Effect of the economic recession on pharmaceutical policy and medicine sales in eight European countries

Abstract Objective To identify pharmaceutical policy changes during the economic recession in eight European countries and to determine whether policy measures resulted in lower sales of, and less expenditure on, pharmaceuticals. Methods Information on pharmaceutical policy changes between 2008 and 2011 in eight European countries was obtained from publications and pharmaceutical policy databases. Data on the volume and value of the quarterly sales of products between 2006 and 2011 in the 10 highest-selling therapeutic classes in each country were obtained from a pharmaceutical market research database. We compared these indicators in economically stable countries; Austria, Estonia and Finland, to those in economically less stable countries, Greece, Ireland, Portugal, Slovakia and Spain. Findings Economically stable countries implemented two to seven policy changes each, whereas less stable countries implemented 10 to 22 each. Of the 88 policy changes identified, 33 occurred in 2010 and 40 in 2011. They involved changing out-of-pocket payments for patients in 16 cases, price mark-up schemes in 13 and price cuts in 11. Sales volumes increased moderately in all countries except Greece and Portugal, which experienced slight declines after 2009. Sales values decreased in both groups of countries, but fell more in less stable countries. Conclusion Less economically stable countries implemented more pharmaceutical policy changes during the recession than economically stable countries. Unexpectedly, pharmaceutical sales volumes increased in almost all countries, whereas sales values declined, especially in less stable countries.


Introduction
European public authorities struggle to maintain a high level of access to health care while restraining increases in expenditure associated with an ageing population and higher demand. [1][2][3][4] The recent global economic recession has put additional pressure on public budgets. 5,6 In 2008, Europe was affected by the financial crisis. As the recession in Europe continued, the effect was felt especially in southern European countries and Ireland in 2010 and 2011. Soon the problem of financial debt for individual European countries developed into a crisis in the Eurozone, which then became a high priority for the European Central Bank and the European Parliament. All countries were urged to implement cost-saving measures that affected public financing for health care. 7 Recession, which is defined as two successive quarters of negative growth in gross domestic product (GDP), can have a detrimental effect on the health of the population because economic downturns are strongly associated with a decline in health-care utilization and a deterioration in health outcomes. 8 For example, suicides and homicides increased among working-age men and women when unemployment rose rapidly during past recessions in Europe. 9 In the current recession, the number of uninsured non-elderly Americans increased by 5.6 million between 2007 and 2009 10 and over a quarter of Americans reported reduced routine use of medi-cal care. 11 Over the same period, insurance policy deductibles and copayments for visits to physicians and for prescription medicines increased, leading to a greater cost burden for patients. [12][13][14] Similar effects were seen in Greece. Studying the health effects of the economic crisis in the country it was found that patients had less access to care and preventive services and, consequently, faced higher risks of infection with sexually transmitted diseases. 15 The World Health Organization examined the influence of the recession on expenditure on, and the sales and prices of, medicines between 2007 and 2009 in 84 countries. It found that the economic recession had mixed effects and that the largest declines in medicine sales occurred in high-income countries and in Europe, particularly in the Baltic states. 16 It has been shown that countries that were seriously affected by the crisis, such as the Baltic countries, Greece, Portugal and Spain, abruptly implemented several pharmaceutical policy measures between 2010 and 2011. This included price cuts, changes in reimbursement rates and the imposition of value-added tax on medicines. 17 In other European countries, such as Italy, in which cost-containment measures were already in place when the crisis began, the implementation of planned policy changes was accelerated. 18 Because different countries were affected differently by the recession and attempted to overcome budgetary constraints in different ways, we decided to analyse systematically how European pharmaceutical policies were affected by the reces-Objective To identify pharmaceutical policy changes during the economic recession in eight European countries and to determine whether policy measures resulted in lower sales of, and less expenditure on, pharmaceuticals. Methods Information on pharmaceutical policy changes between 2008 and 2011 in eight European countries was obtained from publications and pharmaceutical policy databases. Data on the volume and value of the quarterly sales of products between 2006 and 2011 in the 10 highest-selling therapeutic classes in each country were obtained from a pharmaceutical market research database. We compared these indicators in economically stable countries; Austria, Estonia and Finland, to those in economically less stable countries, Greece, Ireland, Portugal, Slovakia and Spain. Findings Economically stable countries implemented two to seven policy changes each, whereas less stable countries implemented 10 to 22 each. Of the 88 policy changes identified, 33 occurred in 2010 and 40 in 2011. They involved changing out-of-pocket payments for patients in 16 cases, price mark-up schemes in 13 and price cuts in 11. Sales volumes increased moderately in all countries except Greece and Portugal, which experienced slight declines after 2009. Sales values decreased in both groups of countries, but fell more in less stable countries. Conclusion Less economically stable countries implemented more pharmaceutical policy changes during the recession than economically stable countries. Unexpectedly, pharmaceutical sales volumes increased in almost all countries, whereas sales values declined, especially in less stable countries.

