Numerous studies have shown substantial variation in hospital expenditure across OECD countries. However, there is only limited understanding of the potential sources of this variation. Explaining hospital (or more broadly health care) expenditure in terms of volumes and prices requires expenditure data to be converted using a common currency. Purchasing Power Parities (PPPs) are commonly used to convert national currencies to a common unit. One of the main advantages of using PPPs is that they are both currency converters and price deflators. Therefore, when PPPs are used to convert expenditure to a common unit, the results are valued at a uniform price level and should reflect only differences in the volumes of goods and services consumed in countries.
The OECD and Eurostat have compiled PPPs for a number of years. PPPs are usually calculated for products groups (rather than individual products), which are then aggregated for the entire economy, and are termed GDP PPPs. Using GDP PPPs for converting health products and services to a common unit has been criticised in the literature. The primary reason is because GDP PPPs do not take account of the relative price levels of the health care sector compared with other sectors in the economy. An alternative to GDP PPPs is to use health-specific PPPs. These have also been developed, although the data to calculate these have been based on the prices of health care inputs for hospital services. Typically, this is dominated by the wage rates of health care providers such as doctors and nurses. The use of input prices to calculate hospital PPPs breaks with conventional methods, which are generally based on the prices of outputs. The main critique for using input based prices is that it ignores the relative productivity of the inputs (in this case labour). For these reasons, the OECD and Eurostat have embarked on a project to calculate PPPs on the basis of hospital output prices.
An OECD Task Force on Health PPPs was set up in 2007 and established the methodological principles for the project. Since then, five pilot studies have been carried out between 2008 and 2012 which have confirmed the feasibility and the general suitability of the methodology. The methodology is intentionally flexible to allow for some definitional differences in the data collected across countries. Hospital-specific output-based PPPs require data on the quantity of outputs produced as well as the price of these outputs. This may sound relatively simple but there are a number of complicating factors. First, hospitals produce many different types of outputs and the quantities and prices for each of those products are likely to vary country by country. This implies that the data needs to be provided at the product level. Second, the products need to be comparable across hospitals and countries. Finally, given that hospitals frequently operate outside of the market place, price data may not be easily obtained. Even when they are available, there may be issues of comparability across countries.
The basic principle for price comparisons is that product types should be comparable and representative. This means that the hospital products should be similar from a clinical perspective across countries. With the advent of output-based hospital funding, it has become feasible to define hospital products. The OECD Task Force on Health PPPs identified a set of criteria to select representative and comparable hospital products. Hospital product types were selected on the basis that they: (i) represent common procedures or diagnoses; (ii) account for a significant percentage of hospital expenditures; (iii) represent procedures which are likely to be the principal procedure within one hospitalisation (for surgical hospital products); and (iv) represent well-identified conditions (for medical hospital products).
To reflect the true value and opportunity costs of products and services, the prices used in comparative studies should be observed from transactions in competitive markets. Primarily this means that it should reflect the price that the seller and purchaser agree on for a given product or service. Such price observations are not always readily although many OECD countries have put systems in place to approximate the monetary value of services provided by hospitals.
The results of this project will provide policy makers with a greater understanding of the role of price and volumes when comparing hospital expenditures internationally. They will also provide researchers with a valuable new tool to compare health care prices across countries.
Mr. Luca Lorenzoni, Luca.Lorenzoni@oecd.org.