116British Journal of General Practice, February 2000
Background. The primary care system in the United
Kingdom, involving the general practitioner (GP) as gate-
keeper to further services, has helped to keep health care
costs down. Despite this, unexplained variation in referral
rates and increasing health care costs have led to the
search for methods of improving efficiency. There is rela-
tively little recent descriptive data on the processes of care
at the primary–secondary care interface. The study reported
here provides information about this.
Aim. To analyse the patterns and process of care for the
referral of outpatients, together with the views of patients,
their GPs, and specialists.
Method. A questionnaire survey of outpatients, their hospital
specialists, and GPs in randomly sampled district health
authorities in the North Thames Region. The measures
included items and scales measuring satisfaction and
Results. Almost all of the outpatients thought that their con-
sultation with the specialist was ‘necessary’ and ‘worth-
while’. Most of the GPs felt that they could not have given
the study patients the care, treatment, and investigations
they received in hospital, and most of the sampled patients’
attendances were rated by the specialists as ‘appropriate’.
However, for just over one-fifth of new patients, the special-
ists reported that the GP could have done more tests and
examinations prior to referring the study patient. Large pro-
portions of GPs in this survey also reported having technical
equipment in their practices, as well as direct access to a
range of services and hospital-based facilities.
Conclusion. A large amount of work is carried out in gener-
al practice prior to the hospital referral of patients, and GPs
have direct access to some technologies and services that
can act to reduce the burden on hospitals. The discrepancy
between GPs’ and specialists’ perceptions about the poten-
tial for further investigative work prior to patient referral mer-
its further investigation.
Keywords: referral; patient satisfaction; general practition-
ers; hospital specialists.
tered with a National Health Service (NHS) general practi-
tioner (GP), and 90% per cent of health care contact takes place
in primary care. The well-developed primary care system,
INETY-EIGHT per cent of the British population are regis-
involving the general practitioner (GP) as gatekeeper to further
services, has helped to keep health care costs in the United
Kingdom down.1Despite this, unexplained variation in hospital
referral rates2-4 and increasing health care costs have led to a
search for methods of improving efficiency and effectiveness,
and of shifting the balance of care further to the primary care
sector.1,5-10Policy makers have encouraged developments in pri-
mary care with the aim of increasing efficiency and curbing the
increasing costs of health services, particularly in the secondary
sector. The most recent example is the replacement of GP fund-
holding schemes with new local commissioning schemes within
primary care groups.11
In England and Wales there are over 37 million recorded out-
patient attendances in NHS hospital clinics per year, at an annual
cost of around £2.5 billion. Therefore, even a small reduction in
outpatient attendances has the potential to make substantial sav-
ings for the NHS. Over 8.5 million (23%) of outpatient atten-
dances are new outpatient attendances and the remainder are fol-
low-up visits.12Cartwright and Windsor13reported, on the basis
of their national survey of outpatients in 1989, that 11% of all
adults will have attended a hospital outpatient department within
a three-month period. They concluded that there were too many
continuing attendances over long periods. Others have estimated
that around one-quarter of follow-up outpatient attendances are
unnecessary, and could perhaps be avoided or managed in gener-
al practice by different methods of organising care,6-8,14although
these schemes are largely unevaluated to date. Many studies have
sought and failed to identify the reasons for variations in GPs’
referral rates to outpatient departments. Faulkner et al14reported
that 38% of surgical outpatient attendances in the southwest of
England were still perceived by hospital doctors to have been
manageable by GPs, and 45% of these attendances were judged
to have been of marginal or little value. However, while it has
been argued that primary care-based health systems have lower
costs than others, for health care to be more ‘primary care-led’,
greater investment and transfer of resources from the secondary
to primary care sector is required, particularly in socially-
deprived areas.10,15-17Despite ongoing debate, there is little con-
sistent information on the extent to which organisational devel-
opments in primary health care have had an impact on the work-
load of GPs or hospital outpatient departments.1
Aims of the survey
This study aimed to describe the patterns and processes of care in
general practice prior to patients’ referral to hospital outpatient
clinics, together with the views of patients, their GPs and special-
ists. A separate paper focuses on the characteristics of fundhold-
ing practices and is published elsewhere.18
Eligible sample members were aged over 18 years and were cur-
rent outpatient attenders. Both first and follow-up attenders were
included in the study in order to be able to provide a profile on
all attenders. The study was based in five, randomly sampled dis-
trict health authorities in the North Thames Region, with stratifi-
The process of outpatient referral and care: the
experiences and views of patients, their general
practitioners, and specialists
A Bowling, MSc, PhD, Professor of Health Services Research; and J
Redfern, BSc, Research fellow, CHIME/Primary Care and Population
Studies, Royal Free and University College London Medical School,
Submitted: 8 May 1998; final acceptance: 16 June 1999.
