian Health Policy Institute (CHPI). www.canadianhealthpolicy.com.
[Download PDF to access all exhibits.]
INTRODUCTION
The federal government is implementing changes to the rules used by the Patented Medicine Prices Review Board (PMRPB) to regulate the prices of new medicines sold in Canada. The regulatory changes are intended to further depress the prices of patented drugs. In its Regulatory Impact Analysis Statement, the PMPRB stated:
“It is not anticipated that these amendments would generate adverse impacts on industry employment or investment in the Canadian economy. Although when the current regulatory framework was first conceived 30 years ago, policy makers believed that patent protection and price were key drivers of medicine research and development (R&D) investment, there is no evidence of this link. The level of industry R&D investment relative to sales by medicine patentees in Canada has been falling since the late 1990s and is now at a historic low despite Canada having among the highest patented medicine prices in the world. These amendments would aim to align Canadian prices with those in countries that, despite having lower prices, receive higher medicine industry investment.”
The objective of this study is to examine whether there is any empirical link between price and industry investment in clinical R&D. The analysis tests for statistical correlations between the geographic distribution of industry-funded clinical trials across OECD countries and variation in drug price levels, controlling for differences in GDP and market size. Policy implications for Canada are briefly discussed with reference to pending changes to federal drug price regulations by the PMPRB.
DATA
Table 1 displays the data that were collected for this study, sorted alphabetically by country. All data cover the calendar year 2017, which was the most recent year available that was common across all variables. The number of countries included for analysis was determined by the availability of comparable pricing data. Proxy data were available from the PMPRB that compared average gross prices (ex-factory list prices before rebates) in 2017 for patented (defined by single-source brands in Canada) drugs across 31 Organisation for Economic Co-operation and Development (OECD) countries in bi-lateral ratios of foreign-to-Canadian prices (Variable label “F:C IPRX$”, Canada = 1.00). The PMPRB uses Canadian and international prices reported in the IQVIA MIDAS™ database at the ex-factory manufacturer level, reflecting all sales to the pharmacy and hospital sectors in all OECD countries with available MIDAS™ data. The average foreign-to-Canadian price ratios are constructed using Canadian sales-weighted arithmetic averages of the corresponding foreign-to-Canadian price ratios for individual medicines and adjusted for the Purchasing Power Parities (PPP) of international currencies.
Data on the number of industry-funded clinical trials (Variable label “IND CTs”) in the 31 countries matching the PMPRB’s patented drug price-level data were obtained from ClinicalTrials.gov, an online database of clinical trial activity around the world operated by the U.S. National Library of Medicine (NLM), National Institutes of Health (NIH) since 2008. The database query specified all industry-funded clinical trials with a registered start date from 01/01/2017 to 12/31/2017. The query returned results on 6674 registered clinical trials across 100,691 sites in 101 countries. The data were then sorted and collapsed to include only the 5,251 registered clinical trials in the 31 OECD countries matching the PMPRB data.
Corresponding 2017 data on Gross Domestic Product (GDP) per capita [Variable label “GDP percap”] and total national population in thousands [Variable label “POP (‘000s)”] in each of the 31 countries were obtained from the OECD online database.
ANALYSIS
A statistical analysis was conducted to test for correlations between variation across countries in the number of industry-funded clinical trials (IND CTs) and differences in prices (F:C IPRX$), economies (GDP percap) and market size (POP ‘000s). Charts 1-3 show scatter plots illustrating the positive (i.e. direct) statistical relationship between the dependent variable (IND CTs) and each of the independent variables (F:C IPRX$, GDP percap, POP ‘000s). However, the results suggest that only price and market size have correlations of statistical significance (R21.000).
Table 2 displays the results of a multi-variable regression analysis to test the statistical significance of the set of independent variables (model) as a predictor of the dependent variable; and the correlations between the dependent variable and each independent variable, controlling for the other independent variables in the model. Statistically significant results (Adj.RSq.1.000; Sig. F<0.050; P<0.050) are highlighted, as are the regression coefficients showing the predicted variation in the dependent variable associated with a 1-unit change in each independent variable.
As a set, the independent variables in the model were a statistically significant (Sig. F=0.000) predictor of the dependent variable, explaining almost 90% (Adj.RSq.=0.897) of the variation in the number of industry-funded clinical trials between countries.
The analysis of the correlations between the dependent variable and each independent variable, showed that only price (P=0.000) and market size (P=0.004) remained statistically significant predictors of the dependent variable after controlling for the other independent variables in the model. The coefficient results show that a 1-unit variation (+/-1.00) in the average foreign-to-Canadian price ratio for patented medicines is associated with a variation of +/-613.355 industry-funded clinical trials. Similarly, the coefficient results show that a 1-unit variation (+/-1,000) in market size (defined by the population stated in thousands) is associated with a variation of +/-0.002 industry-funded clinical trials (or 2 trials per million population variation in market size).
POLICY IMPLICATIONS
Industry decisions about where to invest in pharmaceutical R&D are driven by a complex set of variables. This study provides empirical evidence of a link between the price level for patented medicines and the number of industry-funded clinical trials across economically comparable countries. The results suggest that a lower price ceiling resulting from the PMPRB regulatory changes will likely cause a substantial decline in the number of industry-funded clinical trials in Canada. Other data from the PMPRB suggest clinical trial activity could already be declining over time and that this is linked to a simultaneous decline in the Canadian price level for patented medicines relative to competing markets.
An analysis of data published in the PMPRB’s annual reports confirmed that Canadian gross (i.e. list) prices for patented medicines have been declining relative to international prices in the PMPRB7 countries for the last 11 years. Table 2 shows PMPRB data comparing Canadian prices for patented medicines to international prices averaged across the PMPRB7 countries and stated as foreign-to-Canadian (F:C) ratios. The PMPRB publishes median and mean figures for international prices adjusted for market exchange rates (MER) and purchasing power parities (PPP) of currencies. Table 2 also shows the inverse equivalent Canadian-to-foreign (C:F) ratios (C:F = 1/F:C). Measured at median prices adjusted for MER, the C:F price ratio fell from .97:1 in 2007 to .79:1 in 2017. Measured at mean prices adjusted for PPP, the C:F price ratio fell from .96:1 in 2007 to .65:1 in 2017. [CHART 4]
PMPRB data also show [CHART 5] that from 2007 to 2017 spending on qualifying pharmaceutical R&D in Canada fell from over $1.325 billion to $791.1 million. , The decline of spending on R&D coincides with a deteriorating Canadian price level relative to competing markets in the PMPRB7 countries. To illustrate this, Chart 6 plots R&D spending and C:F price ratios on the same graph using two vertical axes.
It is important to note that there is a statistical lag effect linking industry-funded R&D to the number of clinical trials. Canada accounted for less than 2% (US$22.2 billion sales) of the global pharmaceuticals (patented and non-patented) market (US$1,204.8 billion sales) by 2018, yet since 2008 almost 10% (10,080) of the cumulative global total number of industry-funded clinical trials (103,352) have been located in Canada. , The country’s relative success attracting industry-funded clinical trials is a legacy from earlier investment decisions when Canada’s price level was higher relative to the PMPRB7 comparator countries. A statistical lag between data for R&D spending and the number of clinical trials occurs because clinical trials can take more than 6 years to complete. Declining recent (2008 to 2017) trends in industry investment in pharmaceutical R&D will be reflected in future statistics which should be expected to show a subsequent parallel decline in the number of industry-funded clinical trials located in Canada. The federal government’s pending PMPRB regulatory changes will probably exacerbate this trend.
