Estimating the 1997 to 2012 UK Health Accounts time series – methodology guidance

For the first time, the Healthcare expenditure, UK Health Accounts: 2018 bulletin contains a UK time series for healthcare expenditure running back to 1997, broken down by financing schemes and revenues of financing schemes, consistent with the internationally standardised definitions of the System of Health Accounts (SHA 2011). This replaces the previous series reporting the long-running trends in healthcare spending, ‘Expenditure on health care in the UK’, which is not comparable with data produced by other countries. The new measures will mean that from later this year, UK healthcare expenditure data on international datasets, such as OECD.Stat, will be consistent between 1997 and 2018, rather than just between 2013 and 2017, as data are presently. As shown in Figure 17, this will correct a current break in the time series displayed on international statistics databases, such as OECD.Stat.

It is important to note, however, that the full UK Health Accounts, including breakdowns of healthcare spending by function (type of health care) and provider, are still only available from 2013. The new data from 1997 to 2012 for healthcare financing are estimates based on the growth in healthcare spending using appropriate data sources, used in the existing time series beginning in 2013. Each financing scheme is based on a different methodology, which will be outlined in this annex, but healthcare financed by government, out-of-pocket payments and voluntary health insurance are estimated based on growth in long-running time series for equivalent services measured as components of healthcare expenditure in the UK Health Accounts from 2013 onwards.

The 1997 to 2012 time series for government-financed healthcare expenditure is produced by taking the existing 2013 UK Health Accounts spending figures as a starting point and then using estimates of annual growth back to 1997 to produce the historical time series.

The historical growth rates use national accounts data recording general government final consumption expenditure for health, as reported under the Classification of the Functions of Government (COFOG 7), as the base of the measure. This was also the basis for reporting public healthcare expenditure within the ‘Expenditure on health care in the UK’ series and the control total for government-financed healthcare expenditure in the UK Health Accounts.

However, the SHA 2011 definition of healthcare expenditure also includes a range of long-term care services that would be considered social protection (COFOG 10) under the COFOG classification system. Long-term care components are added to our growth index to take into account the different rate of growth in long-term care services. The long-term care services covered in our measure include:

• local authority-funded social care that is considered health-related, taken from NHS Digital data for England and StatsWales data for Wales

• expenditure on the Carer’s Allowance

Social care services that are considered health-related for Scotland and Northern Ireland are estimated based on the growth in services in England and Wales.

Government expenditure as reported in our UK Health Accounts also includes a range of preventive healthcare services provided by other government departments. However, these are not separately estimated in the historical growth rates, as the components are too small to have a meaningful impact on the growth in healthcare expenditure.

The time series for health care financed through individual out-of-pocket payments is also estimated based on growth in the components that make up healthcare expenditure in the UK Health Accounts from 2013 onwards. This is based on the same data sources used to produce estimates of healthcare expenditure between 2013 and 2018. In instances where the data sources do not extend back to 1997, growth is estimated based on a suitable proxy source, such as components of household final consumption expenditure (HHFCE) produced for the UK National Accounts.

Out-of-pocket spending can be separated into a number of distinct components, including:

• medical goods spending, derived from Office for National Statistics (ONS) HHFCE estimates
• medical services expenditure, largely derived from statistics produced by private healthcare organisations, such as LaingBuisson, and including individuals’ spending on acute care, dental services, primary care and mental health services
• spending on health-related long-term care services, derived from LaingBuisson estimates of the value of the self-pay market for residential and nursing care and home care services as well as government data on client contributions to social care services in England, Wales, Scotland and Northern Ireland

For voluntary health insurance schemes, the growth index used to estimate expenditure between 1997and 2012 is based on extending expenditure on the existing insurance products identified as relating to health care back to 1997. The products included in our estimates are:

• private medical insurance
• business self-insurance
• health cash plans
• dental insurance
• dental capitation plans
• an estimated health-related component of travel insurance

A share of investment income associated with health insurance and Insurance Premiums Tax (IPT) is added to net premiums for private medical insurance, health cash plans, dental insurance and travel insurance. Business self-insurance and dental capital plan spending are unadjusted.

For the remaining financing schemes, the existing methods are extended back to 1997. Healthcare expenditure from enterprise financing is based on ONS supply and use tables, which mean that our existing methodology can be extended back to 1997.

Our new method for estimating charity (non-profit institutions serving households, NPISH) expenditure uses national accounts data for the growth in NPISH final consumption expenditure associated with healthcare industries to estimate growth between 1997 and 2012. These growth rates are applied to our initial analysis into charity expenditure in 2012.