- Scanned Archives 1989 - 2007
- Donate to Us
- Contact Us
Effect of Government-Mediated Access Pricing on Availability of Directly Affected Drugs in Retail Drug Stores in the Philippines from 2009 to 2011
Jesus N. Sarol, Jr.1,2
1National Teacher Training Center for the Health Professions,
University of the Philippines Manila
2Rainiers Contract Research Services, Inc.,
M2 Strata Suites 300 P Guevarra St., San Juan City, 1500 Philippines
Availability and access to quality medicines form an important component of a country’s health service delivery. Many major diseases are treated effectively with drugs. Access to cheap quality drugs continues to be a serious problem especially among the poor in developing countries.1 Low drug availability has been reported in the Philippines.2,3
Drug stores perform a critical role in providing access to drugs to the public. Ball and Tisocki provided an estimate of 30,000 retail outlets, of which 70% were private drug stores in 2008 in the Philippines.4 Citing the Pharmaceutical and Healthcare Association of the Philippines (PHAP) report in 2008, Ball and Tisocki further related that 80% of the pharmaceutical market was channeled through private chain and independent drug stores.4 Poor households spent substantial proportions of their income on out-of-pocket drug expenditures in the Philippines.5,6
The Philippines passed a law in 2008 known as “The Universally Accessible and Quality Medicines Act of 2008” and more popularly called the “Cheaper Medicines Act of 2008.”7 This law empowered the government to regulate drug prices in order to achieve full effective competition in drug supply and demand, thus ensuring access to affordable quality drugs to its constituents. One of the most visible actions that the government has done in implementing this law is the enforcement of the maximum drug retail pricing (MDRP) and government-mediated access pricing (GMAP) policies for selected drugs. Under MDRP, the maximum price was set by the Philippine Department of Health (DOH) for drug products that carried five identified drug molecules. On the other hand, some drug companies voluntarily entered specific products for the GMAP listing wherein these products were to be sold at maximum reduced prices, usually at half of their current prices. Drug companies that distributed other drugs that carried the same molecule in the GMAP-listed drug are not required to sell their drugs at the volunteered reduced price. To effectively inform the public of these benefits, the government required drug stores to post the reduced prices of these MDRP/GMAP drugs on their premises.
The action of the Philippine government to impose MDRP and GMAP pricing constitutes a price regulation approach to control drug prices. Studies on price regulation have often looked into its effects on drug price and competition8,9 and related outcomes such as drug use, healthcare utilization, health outcomes and expenditures,8,9,10 but rarely on drug availability. Danzon and Ketcham provide scant evidence that reference pricing reduces drug availability, in particular more expensive brands, in the case of New Zealand.11 Moreover, these studies concentrated on the effects of price cap8 and reference pricing approaches8,9,10 on price regulation but not on government mediation or negotiation approaches. Under price cap regulation, a maximum price value is set under which drug stores must sell all similar drugs. Reference pricing deals with maximum reimbursements for patients’ drug expenses. Reimbursements for drug purchases are allowed only up to the stipulated reference prices for those drugs for treatment of their particular medical condition. As a mechanism for price regulation, government-mediated pricing, such as the one employed in the Philippines, works in a different way. Drug companies would offer to reduce prices of their drugs that are acceptable to the government. Other companies that distribute similar products are not required to comply with this voluntarily reduced price. However, price competition is anticipated with the consequential effect of bringing down drug prices.
Many countries in Europe also practice government price negotiation.12 However, the effect of this price regulation approach on drug availability has not been studied extensively. The effect on prices of directly affected drugs after implementation of MDRP and GMAP has been reported from the same study in another report.13 This study looked into the effect of government-mediated access pricing on drug availability in drug stores in the Philippines. Specifically, this study compared the change in availability in drug outlets of innovator brands, competitor brands, and generic versions of selected drug molecules affected by the government-mediated access pricing in 2011 using baseline levels in 2009. Differences in the trends of availability of these drugs across location and type of drug stores were also investigated. Large changes in the level of drug availability in retail stores that would occur in a considerable proportion, say greater than 25% of the concerned drugs, could be attributed to the effect of government mediated access pricing.