Research
Economic recession and pharmaceutical policies sion by comparing changes in pharmaceutical pricing and reimbursement policies between economically stable and economically less stable countries. In addition, we investigated changes in the sale of pharmaceuticals in major therapeutic classes before and after the recession in these two types of countries. We expected that some of the cost-containment policies, such as those affecting out-of-pocket payments, would shift the financial burden of medicines onto patients and hypothesized that pharmaceutical sales would decline during this period, especially in less economically stable countries.

Data sources
For this longitudinal study, we used data from two sources to derive information on pharmaceutical policies: (i) the Pharmaceutical Pricing and Reimbursement Information Network (Austrian Health Institute, Vienna, Austria), which collects information from experts in national pharmaceutical pricing and from authorities responsible for reimbursement -the latter provide regular pharmaceutical policy updates; and (ii) the PharmaQuery database (IMS Health, Philadelphia, United States of America), which contains data on pharmaceutical policies. In addition, we included information on policy changes reported in the published literature. We grouped policy changes into 6-month implementation periods from January 2008 until December 2011 and we categorized policy as relating to one of three main areas: (i) pricing; (ii) reimbursement; and (iii) generic drugs. Table 1 defines the policy measures in these three areas.
Quarterly pharmaceutical sales data for the period January 2006 to December 2011 were obtained from the IMS MIDAS (Multinational Integrated Data Analysis System) Quantum pharmaceutical market research service (IMS Health, Philadelphia, USA). Data were expressed in standard units for the volume of sales and in constant United States dollars (US$) for the value of sales. A standard unit, as defined by IMS Health, is the smallest dose of a product -it may be one tablet or capsule for oral preparations, one teaspoon (i.e. 5 mL) for a syrup or one ampoule or vial for an injectable product. The value of sales was derived from the price deemed most accurate for the relevant country and was expressed in constant US$, which were calculated by converting the local currency into United States dollars at a constant exchange rate. In most countries, the price used was the ex-factory price; in Estonia, Finland, Greece and Ireland, ex-factory prices were derived from wholesale prices. Average standard conversion factors, which were determined with the co-operation of the pharmaceutical industry for each country, were applied to estimate prices at various points along the distribution chain. The price calculations did not take into account any discounts between manufacturers, wholesalers and payers and were not adjusted for inflation.
Our study considered only prescription medicines, whether on or off patent, that were available in the retail market for the 10 highest-selling therapeutic classes. We identified the 10 highestselling classes by ranking therapeutic classes according to their sales volume in each country. Together the combined sales volume of products in these 10 classes accounted for at least 50% of the total sales volume of all medicines in each of the eight countries from 2008 to 2011 (Table 2). Data were aggregated by therapeutic class for each country. We had no data on individual drugs.

Country groups
We considered eight European countries in which the majority of the population was covered by a social security system or national health service: Austria, Estonia, Finland, Greece, Ireland, Portugal, Slovakia and Spain. We selected these countries because they represented a variety of geographical regions and levels of economic wealth and stability and had been affected by the recession to different degrees. We classified them as either economically less stable or economically stable using categories defined by the Organisation for Economic Cooperation and Development (OECD) for the level of fiscal consolidation in 2012. Fiscal consolidation was judged according to whether the country had adopted either concrete policies aimed at stabilizing general government gross debt or a long-term target for the debtto-GDP ratio of 60%. There were four categories of country: (i) those that had adopted a programme proposed by the International Monetary Fund, the European Union and the European Commission (e.g. Greece, Ireland and Portugal); (ii) those that were under clear market pressure (e.g. Belgium, Hungary, Italy, Slovakia and Spain); (iii) those that had a substantial deficit or debt but which were under less market pressure (e.g. Austria, Denmark, Finland, France and Germany); and (iv) those that had no or only a marginal need for consolidation (e.g. Norway, Sweden and Switzerland). 21 In this study, we regarded economically less stable countries as those belonging to the first two categories (i.e. Greece, Ireland, Portugal, Slovakia and Spain) and economically stable countries as those belonging to the third and fourth categories (i.e. Austria, Estonia and Finland).

