Background
Promotion of self medication has been enthusiastically embraced within the UK in recent years, in a bid to facilitate improved technical and allocative efficiency in the provision of primary care services. Technical efficiency improvements would be realised by optimising lay and professional skill mix in responding to minor illness, by encouraging appropriate self medication through the purchase of pharmacy available medicines, supported by the community pharmacist's expertise and advice. This will secure saving for the NHS by simultaneously dampening demand for GP consultations and alleviating pressure on the NHS drugs budget. Allocative efficiency could be potentially enhanced by freeing up GP time for relatively 'needier' patients, facilitating redeployment of saved resources to improve existing and develop new primary care services. A key assumption underpinning this is the belief that self medication will obviate the need for GP consultations, facilitating a direct substitution between the GP and community pharmacist in the treatment of minor illness. However, this substitution hypothesis is unfounded in evidence.

Objectives

1. To assess the extent to which such self medication results in substitution of the pharmacist in place of the GP; and
2. To develop an explanatory model aiming to predict and specify the relationship between factors influencing users' choice of consultation career paths to access P medicines.

Methods
Design: Cross sectional descriptive study with prospective follow up of users. Setting: 15 community pharmacies within the Lothian Health area in Scotland. Subjects: 1185 users accessing P medicines either on prescription (Rx) or over-the-counter (OTC) in community pharmacies. Research Instruments: semi-structured pharmacy and telephone follow-up interviews. Analyses: Univariate (chi-square) and multivariate (logistical regression) statistical techniques to explore associations and construct an explanatory model including access and use of GP and community pharmacy services, types of medicines and socio-demographic indicators as explanatory variables.

Results
A 75% participation rate was achieved, with a 61% follow up. One in four (23%) users revisited either the doctor or the pharmacist to follow up on the illness episode for which they obtained the P medicine. One in ten users (11%) visited the GP for a follow up consultation after their initial pharmacy visit to buy a P medicine OTC. Univariate analyses confirmed the prior hypothesis that users of higher socio-economic status were more likely to adopt a pharmacy first consultation career path, accessing P medicines OTC. Multivariate analyses supported the univariate associations. Users who visited the pharmacy first, compared to the GP first, were significantly less likely to: report any long-standing illness; consult the GP as frequently; be exempt from the prescription charge; or to be put off by the cost of medicines. They were significantly more likely to: consult the pharmacist regularly; be using a medicine on the NHS blacklist; and be off higher socio-economic status. Contrary to expectations, usual waiting time to see the GP was not a significant predictor of users' consultation career paths.

Discussion
The GP/pharmacist substitution hypothesis inherent to recent policy imperatives encouraging increasing self medication of minor illness using pharmacy available P class medicines is affirmed. The resource implications associated with the changing distribution of costs and benefits accruing to key stakeholders are outlined and the results contextualised to recent policy debates.