J. Bradlow et A. Coulter, EFFECT OF FUNDHOLDING AND INDICATIVE PRESCRIBING SCHEMES ON GENERAL-PRACTITIONERS PRESCRIBING COSTS, BMJ. British medical journal, 307(6913), 1993, pp. 1186-1189
Objective-To compare general practitioners' prescribing costs in fundh
olding and non-fundholding practices before and after implementation o
f the NHS reforms in April 1991. Design-Analysis of prescribing and co
st information (PACT data; levels 2 and 3) over two six month periods
in 1991 and 1992. Setting-Oxford region. Participants-Three dispensing
fundholding practices; five non-dispensing fundholding practices; and
seven non-dispensing, non-fundholding practices. Main outcome measure
s-Percentage change in net cost of ingredients, number of items prescr
ibed, average cost per item, and proportion of generic drugs prescribe
d after NHS reforms. Results-Prescribing costs increased in all practi
ces in the six months after the reforms. The net costs of ingredients
increased among dispensing fundholders by 10.2%, among non-dispensing
fundholders by 13.2%, and among non-fundholders by 18.7%. The number o
f items prescribed also increased in all three groups (by 5.2%, 7.5%,
and 6.1% respectively). The increase in average cost per item was 4.8%
for dispensing fundholders, 5.3% for non-dispensing fundholders, and
11.9% for non-fundholders. Dispensing fundholders increased the propor
tion of generic drugs prescribed from 26.9% to 34.5% and non-dispensin
g fundholders from 44.5% to 48.7%; non-fundholders showed no change (4
7%). Five of the eight fundholding practices made savings in their dru
gs budgets at the end of the first year of fundholding (range 2.9-10.7
%; the three other practices overspent by up to 3.6%). All non-fundhol
ding practices exceeded their indicative prescribing amounts (range 3.
2-20.0%). Conclusions-Fundholding has helped to curb increases in pres
cribing costs, even among dispensing general practitioners, for whom t
he incentives are different. Indicative prescribing amounts for non-fu
ndholding practices do not seem to have had the same effect.