Volume measurement for non-market health and care services in the Norwegian national accounts

The total number of beds in institutions is currently used as a volume indicator for the institution services. One of the problems with this methodology is that the volume indicator does not distinguish between private non-market and public institutions. Another weakness with the methodology is that there is no split between different institution types such as nursing homes and old people’s homes. It is reasonable to assume that a bed in a nursing home is more expensive than a bed in an old people’s home. 

The project showed that the volume estimates can be improved by separating the private non-market institutions and the public institutions. Using the number of bed-days instead of number of beds will also improve the estimates. The estimates can be improved further by carrying out the calculation at the most detailed level (by splitting nursing homes and old people's homes, and by type of service, e.g. long-term stay, short-term stay, rehabilitation). The detailed breakdown of the output is a way to indirectly take quality changes into account. 

For the home-based care services, the total number of recipients is currently used as a volume indicator. This methodology does not take into account that there are large variations in the recipients’ need for assistance. There is also variation in what kind of service the user receives. 

The project showed that the volume estimate can be improved by using the number of hours assigned to each user. The growth rate increases significantly when using the number of hours. 

A breakdown by type of service could further improve the estimates. However, Statistics Norway concluded from the data that distinguishing between professional nurses and lower skilled workers is currently not bringing additional quality to the calculations. Furthermore, a split between non-market private providers and public providers is currently not possible with the data sets available. Further investigations are considered necessary to achieve a split between local government and NPISH data in the estimates. 

The change in methodology will be implemented in the annual national accounts of Norway for 2016. The time series from 2009-2015 will be revised in the in the next benchmark revision (in 2019).