© British Journal of General Practice, 2000, 50, 116-120.
British Journal of General Practice, February 2000 117
A Bowling and J RedfernOriginal papers
cation by area (the 15 district health authorities were divided into
Greater London/outside London, and then West to East). The
aims of the study were descriptive, and a minimum of 120
patients per specialty was aimed for in the sampling (720 in total
across all specialties). This number was sufficient to permit
analyses by specialty and was manageable within the study period.
Within sampled districts, hospitals that included all of the com-
mon study specialties on site were included in the study. The
study specialties (common specialty areas) were general medi-
cine, general surgery, dermatology, orthopaedics, rheumatology,
and respiratory medicine.
Two researchers attended each outpatient clinic and recruited
all patients as they attended consecutively on the study visit date.
The patients were informed about the study, given an explanat-
ory leaflet, and asked to sign the consent form to participate.
Those who consented were given a questionnaire to complete at
home after the clinic, and were asked to return it in a pre-paid
envelope. Questions for patients covered process, satisfaction,
and attitudes in relation to outpatient care and the interface with
primary care.9,13,19The satisfaction instrument (ratings on a five-
point scale from ‘excellent’ to ‘poor’) covered satisfaction with
waiting times for the appointment, convenience of the location,
waiting times at the clinic, time spent with the specialist, expla-
nations given, competence and personal manner of staff, and
other items relating to facilities and convenience.19Patients were
asked for their signed consent for us to contact their GPs and
specialists with a questionnaire about their referral and for their
GPs’ names and addresses. The specialists were asked to com-
plete a general attitude questionnaire about the organisation of
outpatients and the relationship between specialists and GPs, and
a questionnaire about the study patient (new attenders only). The
GPs were asked to complete a general attitude questionnaire
about the organisation of outpatients; the relationship between
GPs and specialists; practice characteristics, facilities, and equip-
ment; and a questionnaire about the study patient. The data were
analysed using SPSS for Windows; statistical tests included chi-
square tests and t-tests. Location of practice (Greater London or
outside London) was controlled for in all analyses presented
here. GPs’ patient list size and the casemix of outpatients (mea-
sured by patients’ ratings of effect of condition on their lives)
was controlled for where appropriate.
The response rates to the study were: patients, 74% (982); spe-
cialists to the general questionnaire, 100% (34), and to the indi-
vidual (‘new’ patients as defined by the hospital) patients’ ques-
tionnaire, 91% (184); GPs to the general questionnaire, 64%
(393), and to the individual (all) patients’ questionnaires, 64%
(552); and the outpatients’ managers, 61% (27). The GPs’
response rate was slightly higher for out of London than London
GPs at a ratio of 66%:61% (n = 194:195 respectively). The pro-
fessional and practice characteristics of the responding GPs were
compared against national data from the Royal College of
General Practitioners (RCGP)20-22(including practice size, mem-
bership of RCGP, percentage of fundholders, percentage of com-
puterised practices), and no differences were found. However,
our sample of GPs is not necessarily representative of all GPs
because they were included in the study on the basis of their
patients being included in our outpatient sample.
The patients and their views
Forty-one per cent (401) of the patients were male, and 59%
(581) were female. Thirty-one per cent (313) were aged under 45
years, 34% (333) were aged 45 to 65, and 35% (336) were aged
65 or over. Eighty-eight per cent (847) were white and 12%
(115) were in black and other ethnic minority groups. Table 1
shows the results of patients’ reports of the referral and atten-
dance. This shows that most patients reported that their GPs had
examined them thoroughly before their referral to an outpatient
department, and over one-quarter in each case reported having
X-rays or other prior investigations. Almost three-quarters of
patients had been referred to an outpatient department by their
GPs. Table 1 also shows that most patients reported being seen in
an outpatient department within three months of their referral,
although 6% waited six months or more. Most (60%) patients
were re-attenders at the time they were sampled for inclusion in
the study, although just 10% of all patients were long-term atten-
ders of five years or more.
The average number of attendances for all patients was 4.8
(including the sampled attendance). The average waiting time in
the clinic was 36.12 minutes: mean waiting times were longest in
London at 41.3 minutes (SD = 41.70) in comparison with 31.6
minutes (SD = 29.37) outside London (P<0.001).
Over two-thirds of patients were retained for outpatient fol-
low-up after the sampled consultation, and 36% of new attenders
and 52% of follow-up attenders reported that their condition had
‘improved’ or ‘cleared up altogether’ after the sampled consult-
ation (Table 1).
