NB. If a
‘PPP value’ or ‘GDPD value’ is displayed in red type in the ‘Results’ panel,
this indicates that the value is based on IMF staff estimates. In this case, the
‘Final result’ will be based partly or wholly on value(s) based on IMF staff
estimates.
Important information for users
Full
details of the development, underlying methods and data, user interface and
applications of this web-based tool are described in a paper published in the
journal Evidence and Policy (Volume 6, Number 1, 2010, pp. 51-59). This
paper is available electronically via IngentaConnect via this link. Important information
for users is summarised below.
International
comparisons of the costs of health care, social and behavioural
interventions are increasingly important for a number of different applications
[1]. One specific application relates to systematic reviews of interventions
that include an aim to incorporate evidence on costs drawn from published or
unpublished studies. Since studies included in such reviews are often conducted
in different countries and/or at different times, estimates of costs are often
expressed in different currencies and/or price years.
For
example, ‘Study A’, conducted in the United States in 2000, reports an estimate
of the mean total direct costs of a drug rehabilitation programme per
participant that is expressed in 2000 US Dollars ($), whilst ‘Study B’,
conducted in the United Kingdom in 2005, reports an estimate of the same cost
measure for a comparable programme that is expressed in 2005 UK Pounds Sterling
(£). In these circumstances, in order for end users of reviews to make
meaningful comparisons between two (or more) estimates of the costs of a
programme, technology or intervention, it is preferable for these estimates to
be expressed using a ‘common metric’, which requires their adjustment to a
common currency and price year [2, 3].
Cost
adjustments are also sometimes applied in decision models for economic
evaluation. A decision model is an analytic tool used to support systematic
approaches to evaluating the impact of alternative interventions on costs and
other outcomes under conditions of uncertainty [4]. Decision models synthesise
data that may be collected from several different sources to inform specific
decisions. Ideally, these data should be assembled using a systematic but
necessarily iterative approach to searching, with the aim of identifying the
‘best available’ source(s) of data to estimate each model parameter. Unit costs
are one of the data components needed to estimate model parameters. Cost
calculations based on reliable databases or data sources conducted for the
specific study and in the same jurisdiction have been proposed as the preferred
source of unit costs data for use in decision models [5]. However, if the
preferred source is not available, it may be necessary to utilise unit costs
obtained from previously published sources. In these circumstances, if the
source unit costs are expressed in a different currency and/or price year from
those applicable to the decision model, some adjustment of these data is
required.
This
web-based tool adjusts estimates of costs for currency and/or price year using a
two-stage computation. Stage 1 adjusts the original estimate of cost from the
original price year to a target price year, using a
Gross Domestic Product deflator index (‘GDPD values’).
‘GDPD values’ are a measure of the change over
time in prices of all new, domestically produced, final goods within that
economy. This can be viewed as a measure of general inflation within an economy
over time, which takes account of inflation across a broad range of economic
sectors.
Stage 2
converts the price-year adjusted cost estimate from the original currency to a
target currency, using conversion rates based on
Purchasing Power Parities for GDP (‘PPP values’). ‘PPP values’ adjust
appropriately for differences in current price levels between countries, thus
allowing comparisons based on a common set of average international prices; this
is an advantage over pure exchange-rate conversion and GDP per capita
approaches. In other words, PPPs eliminate differences in price levels between
countries in the process of conversion, whereas pure exchange-rate conversion
and GDP per capita approaches do not. The price levels underpinning ‘PPP values’
are measured based on a general ‘basket’ of goods and services covering a broad
range of economic sectors.
‘GDPD values’ used in Stage 1 of the computation are obtained from the
International Monetary Fund (IMF) World Economic Outlook Database ‘GDP deflator
index’ dataset [6]. This dataset
contains ‘GDPD values’ for 184 countries
(currencies) from 1980 onwards. It is updated biennially in April and October
and each new release dataset is imported into the database underlying this
web-based tool.
You can select one of two alternative source datasets for ‘PPP values’ for use in Stage 2 of the computation. The first is the ‘Implied PPP conversion rate’ dataset, obtained from the IMF World Economic Outlook Database [6]. This dataset contains ‘PPP values’ for 191 countries (currencies) from 1980 onwards, updated
biennially in April and October. The second is the OECD ‘Purchasing Power
Parities for GDP’ dataset, published as part of the
OECD.Stat series [7]. This dataset contains ‘PPP values’ for all current OECD
countries (plus selected non-OECD member economies) from 1980 to the present. It is updated triennially in February, June
and December. As with ‘GDPD values’, each new release
of these two ‘PPP values’ datasets is
imported into the database underlying this web-based tool. Note that the ‘PPP
values’ contained in the IMF and OECD datasets differ slightly with respect to
the OECD countries in all price years, due to variations in the detailed
methodologies used to generate these data. Therefore, different results are
produced by this tool depending on the choice of dataset. There is no normative
basis for the choice between these two datasets.
