Deprivation and volunteering by general practices: cross sectional analysis of a national primary care system

Since the inception of the NHS, primary medical services in the United Kingdom have been mainly delivered by a large number of general practices operating as independent businesses. As well as providing general medical services, practices can participate in several voluntary activities associated with quality care and service development.

Methods and results

Our study population comprised 5.35 million people served by 1050 general practices and is a complete national sample of patients and practices.

We ranked general practice populations using a modified version of the Scottish Indices of Deprivation 2003,¹ including currently available data for education, income, and employment, but excluding data for access and health. We used practice mean values to divide the population into 10 groups of equal size, from tenth 1 (least deprived) to tenth 10 (most deprived). We analysed the deprivation related distribution of population health indicators, practice characteristics, and participation in voluntary development schemes, using data for 2001-2 (table).

Table 1.

Characteristics of populations, medical manpower and general practices in tenths of the general population stratified by socioeconomic deprivation

	Demography					Population characteristics				Practice characteristics			Practice activities
Deprivation status	Registered patients	Practices	WTE principals	Non-principals	Total GPs	Composite deprivation	SMR<75	SIR	SHR	Single handed	Young GPs	Large list	PA	SPICE	Training	PMS
Tenth:
1 (least deprived)	535 290	94	344.2	87	431.2	−1.18	62.1	65.8	54.7	0.04	0.06	0.15	0.33	0.23	0.51	0.12
2	537 391	112	378.3	88	466.3	−0.76	76.4	77.2	67.6	0.05	0.07	0.15	0.30	0.22	0.43	0.16
3	534 250	114	362.7	73	435.7	−0.52	85.7	83.9	77.1	0.04	0.04	0.15	0.24	0.13	0.40	0.05
4	539 389	95	363.9	73	436.9	−0.35	91.9	88.5	83.5	0.02	0.07	0.09	0.28	0.22	0.45	0.07
5	529 883	97	354.2	61	415.2	−0.17	96.5	95.9	94.1	0.04	0.06	0.16	0.26	0.21	0.29	0.12
6	535 521	108	354.9	58	412.9	0.07	99.00	101.2	101.6	0.05	0.04	0.09	0.30	0.17	0.34	0.08
7	536 082	93	352.0	46	398.0	0.28	105.5	106.5	108.9	0.03	0.10	0.09	0.30	0.13	0.28	0.09
8	535 599	105	338.7	34	372.7	0.51	115.2	114.4	120.4	0.06	0.06	0.20	0.21	0.03	0.21	0.06
9	538 462	105	332.0	39	371.0	0.78	119.9	121.7	132.3	0.07	0.11	0.22	0.15	0.15	0.20	0.05
10 (most deprived)	528 283	124	352.2	53	405.2	1.47	150.1	143.2	170.0	0.07	0.15	0.11	0.12	0.13	0.22	0.04
Sum	5 350 150	1047	3533.10	612	4145.1	—	—	—	—	—	—	—	—	—	—	—
Mean	535 015	—	353.31	—	—	—	100.2	99.8	100.9	0.05	0.08	0.14	0.25	0.16	0.33	0.08
SII§	—	—	—	—	—	—	78.0	73.9	108.6	0.04	0.08	0.02	−0.18	−0.13	−0.34	−0.09
RII¶	—	—	—	—	—	—	0.78	0.74	1.08	0.84	0.99	0.12	−0.74	−0.81	−1.01	−1.05
P value**	—	—	—	—	—	—	<0.001	<0.001	<0.001	0.21	0.03	0.51	<0.001	<0.001	<0.001	0.01

WTE=Whole time equivalent; SMR<75=All cause deaths under 75; SIR=Standardised limiting long term illness; SHR=Standardised “not good” self assessed health; PA=Practice accreditation; SPICE=Scottish Programme for Improving Clinical Effectiveness participant practice; PMS=Personal medical services practice.

^*Mean age <38 years.

^†More than 1900 patients per WTE.

^‡Royal College of General Practitioners accredited training practice.

^§SII is the slope measure of inequality and is obtained by regressing the values of the variable of interest for each tenth on the cumulative proportions of the population. The SII therefore represents the expected difference between the top and bottom tenth as measured by the average slope of the variable across all of the tenths. It is positive (negative) when the variable takes higher (lower) values in more deprived (affluent) tenths.

^¶RII is the SII divided by the mean and can therefore be compared across variables.

^**The significance of the trend is measured using the t ratio for the slope coefficient.

By design, the composite deprivation index increases across tenths with the largest increase between tenths 9 and 10. All three measures of ill health show a significant positive trend and greater than 2.5-fold variation across tenths.

On average, populations of 530 000 people were served by 353 whole time equivalent general practitioner principals, with little variation between tenths. The total whole time equivalent of general practitioners, however, including non-principals and doctors in training, was 11% higher (437.1 v 392.0, P < 0.001) in tenths 1-5 (least deprived) compared with tenths 6-10 (most deprived).

On average, each tenth was served by 105 general practices, with larger numbers of practices in the most rural (tenth 2) and deprived (tenth 10) areas. This reflects the higher proportion of single handed and small practices in such areas. General practitioners' partnerships in deprived areas also have lower average ages and a higher proportion without a female general practitioner.

Variation between affluent and deprived areas is more than twofold (tenths 1-3 v 7-10, P < 0.001) in the proportion of practices involved in training general practitioners. Although younger general practitioners are more likely to work in deprived areas, it is less likely that they could have been trained there.

Potential markers of quality general practice, such as practice accreditation (see www.rcgp-scotland.org.uk/products/practice.asp), and enhanced data collection schemes, such as the Scottish Programme for Improving Clinical Effectiveness (SPICE, see www.ceppc.org/spice/index.shtml), were 80% (P < 0.001) and 90% (P < 0.001) more common, respectively, in the more affluent practices (tenths 1-3) than in the more deprived practices (tenths 7-10).

What is already known on this topic

The aim of the NHS is to provide comprehensive health care according to need

What this study adds

The provision and development of primary medical services varies inversely with need, particularly if promoted on a voluntary basis; the NHS needs to do more in promoting primary care for the population as a whole

Although the Personal Medical Services initiative was launched with the intention of improving clinical care in rural and deprived areas,² participation was 2.2 times more common in affluent areas (P < 0.001).

Comment

Insofar as the leading edge of general practice may be characterised by participation in quality schemes, health service initiatives, and postgraduate training, these activities feature about twice as often in practices serving more affluent areas. Although there are training and pioneering practices in poor areas, they are exceptions to the general rule.

These activities are not centrally distributed but are taken up by practices that volunteer. Practices serving the most deprived areas are less likely to volunteer, possibly because they are so consumed by dealing with increased levels of morbidity, without increased levels of medical manpower, that they are unable or unwilling to take on additional activities.³

The patterns described in this paper are generally hidden from public view, as a result of the convention of reviewing healthcare services at the level of large administrative areas, with substantial social heterogeneity. A different focus is required to monitor the leading edges of primary care development.

Arguably, the greatest challenge facing any national health service aiming for the equitable delivery of high quality care is to develop the best examples of care and the most attractive professional career opportunities in populations where need is greatest.⁴ The NHS has much more to do in tackling this challenge.