Patients’ and doctors’ views of the appropriateness of the
Almost all (95%; 800 out of 842 responders) patients thought
that their consultation with the specialist was ‘necessary’, and
89% (851 out of 857 responders) rated it as ‘worthwhile’.
Consistent with this, of the 540 patients for whom GPs comp-
leted the individual patient questionnaire item, 89% (481) were
rated by GPs as ‘GP not able to give the care, treatment and
investigations received in the hospital’, although 10% (56) felt
that they could have done; 1% (3) were uncertain.
General practitioners were asked to rate their level of agree-
ment with the hospital doctor’s decision to retain, discharge,
admit, or refer their patient. GPs reported that they ‘definitely
agreed’ with the decision for 77% (406) of the patients and
‘probably agreed’ for 14% (74) of the 525 patients they complet-
ed the question for; 6% (29) were rated as ‘uncertain’ and 3%
(14) were rated as ‘disagreed’. Most attendances were rated by
specialists as appropriate (90%; 103 out of 115 attendances
rated). However, for over one-fifth of the new attenders (22%; 39
out of 178 rated), the specialists reported that the GP could have
done more (tests, examinations) prior to referring the study
General practitioners’ views
Table 2 shows that a substantial minority of study patients was
reported by their GPs to have had tests and X-rays prior to their
referral. The main reasons for the referral were diagnosis and
treatment. A relatively large proportion of GPs, in response to a
general question, reported that they were given inadequate inform-
ation about patients from specialists. In addition, most GPs
reported, in response to general questions, unduly long delays
‘very’ or ‘fairly’ often before referred patients were seen in out-
patient departments, and sizeable proportions reported problems
with the grade of hospital doctor (too junior) seeing their patients
and long delays in communications from hospitals. There were
substantial numbers of practices with specialist outreach clinics,
shared hospital–GP patient care schemes (mainly for diabetes),
access to consultant-only clinics, locality outpatient clinics (e.g.
in community or cottage hospitals), and one-stop hospital clinics.
118 British Journal of General Practice, February 2000
A Bowling and J Redfern Original papers
Large proportions reported having specific equipment in the
practices, carrying out minor surgical procedures, and having
direct access to services.
Specialists rarely reported receiving inappropriate referrals from
GPs (6%, two out of 34, reported this ‘fairly’ or ‘very’ often),
although 38% (13 out of 34) reported that they had been provid-
ed with inadequate information by GPs ‘fairly’ or ‘very’ often,
and 42% (14 out of 33) reported inadequate investigations/treat-
ment were carried out by GPs prior to referral ‘fairly’ or ‘very’
often. Specialists were asked about whether there had been any
innovations in their departments that aimed to shift the balance
of care from secondary to primary care. They were given a
Table 1. Processes of outpatient care. Patients’ reports.
% (n) Patient reported items
Action taken by GPs before referral
Do/arrange other investigations
How the patient was referred to the outpatient
Referred by GP
Referred after being inpatient
Referred from other hospital clinic/A&E
Interval between referral and being seen in the
Less than one month
One month but less than three months
Three months but less than six months
Six months or more
New patient status
Follow-up patient status
Followed up for fewer than 12 months
Followed up fewer than five years
Followed up for five years or more
Process outcome of consultation
Retained for follow-up in outpatient department
Follow-up in GP’s surgery
Clinical outcome of consultation
Cleared up altogether
Total clinical outcome of consultation
Cleared up altogether
Average number of attendances
All patients (new and follow-up)
Average waiting time in clinic in minutes 36.12 (36.23)
Table 2. GPs’ reports of process of care and facilities.
