Addressing inequity in general practice income

There is large variation in GP practice income.  Concern has been expressed, both locally and nationally, that this might be contributing to inequitable care.  Dame Barbara Hakin, the Deputy Chief Executive, has announced the Department of Health’s (DoH) intention to move towards equitable contracts starting in April 2014 (ref 1).  She has also signalled the DoH intention to move towards a single weighted capitation contract.  MPIG (minimum practice income guarantee) protected GMS practices incomes which would have fallen under the 2004 contract, and will be phased out completely over 7 years from 2014(ref 2).

In Bristol, there are 48 PMS practices and 7 GMS practices (87% PMS compared with 40% nationally).  For both types of practice, income can be divided into core income, Quality and Outcomes Framework(QOF) payments and other payments (premises payments, IT, extended services,  training , seniority, drugs and dispensing).  In 2011-12 the breakdown of payments was

Income

55 Bristol practices in  2011-2

GMS or PMS core income

£31.7m (54%)

QOF

£8.1m

Other

£14.9m

Total

£58.4m

Practice financial data for this study was obtained from NHS Bristol using the Freedom of Information Act.

Carr-Hill formula

The needs based element of GMS reimbursement(the “global sum”) uses the Carr-Hill formula (ref 3) to weight practice populations adjusting list-size for

·            age and sex

·            Standardised Mortality Ratio and longstanding illnesses

·            rurality

·            market forces factor (to allow for variations in staff pay)

·            turnover

·            nursing and residential home residency

Perhaps significantly, the formula does not adjust for deprivation, although Carr-Hill would argue that the above factors cover workload variations caused by deprivation.    You can see how Carr-Hill adjustments affected list size in the chart below, figure 1.  The 2 practices farthest above the line are in the Bristol University area with a large proportion of students. As you can see, the largest practice had its list size adjusted downwards from 20373 to 15335 (which would have reduced its core income by 26% if it were GMS).  PMS practices do not receive a global sum to provide essential and additional services, but instead receive a baseline allocation, locally agreed.

Figure 1 : unadjusted list sizes compared with Carr-Hill adjusted list sizes, Bristol 2011-2


Within core income, there was a 2 fold variation in practice income per patient (figure 2), using the Carr-Hill formula to adjust for demand. There is a significant negative correlation (-0.3, p<0.05) between list size and income, with (a small cluster of) smaller practices tending to do better per capita. Even when we exclude the outliers, there are a significant number of midsize practices with quite different incomes (range £60-100 per patient). An increase £20 per patient would add £200,000 to a practice budget where the list size was 10,000, enough for 1 additional GP and 2 practice nurses, for example.

Figure 2: Variation in core practice income per Carr-Hill adjusted patient, by list size, Bristol 2011-12

The lowest earning practice received £59.71 per Carr-Hill patient per year, where the highest earning practice received £120.10 per Carr-Hill patient per annum. Of the seven GMS practices, 6 were in the bottom 7 earners, with one GMS practice with a large MPIG (confirmed by NHS Bristol) ranked 35th out of the 55 practices. The only PMS practice in the bottom 7 practices was ranked 53rd out of 55. So, low earning practices are either GMS who were low earning pre the 2004 contract and hence received low MPIG, or PMS with a “poorly negotiated” PMS contract. Within PMS practices there is still wide variation (£61.15 per patient to £120.10 per patient). Within GMS practices, 6 had incomes in a very narrow range between £59.71 to £62.22, with 1 practice achieving an income of £77.16 (thanks to a large MPIG).


Bristol PCT’s attempt to equalise incomes
In 2011-2 Bristol PCT made efforts to equalise incomes, by a collar and cuff method (the equitable pricing project) , reducing the high earning practices to an upper threshold and bringing up the lower earning practices to a minimum income per patient. Other PCTs have used other schemes to try to tackle GP income inequity. The Bristol scheme was implemented in September 2012 via a LES(locally enhanced service), so does not affect the 2011-12 data used here. 15 low earning practices were offered uplifts in funding from £3000-£73000 per practice per year. The PCT intention was to fund these uplifts from clawbacks from the high earning practices. As might have been expected, this move has not gone down well amongst the high earning practices, and has led to some staff redundancies, but it seems to have been accepted. It still leaves a fairly wide variation in pay rates, with clusters at the upper and lower boundaries.


Conclusion
This local analysis demonstrates clear large disparities in practice incomes. Concerns about inequity appear to be well founded. Local attempts to reduce inequity need to be monitored. National efforts to reduce inequity need to be supported as long as they do not destabilise primary care. Resources need to be shown to be following need and performance.

Evidence that income variations reflect need and performance
PMS contracts were intended to pay for extra local services and thus improved care. Could the large variations in income shown here mean that some practices are responding to local need and providing tangible improvements in health care?  In 2  reports to follow, I will be looking at whether there is any significant evidence of correlations between practice income and need, or between practice income and performance.

Conflict of interest

report written by recently retired GP partner from a low earning practice in Bristol

Click here for PDF of this document, with references