Annex 1. Calculation of GP revenue per country

SHORT DESCRIPTION OF THE STUDY OF DELNOIJ, 1975-1991 [1]

Delnoij published in 1994 as part of her thesis a chapter on GP income development from 1975-1991. The study included eight countries: Belgium, Denmark, Germany, Finland, France, The Netherlands, Sweden and the United Kingdom. Delnoij concentrated on two factors that can theoretically affect physician income: economic factors (supply, demand, competition), and political factors (the bargaining power of organized medicine). The relationships were expected to exist with regard to cross-national differences in the level of the GPs’ income as well as the development of their income. It was also expected that the development of GP income between 1975 and 1990 had been conditioned by the room the payment and the negotiation system left GPs as a group to increase their income through increasing the volume of care: if GPs had the room to increase their income through increasing volume, the development of their income between 1975 and 1990 did not depend on the development as regards the number of inhabitants per GP.

Delnoij formulated three hypotheses concerning the level of GP income and the development of their income:

1.  The relationship between the number of inhabitants per GP and GP income is positive, such that (a): GP income is higher in countries where the number of inhabitants per GP is higher, and (b) over time GP income increases more (or decreases less) in countries where the number of inhabitants per GP increases more (or decreases less).

2.  The relationship between GPs’ collective bargaining power and GP income is positive, such that: (a) GP income is higher in countries where GPs collectively hold a stronger position, and (b) over time GP income increases more (or decreases less) in countries where GPs collectively hold a stronger position.

3.  If GPs can increase their income by increasing the quantity of services provided, the development of their income depends less on the development as regards the number of inhabitants per GP (p. 108, [1])

Delnoij used the following data: GP income was defined as average gross revenue from general practice minus practice expenses, expressed in US$ppp. For GP power the position in the UK, Demark and The Netherlands were evaluated as strong, because in these countries, GPs (1) are independent, self-employed contractors with a national health service or health insurance organization, 92) serve as gatekeepers to specialists and hospital care, 93) have a monopoly on the provision of primary medical care, and (4) are represented by their own GP association in income negotiations. Possibilities to obtain higher income through increased volume were established to be moderate in Belgium and France and absent in the other countries.

The results do not provide evidence strong enough for these hypotheses to be rejected, though they do not strongly corroborate them either. GP income is higher where GP density is lower, except for countries with a strongly hospital-dominated health care like Sweden and Finland (until 1980). Gps incomes compared with per capita GDP have decreased since 1975. The decrease is, however, due to economic factors, namely the increase in GP supply. However, if GPs can increase their income by providing more services, which is the case in Belgium and France, they seem to be able to partly offset the negative consequences of increased physician supply. Evaluation of the development of GP income further showed that British GPs did remarkably well between 1975 and 1990. Dutch and German GPs, on the contrary, experienced a serious decrease in their relative incomes, that is, compared with per capita GDP, although they still rank number two and number one on the list of highest incomes in US$ppp.

The results lead to the rejection of hypothesis 1 for countries with a strongly hospital-dominated health care, such as Sweden and Finland (until 1980). In other countries, however, GP income is higher where GP density is lower. Hypothesis 2 must be rejected as far as the level of income is concerned. The GPs’ position is, however, a significant factor in explaining the development of income. In general GP income compared with per capita GDP has decreased since 1975, but this decrease has been smaller for GPs who collectively hold a strong political position. Apart from political factors, decrease of GP incomes can be explained by economic factors, namely the increase in GP supply. Finally, if GPs can increase their incomes by providing more services, which is the case in Belgium and France, they can partly offset the negative consequences of increased physician supply. The latter finding suggests that hypothesis 2 cannot be rejected. It should, however, be noted that the number of observations in this study is low. Income data could be analyzed for eight countries only.

TECHNICAL NOTES

The income per country will be provided in pppUS$. When selected sources resulted in different income estimates, we used the lowest estimate. In the overview below, the income in pppUS$ that is used in the paper is printed bold.