Reported process in relation to study patients % (n)
Tests carried out
Reasons for referral
Advice on management
GP reported ‘very’/‘fairly often’
Inadequate information being given to GPs by
specialists about outpatients
Unduly long delays between referral and
Unduly long delays from hospitals in getting
information about patients
Patients been seen by insufficiently
GP has access to
Shared GP–hospital care (e.g. for diabetes,
One-stop hospital outpatient clinics
Specialist outreach clinics
Locality outpatient clinics
Consultant-only outpatient clinics
GP has the following equipment in practice
GP has direct access to
GP carries out minor surgery in the practice for
Cutting out cyst
Cuts that need stitching
British Journal of General Practice, February 2000119
A Bowling and J Redfern Original papers
checklist to tick, with provision for other responses. The most
commonly reported innovations are reported here (the item
response ranged between 31 and 33 doctors). Sixty-six per cent
(21 out of 32 responders to this item) reported having written
guidelines for GPs for complex cases, 64% (21 out of 33) report-
ed having consultant-only clinics, 47% (15 out of 32) had locali-
ty outpatient clinics, 42% (14 out of 33) had shared-care schemes
with GPs, 30% (10 out of 33) reported having one-stop clinics,
15% (five out of 33) had outreach clinics in GPs’ surgeries, and
10% (three out of 31) reported having community-based follow-
Comparisons between surveys
Table 3 makes cautious comparisons between some of the survey
data reported here and Cartwright and Windsor’s 1989 national
survey of outpatients.13Although the study reported here used
questions from this 1989 survey, comparisons can only be tenta-
tive because of differences in methodology and areas studied (the
latter was based on a national population screen to identify out-
patients). Bearing this caution in mind, Table 3 does suggest
large increases in the proportions of GPs who have equipment in
their surgeries, direct access to facilities, and an increase in the
investigations performed (e.g. X-rays) before the referral of
patients to outpatient clinics.
Before interpreting the results presented here, some caution
should be taken. This was a cross-sectional study and can there-
fore only demonstrate associations rather than impute causality.
This survey was also limited to sampling GPs via their referred
outpatients, suggesting potential for sampling bias, and was lim-
ited to one geographical region. However, there were no differ-
ences in the characteristics of GPs who responded and those who
did not, and in the characteristics of GPs nationally. This
enhances confidence in the findings reported here — although
the response rate was slightly higher for out-of-London GPs (by
5%), area was controlled for in all analyses.
The survey provided evidence that GPs are undertaking inves-
tigations for a sizeable proportion of their patients prior to refer-
ral, and most referred patients regarded the pre-referral examina-
tion by their GPs as thorough. Large proportions of GPs in this
survey also reported having technical equipment in their prac-
tices, performing minor surgery, and having direct access to ser-
vices and hospital facilities, including specialist outreach clinics.
These developments were all designed to improve patient access
and develop closer relationships between GPs and specialists, as
well as to discourage the number of reattendances by outpatients.
However, for over one-fifth of the new attenders the specialists
reported that the GP could have done more (tests, examinations)
prior to referring the study patient. In addition, a sizeable propor-
tion of specialists reported that GPs (in general) carried out inad-
equate investigations/treatment prior to referring patients ‘very’
or ‘fairly often’.
Almost all patients thought their outpatient consultation was
‘necessary’ and ‘worthwhile’, and most of their GPs felt that, in
retrospect, they could not have given the study patients the care,
treatment, and investigations that they received in the hospital. In
contrast to the literature cited earlier,6-8,14most outpatient atten-
dances were rated by specialists as appropriate, and less than
one-third of patients were discharged from outpatient clinics
after the sampled attendance. Most of the GPs agreed with the
decision of the hospital doctor to retain, discharge, admit, or re-
refer the study patient.
Although the indication is that the balance of care is shifting
from the secondary to the primary care sector, there is scope for
further development; e.g. possibly more investigations should be
carried out in general practice prior to a patient’s referral. Two-
thirds of specialists reported having developed guidelines for use
in general practice that could lead to shifts in the balance of care.
Evaluative research on the uptake and effectiveness of guidelines
in routine practice, and on other innovations such as one-stop
clinics, outreach clinics, telemedicine, and telephone consulta-
tions with specialists, is gradually being published.6-9There is
likely to be scope for the new primary care group commissioning
schemes11to influence criteria for outpatient referrals and effi-
Table 3. A comparison of North Thames Regional Health Authority (NTRHA) survey data (1995–1996) with Cartwright and Windsor’s13national
survey data (1989).
NTRHA data 1995–1996
National data 1989
% (n) GP reported
Availability of equipment and services in their surgeries
Nebuliser — yes
Computer — yes
Direct access to
Minor surgery performed in the practice
Carried out investigations of study patient prior to referral to hospital outpatient clinic
NB: Significance testing was not performed as the surveys are not strictly comparable; totals do not equal 100% as all items were multicoded.
120 British Journal of General Practice, February 2000
A Bowling and J Redfern Original papers
ciency targets for hospitals.
1. Coulter A. Evaluating general practice fundholding in the United
Kingdom. Eur J Pub Health 1995a; 5: 233-239.
2. Hippisley-Cox J, Hardy C, Pringle M, et al. The effect of deprivation
on variations in general practitioners’ referral rates: a cross-sectional
study of computerised data on new medical and surgical outpatient
referrals in Nottinghamshire. BMJ 1997; 314: 1458-1461.