This
web-based tool is a generic tool intended to be applicable across a large number
of different countries and all sectors, including (but not limited to) health
care, social welfare, education and criminal justice. It is important to be aware that the tool utilises one
of several methods currently available to convert costs to a target (common)
currency (that is, use of Purchasing Power Parities for GDP). As well as the
alternative methods mentioned above (that is, pure exchange-rate conversion and
GDP per capita approaches), health economists have developed and applied health
care-specific PPPs [8], technology-specific PPPs [9] and episode-specific PPPs
[1] for this purpose. An advantage of using these ‘context-specific’ methods in
health care applications is that they derive conversion rates based on
cross-country comparisons of the relative prices of, respectively, baskets of
health care services, specific health care technologies and specific health care
episodes, whereas Purchasing Power Parities for GDP derive conversion rates
based on comparisons of the prices of a larger and more diverse basket of goods
and services covering a broad range of economic sectors. Authors of systematic
reviews of health care interventions aiming to summarise evidence on costs, and
researchers developing decision models for health economic evaluation, should
therefore consider whether it is feasible to apply these more sophisticated
‘context-specific methods’ in preference to those underpinning this web-based
tool. The availability of detailed descriptive data on health care interventions
being examined (which are needed to inform calculation of technology and
episode-specific PPPs) is likely to be a key factor to be taken into account in
making this assessment.
No equivalent alternative ‘context-specific’ methods have yet been
developed based on cross-country comparisons of the relative prices of
non-health technologies, programmes, policies or services, such as those
implemented within and across the education, crime and justice or social welfare
sectors. Therefore, methods based on Purchasing Power Parities for GDP,
including those underlying this web-based tool, may currently be regarded as the
first-line approach for use in applications outside health care. Such methods
may also be viewed as a more methodologically straightforward alternative for
use in applications within the health care sector, for use in circumstances in
which context-specific methods are not judged feasible.
This web-based tool is developed as a joint initiative between The Campbell
and Cochrane Economics Methods Group (CCEMG) and the
Evidence for Policy and Practice Information and Coordinating Centre
(EPPI-Centre).
Please direct any questions or comments to Ian Shemilt, CCEMG Co-convenor (e-mail:
Ian.shemilt@medschl.cam.ac.uk
).
References
[1] Schreyogg J, Tiemann O, Stargard T, Busse R. Cross-country comparisons
of costs: the use of episode-specific transitive purchasing power parities with
standardised cost categories. Health Economics 2008: 17(S): S95-S103.
[2] Shemilt I, Mugford M, Byford S, Drummond M, Eisenstein E, Knapp M,
Mallender J, McDaid D, Vale L, Walker D (2008a). Chapter 15: Incorporating
economics evidence, in JPT. Higgins and S. Green (eds) Cochrane Handbook for
Systematic Reviews of Interventions.Chichester : John Wiley & Sons. Available from:
http://www.cochrane-handbook.org
[3] Shemilt I, Mugford M, Byford S, Drummond M, Eisenstein E, Knapp M,
Mallender J, Marsh K, McDaid D, Vale L, Walker D (2008b). Campbell Collaboration
Methods Policy Brief: Economics Methods (updated
April 2008). Oslo : The Campbell Collaboration.
[4]
Briggs, A., Claxton, K. and Sculpher, M (2006)
Decision modelling for health economic evaluation,
Oxford: Oxford University Press.
[5] Cooper
NJ, Sutton AJ, Ades AE, Paisley S, Jones DR. Use of evidence in economic
decision models: practical and methodological issues. Health Economics 2007; 16:
1277–86.
[6] International Monetary Fund. World Economic Outlook Database (October 2018). Available from: https://www.imf.org/external/pubs/ft/weo/2019/01/weodata/index.aspx (Downloaded 23 April 2019).
[7] Organisation for Economic Co-operation and Development. Purchasing Power Parities for GDP dataset (April 2019). Available from: https://stats.oecd.org/Index.aspx?datasetcode=SNA_TABLE4# (Downloaded 23 April 2019).
[8] Busse R, Schreyogg J, Smith PC. Variability in
healthcare treatment costs amongst nine EU countries - results from the
HealthBASKET project. Health Economics 2008; 17: S1-S8.
[9] Wordsworth S, Ludbrook A. Comparing costing results in across country
economic evaluations: the use of technology specific purchasing power parities.
Health Economics/i> 2005; 14: 93-99.