Conversion rates

The following conversion factors were used throughout the document:

Table 1. Power Purchasing Parities (ppp) and Euro conversion rates

Local currency to pppUS$ / Euro to local currency
Based on ‘old’ currency / Based on Euro
Country / 1995 / 2000 / 2005
Belgium / 35.6 / 35.6 / 0.9 / 40.3399
Denmark / 8.3 / 8.9 / 8.4 / 7.4538
Finland / 5.7 / 6.1 / 1 / 5.94573
France / 6.1 / 6.5 / 0.9 / 6.55957
Germany / 1.9 / 1.9 / 0.9 / 1.95583
Netherlands / 2.0 / 2.1 / 0.9 / 2.20371
Sweden / 8.8 / 9.7 / 9.6 / 8.4452
United Kingdom / 0.6 / 0.7 / 0.6 / 0.6095

( Table 1 continued)

Sources:

ppp 1995: / World development indicators 1997, The World Bank, Washington, DC, USA, Table 5.5 (for local currencies)
Germany: 1995 is estimated from the figures for 2000 and 1990, which are both 1.9
ppp 2000: / World development indicators 2002, The World Bank, Washington, DC, USA, Table 5.6 (for local currencies)
ppp 2005 / World development indicators 2006, The World Bank, Washington DC, USA, Table 4.14 (figures for 2004)
Euro conversion rate (euro to local currency) for euro-countries: / http://www.euro.ecb.int/en/section/conversion.html
Euro conversion rate: for other countries: / http://www.statistics.dnb.nl

Where data were provided in Euro’s in the original source in 1995 and 2000, the data were converted into local currency and then to ppp$US.

Differences with Delnoij:

Belgium:

Delnoij used an estimate for practice expenses that on average amounted 25-30% of the total income. We used a more detailed foundation of the practice expenses, based on a policy-paper of the Belgian Association of Physicians (ASGB: Algemeen Syndicaat van Geneeskundigen van België) concerning the income of Belgian GPs [2], information on the internet and email correspondence with ASGB.

Finland:

We used the income figure expressed in the European Observatory series [3]. Delnoij used national statistics, but they will probably be the same source as the data form the European Observatory series. For 2005 we used information provided by the country expert.

France:

In our study, the practice expenses for 1995, 2000 and 2005 were based on tax-figures. In Delnoij’s study, the estimate for practice expenses was 40% of the total income. In our study, this estimate is 46%.

The Netherlands:

Delnoij estimated the income of Dutch GPs based on the capitation fee for publicly insured patients. Revenue from privately insured patients was assumed to be equal to the full capitation fee paid for publicly insured patients. In our estimate, we used workload as the basis for calculation for privately insured patients. Workload was measured in the second Dutch National Study on General Practice [4,5]. The estimate for practice expenses used by Delnoij was based on the figure used in the tariff negotiations. We used an estimate based on real expenses, which were derived from tax-forms [6].

Sweden:

Delnoij used an estimate of the ‘average’ income based on calculations of the Federation of County Councils. We also used an estimate of the ‘average’ income, but the figure was derived from the questionnaire filled out by the Swedish Medical Association.

United Kingdom:

For 1995 and 2000, the same type of sources was used compared with Delnoij. Due to a change of payment system, for 2005 other sources were consulted

COUNTRY DESCRIPTIONS

Belgium:

Type of remuneration

The Belgium GP is remunerated on the basis of a fee-for-service system.

Establishment of tariffs or salary

The fees result from negotiations within the Committee of mutualities (these are sick funds) and physicians. The fees require endorsement of the Minister of Social Affairs. Besides this, a majority of physicians should agree with the fees. The fees are normally set for two years. Physicians are allowed to charge more than the negotiated fee. However, they risk suspension from practice when charging under the set level [7].

Sources and calculation of revenue and income

The income figures for Belgium were based on a policy-paper of the Belgian Association of Physicians (ASGB: Algemeen Syndicaat van Geneeskundigen van België) concerning the income of Belgian GPs [2] and email correspondence with ASGB.