3. Surrender R, Bradlow J, Coulter A, et al. Trends in referral patterns
in fundholding and non-fundholding practices in the Oxford Region,
1990-4. BMJ 1995; 311: 1205-1208.
4. Coulter A, Bradlow J. Effect of NHS reforms on general practition-
ers’ referral patterns. BMJ 1993; 306: 433-437.
5. Coulter A. Shifting the balance from secondary to primary care. BMJ
1995b; 311: 1447-1448.
6. Weingarten MA. Telephone consultations with patients: a brief study
and review of the literature. J R Coll Gen Pract 1982; 32: 766-770.
7. Harrison R, Clayton W, Wallace P. Can telemedicine be used to
improve communication between primary and secondary care? BMJ
1996; 313: 1377-1381.
8. Grimshaw JM, Russell IT. Effect of clinical guidelines on medical
practice: a systematic review of rigorous evaluations. Lancet 1993;
9. Bowling A, Stramer K, Dickinson E, et al. Evaluation of specialists’
outreach clinics in general practice in England: process and accept-
ability to patients, specialists, and general practitioners. J Epidemiol
Community Health 1997; 51: 52-61.
10. Hughes J, Gordon P. An optimal balance? Primary health care and
acute hospital services in London. [King’s Fund London Acute
Services Initiative. Working Paper No. 11.] London: King’s Fund,
11. Department of Health. The New NHS. [Cm. 3807.] London: HMSO,
12. Department of Health. Health and personal social services statistics
for England. London: HMSO, 1992.
13. Cartwright A, Windsor J. Outpatients and their doctors. A study of
patients, potential patients, general practitioners and hospital doc-
tors. London: HMSO, 1993.
14. Faulkner A, Saltrese-Taylor A, O’Brien J, et al. Outpatients revisit-
ed: subjective views and clinical decisions in the management of
general surgical outpatients in South West England. J Epidemiol
Community Health 1995; 49: 599-605.
15. Jarman B, Bosanquet N. Primary health care in London. Changes
since the Acheson Report. BMJ 1992; 305: 30-36.
16. Leese B, Bosanquet N. Family doctors and change in practice strate-
gy since 1986. BMJ 1995; 310: 705-708.
17. Starfield B. Is primary care essential? Lancet 1994; 344: 1129-1233.
18. Redfern J, Bowling A. Efficiency of care at the primary-secondary
interface: variations with fundholding status. Health and Place (in
19. Davies AR, Ware JE. GHAA’s consumer satisfaction survey and
manual. Washington: Washington Group Health Association of
20. Royal College of General Practitioners. General practice computeri-
sation. [RCGP Information Sheet No. 7.] London: RCGP, 1995.
21. Royal College of General Practitioners. Profile of UK Practices.
[RCGP Information Sheet No. 2.] London: RCGP, 1996.
22. Royal College of General Practitioners. Fundholding. [RCGP
Information Sheet No. 6.] London: RCGP, 1997.
We are grateful to Sally Anne Francis, Lesley Marriott, Orla Murphy, and
Maria Clement for their assistance with setting up this project; to Joy
Windsor and Ann Cartwright for permission to use questions from their
survey of outpatients and access to their original dataset, and for their
valuable advice; and to the patients, doctors, and managers who gave up
their time for this study. The study was funded by the Organisation and
Management Group, Research and Development, North Thames Regional
Health Authority (grant no. P/95/202), and the authors thank the members
of the committee. Crown Copyright reserved, 1995. The views reported do
not necessarily represent those of the funding body.
Address for correspondence
Professor A Bowling, CHIME, Royal Free and University College London
Medical School, 4th floor Archway Wing, Whittington Hospital, Highgate
Hill, London N19 5NF.
• Most patients felt their outpatient attendance was ‘neces-
sary’ and ‘worthwhile’; most of their GPs felt that, in ret-
rospect, they could not have given the study patients the care,
treatment, and investigations they received in the hospital; most
attendances were rated by specialists as appropriate.
• Just 29% of patients were discharged from outpatient departments
after the sampled attendance. However, most of the GPs agreed
with the decision of the hospital doctor to retain, discharge, admit,
or re-refer the study patient.
• Large proportions of GPs in this survey also reported having tech-
nical equipment and computers in their practices, as well as
direct access to services such as physiotherapy.
• Between around one-fifth and just over one-half of GPs
reported having access to specialist outreach clinics in general
practice, consultant-only clinics, locality out-patient clinics, and
one-stop hospital outpatient clinics.
• There was a difference of opinion between GPs and specialists
about whether much more preliminary investigation could be
carried out by GPs prior to referral of patients to hospi-