Of the variables studied, the UPA(8) score was the strongest predictor of variations in referral rates. This association is most likely to be through a link with morbidity, although it could reflect differences in patients' perceptions, doctors' behaviour, or the use and provision of services.

Objectives: To investigate the relation between out of hours activity of general practice and accident and emergency services with deprivation and distance from accident and emergency department. Design: Six month longitudinal study. Setting: Six general practices and the sole accident and emergency department in Nottingham. Subjects: 4745 out of hours contacts generated by 45 182 patients from 23 electoral wards registered with six practices. Main outcome measures: Rates of out of hours contacts for general practice and accident and emergency services calculated by electoral ward; Jarman and Townsend deprivation scores and distance from accident and emergency department of electoral wards. Results: Distances of wards from accident and emergency department ranged from 0.8 to 9 km, and Jarman deprivation scores ranged from − 23.4 to 51.8. Out of hours contacts varied by ward from 110 to 350 events/1000 patients/year, and 58% of this variation was explained by the Jarman score. General practice and accident and emergency rates were positively correlated (Pearson coefficient 0.50, P = 0.015). Proximity to accident and emergency department was not significantly associated with increased activity when deprivation was included in regression analysis. One practice had substantially higher out of hours activity (B coefficient 124 (95% confidence interval 67 to 181)) even when deprivation was included in regression analysis. Conclusions: A disproportionate amount of out of hours workload fell on deprived inner city practices. High general practice and high accident and emergency activity occurred in the same areas rather than one service substituting for the other.

Background: Accurate spirometry is important in the management of COPD. The UK Quality and Outcomes Framework pay-for-performance scheme for general practitioners includes spirometry related indicators within its COPD domain. It is not known whether high achievement against QOF spirometry indicators is associated with spirometry to BTS standards.

Different types of primary care activity and contacts for different morbidities had different associations with deprivation. This makes it difficult to recommend a simple list size adjustment; however, increased activity in deprived wards needs to be recognized in resource allocation, service configuration and performance management in primary care.

Given the differences in presentations to both general practice and A & E services there may be limited scope for altering patients' consulting patterns without making significant changes to service provision. However, there may be scope for increasing the proportion of general practice contacts dealt with by telephone alone.

Increased private health care activity does not reduce the demand for NHS care: NHS and private referral rates were positively associated with each other after adjusting for age, deprivation and practice.

There is a widespread problem in recruiting general practitioners. Recruitment is hardest in areas with the greatest health needs.

Objective-To compare night visit rates in different electoral wards of one general practice with the Jarman and Townsend deprivation scores and unemployment rates.Design-Analysis ofcomputerised workload data. Setting-General practice in centre of Mansfield, Nottinghamshire.Outcome measure-Visits made in 588 nights to the 11998 patients on the practice list.Results-Night visit rates in 15 electoral wards varied from 19-6 to 55 3 visits per 1000 patients per year. The rates showed a significant association with the Townsend score (p=0.004) and the unemployment rate (p=0.03) but not with the Jarman score (p=03). The Townsend score explained 49% of the variability; unemployment explained 31% and the Jarman score explained 9%.Conclusions-Even in a general practice not eligible for deprivation payments there was a 2X8-fold variation in night visit rates between wards. In this practice the Townsend score was significantly better at predicting night visit rates than the Jarman score. This method of looking at internal variation in workloads in computerised practices could give more direct data on the relation between deprivation and general practice workload than has previously been available.