Data analysis
First, we described and analysed the number of policy measures implemented per year, per country group and per policy category. Next, we determined the volume and value of the sales of drugs in each therapeutic class between 2006 and 2011 in each country and, then, we calculated the combined volume and value of the sales of drugs for all 10 therapeutic classes for each country. Since our findings for individual therapeutic classes and for all therapeutic classes combined were similar, we present only the results for all therapeutic classes combined.
For this analysis, we divided the volume and value of sales by the size of the country's population to control for population growth; annual population figures were obtained from the OECD. 22 We derived the annual and average growth rates over the study period using both the volume and value of pharmaceutical sales per capita: where AGR is the annual growth rate, S y is the per capital sales in a year, S y-1 is the per capital sales in the previous year, AAGR is the average annual growth rate and n is the number of years.
To compare changes in the volume and value of sales, we calculated the dif-

Pricing
Price cut A cost-containment measure whereby the set price of a medicine is reduced by the authorities. External price referencing External price referencing is the practice whereby the price of a medicine in one or several other countries is used to derive a benchmark or reference price for the purpose of setting or negotiating the medicine's price in a given country. Policy changes in external price referencing include the introduction or abolition of this pricing policy and altering the methodology (e.g. changing the basket of reference countries or the way of calculating the benchmark price). Distribution remuneration (i.e. mark-ups, margins and fees for service) Distribution remuneration is the payment of a health-care provider, whether an individual or an organization, for the services provided. In the distribution of pharmaceuticals, wholesalers and pharmacies are remunerated using mark-ups or regressive margin schemes or, for pharmacies alone, by paying a "fee for service". With mark-ups, a defined linear or percentage amount is added to the cost of a good to ensure a profit at the wholesale or retail level or both. With regressive margin schemes, the margin is expressed as a percentage of the selling price. Policy changes in distribution remuneration include adjusting the mark-ups or margins used for wholesalers or pharmacies or changing the type of distribution remuneration for a defined actor. Changes may also be made to the types of medicines (e.g. reimbursable medicines or prescription-only medicines) to which distribution remuneration applies. VAT on medicine VAT is a sales tax on products that is collected in stages. It is a wide-ranging tax that is usually designed to cover most or all goods and services, including medicines. Policy changes in VAT include the introduction or abolition of VAT on medicines and altering the VAT rate on medicines. Extraordinary price review Price reviews involve reviewing the process by which the set price of a medicine was established. Reviews may or may not be performed in combination with reimbursement reviews. Reviews can be performed systematically (e.g. once a year) for all reimbursed medicines or for a group of medicines (e.g. for a specific indication) or at any time. Reimbursement

Reference price system
With a reference price system, which is also referred to as internal or therapeutic reference pricing, the third party payer determines a reference price for the reimbursement of medicines with a particular active ingredient or in a given therapeutic class. If the price of the medicine exceeds the reference price, the health-care consumer must pay the difference between the fixed reimbursed amount (i.e. the reference price) and the actual pharmacy retail price in addition to any copayments (e.g. prescription costs and percentage copayment rates). Policy changes in the reference price system include the introduction or abolition of a reference price system and changing the methodology by which clusters of medicines are established for determining a reference price (e.g. by grouping identical or similar medicines). Out-of-pocket payments Out-of-pocket payments are payments made by health-care consumers that are not reimbursed by a third-party payer. They include cost-sharing, fixed or percentage copayments and informal payments to health-care providers. Delisting Delisting is the exclusion of a medicine from a reimbursement list (e.g. a positive list), which often results in exclusion from reimbursement.

