Cross sectional study of social variation in use of an out of hours patient transport service

BMJ 1999; 318 doi: https://doi.org/10.1136/bmj.318.7183.566 (Published 27 February 1999) Cite this as: BMJ 1999;318:566
  1. Catherine AO Donnell, lecturer (kate.odonnell{at}udcf.gla.ac.uk)a,
  2. Alex McConnachie, statistician,a,
  3. Katrina Moffat, clinical research fellow,a,
  4. Neil Drummond,, senior research fellow,b,
  5. Philip Wilson, senior clinical research fellowa,
  6. Sue Ross,, lecturera
  1. aDepartment of General Practice, University of Glasgow, Woodside Health Centre, Glasgow G20 7LR
  2. bPublic Health Research Unit, University of Glasgow, Glasgow G12 8RZ
  1. Correspondence to: Dr C O'Donnell, Department of General Practice, University of Glasgow, Glasgow G12 0RR
  • Accepted 23 October 1998

Out of hours primary care has undergone radical reorganisation in recent years, with increasing numbers of general practitioner cooperatives operating from primary care emergency centres.1 A major issue continues to be equity of access, particularly in areas of socioeconomic deprivation where demand is high but access to transport is poor. 2 3 In Glasgow, 52% of the population reside in areas of deprivation (Carstairs and Morris deprivation categories 6or 7).4

The Glasgow Emergency Medical Service was established in February 1996: it covers around 950000patients and 95% of the city's 219general practices, and operates from six centres across the city. The service offers free transport for patients between their homes and the centres.

Subjects, methods, and results

We collected data on all patient contacts with the emergency service over one week in October 1996(n=3193). The socioeconomic category of the patients was derived from their postcode sector of residence (depcats 1and 2,affluent; 3-5,intermediate; 6 and 7,deprived).4 Time of first contact with the service was categorised as evening until midnight, night, and weekend daytime. To standardise the distribution of contacts over time, we calculated rates of contact per million person hours then analysed these by Poisson regression. We analysed service response by logistic regression. Independent variables for both models were age, socioeconomic category, and time of first contact.

Sociodemographic data were available for 2882contacts (90.3%), giving a crude contact rate of 28.1per million person hours (equivalent to 157.8contacts per 1000patients per annum). We found an interaction between socioeconomic category and age group (P=0.002), with 60% higher contact rates for children and adults from deprived areas (aged <5: affluent, 79.4per million person hours; deprived, 130.9; schoolchildren: affluent, 19.4; deprived, 31.3; adults: affluent, 15.1; deprived, 24.1). In elderly people the contact rate was 38% higher for the affluent group (affluent, 51.7; deprived, 37.5).

Of the 3193contacts, 1713(53.7%) attended centres, 726(22.7%) received home visits, 449(14.1%) were given telephone advice, 63(2.0%) were sent an ambulance, and 144(4.5%) did not attend as arranged (unknown for 98(3.1%)). Socioeconomic category influenced the probability of receiving a home visit (P=0.037), with adults and elderly people in deprived areas more likely to receive one. Socioeconomic category did not affect the likelihood of receiving telephone advice (P=0.42) or attending a centre (P=0.29) (table).

Response for different age groups by patient socioeconomic category (full data available for 2641/3193 (82.7%) contacts). Values are percentages (numbers) of patients

View this table:

Full data were available for 1607(93.8%) patients attending centres, of whom 304(18.9%) used the patient transport service. Patients from deprived areas were four times more likely to use patient transport (affluent, 6.3%; deprived, 25.2%: P<0.0001): this trend was most apparent at night, when there was a sevenfold difference between affluent (6.3%) and deprived areas (44.6%: P<0.0001).


Socioeconomic category influenced the use of the Glasgow Emergency Medical Service, with increased contact rates by children and adults from deprived areas. For elderly people, the highest contact rate was among the most affluent. While the effect of affluence on raising expectations and service use has been reported, this association has not previously been linked to age.5

After contacting the emergency service, adults and elderly people from deprived areas were mostlikely to receive a home visit. The reasons for this are unclear and require further investigation. The lack of association between socioeconomic category and centre visit rates may be attributable to the provision of patient transport for anyone asked to attend a centre: those from deprived areas were far more likely to use the service. The cost of this service for 1998-9will be around £240000, yet may still be cheaper than taxi fares or home visits. While this requires economic evaluation, the evidence suggests that a free and accessible patient transport service may contribute to equity of access to out of hours primary care across the socioeconomic spectrum.


We thank Dr James O'Neil, John Easthope, and the staff of the Glasgow Emergency Medical Servicefor their cooperation; staff of the Robertson Centre for Biostatistics for data entry; Department of Health Information, Greater Glasgow Health Board for routine data; Ms Michere Beaumont for secretarial support; Professor Graham Watt for his discussions. Ethical approval was obtained from theGreater Glasgow Community and Primary Care Research Ethics Committee. The Public Health Research Unit is supported by the Chief Scientist Office of the Scottish Office Department of Health, but the views expressed in this paper are those of the authors alone.

Contributors: COD contributed to the conception and design of the study, collection and analyses of the data, and coordinated writing of the paper; she will act as guarantor for the paper. AMcC performed statistical analyses and contributed to writing the paper. KM, ND, and PW contributed to the conception and design of the study, collection of the data, and to writing the paper. SJR contributed to the conception and design of the study, collection and analyses of the data, and to writing the paper.

Funding: Glasgow Emergency Medical Service contributed towards data entry costs.

Conflict of interest: None.


View Abstract