In the ASGB paper, the calculation of the income is based on a working day of 8 hours, 5 days a week, 11 month per year. It is assumed that 1.5 hours per day is spent on non-patient bound activities, like schooling, meetings, and literature. This leaves 6.5 hours for patient contacts, from which 70% are assumed to be contacts at the GPs office and 30% home visits. The average duration of an office contact is 20 minutes, of a home visit 30 minutes. This results in 20 home visits and 68 office consultations in a week. On a monthly basis this sums up to 80 home visits and 272 office consultations. Yearly (based on 11 months) this is 880 home visits and 2992 consultations. Based on an average of 4.5 contacts per person per year, a full-time GP will serve a patient population of 860 persons. Extra income is generated by having patients of 50 years and older on the GP’s list (managing their medical record). The assumption is that 25% of the patients are older than 50 years and 60% of these patients will have appointed the GP as their medical record holder. This results in 126 extra remunerations (so called GMD-vergoedingen: General Medical Record allowances). These GMD-allowances were introduced in 1999 and are included in the figures for 2000 and 2005.

An important remark regarding the above workload calculations is the following: Having 860 persons as patient population is an improbable outcome of the calculation due to the fact that there are on average 479 inhabitants per GP available in Belgium (10,251,250 inhabitants and 21,415 GPs in 2000 (source: OECD Health data files 2005)).

Table 2. Estimation of income of full-time Belgian GP.

In Euro / 1995 / 2000 / 2005
Service / Number per year (2002)
(A) / Fee
(B) / Income
(A*B) / Fee
(C) / Income
(A*C) / Fee
(D) / Income
(A*D)
Home visits / 880 / 16.63 / € 14,634 / 18.74 / € 16,491 / 29.00 / € 25,520
Consultations / 2992 / 13.63 / € 40,781 / 15.29 / € 45,748 / 16.41 / € 49,099
GMD-vergoeding / 126 / € - / 12.51 / € 1,576 / 20.00 / € 2,520
Total Income + practice costs / € 55,415 / € 63,815 / € 77,139
Inflation correction / 92% / 98% / 111%
Source / OECD 2005 / OECD 2005 / Eurostat
Practice costs / € 38,459 / € 41,221 / € 46,747
Income excl. practice costs / € 16,956 / € 22,594 / € 30,392
In pppUS$
Income + practice costs / 62,794 / 72,312 / 85,709
Practice costs / 43,580 / 46,709 / 51,940
Income excl. practice costs / 19,214 / 25,602 / 33,768

The practice costs were also estimated in the ASGB discussion paper and include office costs (building, electricity, telephone, furnishing, information technology, insurances), personnel (practice assistant and cleaner), and the costs of a middle class car [2]. These data were based on the year 2001. For the years 1995 and 2005 the same data were used, corrected for inflation.

Denmark

Type of remuneration

In Denmark, GPs derive their income from a capitation fee, which makes up one third to half of their income, and from fees for services rendered (per consultation, examination, operation etc.).

Establishment of tariffs or salary

The fees are negotiated between the Organisation of General Practitioners (PLO) and the NHSS (National Health Security System) committee, which is run by the Association of County Councils. Costs for housing and staff are included in the fee structure.

Detailed description of (changes in) remuneration system

The Danish GPs serve as gatekeepers (although a very small proportion (1,7%) of the population fits into another scheme, where they are free to choose the physician of their choice. These people have to pay for all services except hospital treatment). There is no differentiation towards patient’s age in fees. Until 1995/96 children under the age of 16 were not individually registered with the GP but along with their parents (mother). If the child had a consultation, the service was registered under the parent’s civil registration number but marked with a “child mark”. After 1995 all children were also registered individually. This had no special impact on the earnings but made it possible to identify each registered person in a practice (e.g. consultations per registered patient). In the period from 1995 to 2005 there has been no change in the remuneration system but there have been small adjustments in the fees and the types of fees. However, the proportion of fee for services has increased due to higher activity in practice.