Generic drugs INN prescribing
With INN prescribing, prescribers (e.g. physicians) are required to prescribe medicines using the INN for the pharmaceutical (i.e. the name of the active ingredient) instead of a brand name. Policy changes in INN prescribing include its introduction or abolition, changing the way INN prescribing is organized (e.g. by imposing or eliminating financial incentives) and changing from indicative to obligatory INN prescribing. Generic substitution Generic substitution is the practice of substituting a medicine, whether marketed under a trade name or generic name (i.e. a branded or unbranded drug), by a less expensive medicine (e.g. a branded or unbranded generic drug), which often contains the same active ingredients. Generic substitution may be encouraged (i.e. indicative generic substitution) or required (i.e. mandatory generic substitution). Policy changes in generic substitution include its introduction or abolition, changing the way generic substitution is organized (i.e. imposing or eliminating financial incentives) and moving from indicative to obligatory generic substitution. Public campaigns Policies, regulations, measures and initiatives promoting the use of generic drugs or licensed, offpatent medicines are typically undertaken by government authorities. Policy on generic drugs may be targeted at prescribers, pharmacists, patients or consumers, or other stakeholders.
INN: international nonproprietary name; VAT: value-added tax.

Research
Economic recession and pharmaceutical policies ference between the annual growth rate in the value of pharmaceutical sales and the annual growth rate in the volume of sales for each country. Table 3, Table 4 and

Changes in sales
The small increase in the volume of pharmaceutical sales in all countries between 2006 and 2011 is shown in Fig. 1, Fig. 2, Fig. 3, Fig. 4 and    In general, between 2006 and 2008, the annual value of pharmaceutical sales increased more than the annual volume of sales in both economically stable and less stable countries, which indicates that the average price per unit increased. From 2009 onwards, during the period when most policy changes were implemented, the growth in the annual value of sales was less than the growth in the annual volume, which indicates a decrease in average price per unit.

Discussion
Although countries adjust their pharmaceutical policy frameworks continuously, a surge of policy changes seems to have taken place during the economic recession, particularly in 2010 and 2011. Unexpectedly, both economically stable and economically less stable countries