This study used secondary data from the IMS Health Philippines (IMS) surveys conducted in 2009 and 2011. IMS Health Philippines was commissioned by the Philippine Department of Health to conduct surveys during these years to monitor the prices and availability of drugs as part of the monitoring of the implementation of the Cheaper Medicines Act of 2008. The 2009 survey reflected baseline levels of drug price and availability while the 2011 survey was done to see changes in these indicators after the implementation of the law. The use of data from IMS was covered in a memorandum of agreement between IMS Health Philippines and Rainiers Contract Research Services Inc (RCRSI) to which the author is affiliated. Ethical approval was granted by the National Ethics Committee of the Philippine Council for Health Research and Development (PCHRD).
The procedures for the selection of drugs and drug stores were presented in greater detail in another publication.13 Briefly, a stratified sample of 600 drug stores was independently drawn by random sampling in 2009 and 2011 from the IMS Drugstore Distribution Database, a nationwide database of Philippine drug stores. Stratification was based on location and retail type. The resulting allocation of the sample by location and retail type is shown in Table 1.
This study covered 11 drug molecules. These were included because data on drug price and availability were available for these molecules in both 2009 and 2011 IMS Health surveys. The criteria for selection of priority drug molecules in 2009 was based on a score obtained by consideration of the current sales value of molecules, the DOH morbidity and mortality data, Philippine Medical Data Index Prescription Counts and Philippine National Drug Formulary (PNDF) Classification. On the other hand, the 2011 survey covered only drug molecules that were carried by drugs in the government-mediated access pricing (GMAP) list. The intersection of the sets of drug molecules in both surveys resulted in the 11 drug molecules in this study.
After the selection of drug molecules, selection of stock keeping units (SKUs) was done. SKUs refer to the different drug brands that carry a specific drug molecule. For each molecule, three SKUs were selected as follows: 1) the most saleable brand; 2) the highest priced brand; and 3) the cheapest generic product in the sample drug store. Coincidentally, this resulted in the inclusion of the same innovator and competitor brands for each molecule in both 2009 and 2011 surveys. The cheapest generic brand varied according to what was available in the sample drug store during a particular visit. Thus, this study allowed direct comparison of the availability of innovator brands and competitor brands. The innovator and competitor drugs are listed in Table 2. The cheapest generic brand for each drug molecule is not identified in the table for obvious reasons. Comparison of results of drug availability of the cheapest generic brand with that of innovator and competitor brands should be done with caution. In effect, the presence of price and availability data for the cheapest generic drug indicated merely that the drug store was selling at least one cheap generic brand of each particular drug molecule under study. The data from the study did not allow determination of trends of availability of any particular cheap generic drug brand.
Data collectors acted as mystery buyers to obtain data on drug availability (and price) of the drugs in the list from the sample drug stores.
The percentage of drug stores where each drug was available was obtained. Statistical significance of the changes in drug availability from 2009 to 2011 was assessed using Chi-square test or Fisher’s exact test. The changes (or trends) in drug availability from 2009 to 2011 were also examined across locations (island groups - Metro Manila, Luzon, Visayas, and Mindanao) and by retail type of drug store (chain or independent). To assess the differences of changes in drug availability by location and type of drug store, a logistic regression model incorporating an interaction term of these variables with year was employed. The likelihood ratio test (LRT) comparing this model to a nested model without the interaction term was used to assess the statistical significance of the interaction. The odds ratios corresponding to the interaction terms were derived by getting the exponent of the estimates of the interaction term in the model.
Further details to illuminate the interactions are presented in tables only when the interaction effects for innovator and competitor drugs were significant. The a=0.05 level of significance was used. The generic drugs were not brand-specific, that is, the cheapest generic brand was not the same across drug stores. This precluded meaningful interpretations of the interaction terms similar to that for innovator and competitor drugs. For obvious reasons, the interaction effects of location and type of drug store with year for the cheapest generic drugs are not shown in the tables.
STATA Ver 10.1 was used in generating the data analyses.14