Research
Economic recession and pharmaceutical policies experienced a slight increase in the consumption of pharmaceuticals in the 10 highest-selling therapeutic classes, as measured in standard units per capita. As expected, the annual growth in the per capita value of medicine sales decreased in economically less stable countries in 2010 and 2011.
Our study shows that economically stable countries implemented fewer policy measures between 2008 and 2011 than economically less stable countries. The most frequently implemented policy changes targeted out-of-pocket payments for patients. Previous studies have shown that increases in copayments, such as prescription fees, tend to lead to lower medicine utilization, especially in times of economic recession and increased unemployment. [23][24][25][26][27][28][29][30] Policy measures such as the medicine price cuts (also applied in the form of discounts) that were implemented in Greece, Portugal and Spain could have had a negative effect on the availability of medicines if they caused pharmaceutical companies to withdraw their products from national reimbursement lists. 31 Contrary to our expectations, however, we did not observe a major decline in the consumption of pharmaceuticals during the recession in the therapeutic categories studied as most countries continued to experience a moderate positive annual growth in sales volume. However, in line with media reports of drug shortages in Greece and Portugal, our data showed that the sales volumes of important medicines for chronic diseases, such as angiotensin-converting enzyme inhibitors and antidepressants, dropped drastically in these two countries in 2010. 31 Hence, although the overall growth in sales volume was positive, the rate of growth appears to have fallen to below the prerecession level, which ranged from 5% to 12%.
In contrast, the rate of growth in the value of pharmaceutical sales declined, especially in economically less stable countries. This decrease may have been due partly to inflation: the average inflation rate in 2010 and 2011 generally ranged between 2.0% and 3.4%, although it was as low as −1.6% in 2010 in Greece and as high as 5.1% in 2011 in Estonia. 32 Our analysis did not take inflation into account. The decrease may also have occurred because the  The case of Estonia needs to be discussed separately. After a decade of rapid growth before the recession, during which public sector expenditure grew 6.5 times, 21 Estonia experienced a major decline in GDP in 2009. Public sector spending was cut by 6.6% -a reduction of 100 million euros compared with 2008 -and there was a 50 million euros reduction in health insurance expenditure. 33 A previous study identified a large decline in the consumption of pharmaceuticals of −18% between 2008 and 2009, 16 which was mirrored in our data. In response, Estonia implemented strict cost-saving measures with respect to medicines, reduced sick leave coverage and increased the workload of clinical staff without increasing their salaries. 16,33 Our data show that, by 2010, the consumption of pharmaceuticals had returned to a level similar to that before the recession, which paralleled Estonia's relatively quick recovery from the recession overall. 21 Early in the recession, countries not only implemented few policies changes overall but also implemented no policies that targeted consumption by specific patient groups or in specific therapeutic areas. Recent studies show that fewer policy changes were implemented in 2012 and 2013 than during the recession and that there was a trend towards policies that targeted high-cost medicines. 34 Several countries have explored alternative policies for sharing the financial risk of selected, new, high-cost medicines, such as value-based pricing models or risk-sharing agreements. [35][36][37][38] The effect of these new approaches still needs to be determined.
Our study had several limitations. We did not take into account the differences in pharmaceutical policy frameworks that existed between all countries before the economic recession or between regions within some countries (e.g. Italy or Spain). Moreover, it was not always clear whether a country implemented a policy as a short-term reaction to recession-related budgetary constraints or whether the policy was part of a planned long-term change to the system. For instance, in Finland, the implementation of internal refer-ence pricing in 2009 had been planned long before the recession. 39 Major policy changes, such as the introduction of reference price systems, may take several years to implement since many stakeholders are involved. 40 However, most of the policy changes related to the recession involved adjusting existing policies and could be implemented relatively quickly. Although these policies might, as desired, contain costs over the shortterm, they could have substantial longterm effects on the use and affordability of medicines and could have negative consequences for health. [41][42][43][44] We focused our analysis on the sale of products that accounted for the majority of pharmaceutical sales by volume. It is possible that policy changes had a differential influence on the sale of less frequently used products, including those used by patients with rare diseases. However, at least one price cut in Greece exempted orphan drugs for rare diseases.
Another limitation is that, since data on the value of sales are expressed in constant US$ and disregard discounts and rebates, they do not reflect actual spending by third-party payers. In addition, an individual country's data might include different products within each therapeutic class. Moreover, medicines sales may also be influenced by other market variables, such as patent expiration. During the study period, patents expired on several highly used products, including diabetes medications, antiulcerants, platelet aggregation inhibitors, lipid regulators, angiotensin-converting enzyme inhibitors and antidepressants. The price reductions accompanying patent expiries may have combined with policies promoting generic prescribing to reduce the value of sales while limiting the decline in sales volume. Finally, the rapid implementation and the different timing of policies in different countries meant that we were unable to attribute the observed change in pharmaceutical sales to any single policy or set of policies or to make statistical comparisons of responses to policy between countries.
We suggest that future research focuses on the effect of policy changes in only a few countries by exploring the relationship between changes in medicine utilization and health outcomes. Moreover, since some of our findings were not in line with our expectations, we recommend that studies of the effect

Research
Economic recession and pharmaceutical policies of new pharmaceutical policies should monitor access to medicines and look for potential barriers to access.
In conclusion, the ways in which countries responded to the recession differed greatly, with less economically stable countries implementing a larger number of policies that affected the pharmaceutical sector than economically stable countries. Our evidence shows that, despite numerous policy changes and contrary to our expectations, overall consumption of pharmaceuticals in the 10 highest-selling therapeutic classes continued to increase in most countries; there was no clear difference between economically stable and less stable countries. The observation that the value of sales declined while the volume was maintained may indicate that pharmaceutical purchasing became more efficient. However, since many policies were designed to shift the financial burden to patients, future research should investigate the effect of changes in pharmaceutical policy, expenditure and utilization on equitable access to medicines, on the affordability of essential medicines for households, on the appropriate use of medicines and on health outcomes. ■ C Leopold was a PhD candidate at Utrecht University and a visiting scholar at the Department of Population Medicine that resides within the Harvard Pilgrim Health Care Institute and is an appointing department of Harvard Medical School.