Reprinted from the BRITISH MEDICAL JOURNAL,

15 October 1983, 287, 1115 -1118

Hospital Topics

General surgical workload inEngland and Wales

TIMOTHY G ALLEN-MERSH, RICHARD J EARLAM

Abstract

An attempt was made to measure the workload of atypical general surgical firm (two part time consultantsand their junior staff) serving a population of 100 000 inEngland and Wales. This provides a background againstwhich to plan curricula for undergraduate and post-graduate teaching, as well as being a guide to the experi-ence that a surgical trainee should get in a suitabletraining post. The effect of changes in surgical staffingon the number of operations done by a surgeon may alsobe estimated.

Introduction

The workload of a hospital doctor is important in deciding onthe required number of hospital consultants and their juniors,training programmes for specialisation, and the teaching ofundergraduates. Although there has been some debate aboutthe number of consultant general surgeons required for Englandand Wales,1 this has not, to our knowledge, been based on aquantitative assessment of the national workload in generalsurgery.

The London Hospital, London El IBBTIMOTHY G ALLEN-MERSH, MD, FRCS, senior surgical registrarRICHARD J EARLAM, MCHIR, FRCS, consultant surgeon

Correspondence to: Mr Timothy G Allen-Mersh.

COPYRIGHT <"<? I1*' ALL RIGHTS OF REPRODUCTION OF THIS R E P R I N T ARE RESERVE! '

BMJ/455/83 IN A U ~ C O U N I K l t b oh r l < E WORLD

Measurement of a doctor's workload is difficult and con-troversial, but in surgical practice components of the workloadthat can be measured are: the number and type of operationsperformed, the number of inpatients, and the number of out-patient and domiciliary consultations. The purpose of thisstudy was to provide a quantitative description of the workdone by the consultant general surgeons in England andWales (almost 1000 serving about 50 million inhabitants). Theresults were related to a population of 100 000 so that individualsurgeons, serving different sized communities, may relate theirown experience to the figures given.

Methods

The most recent year for which complete statistics areavailable is 1978, so this year was chosen for the study. Out-patient attendance and bed occupancy figures in generalsurgery were obtained from SH3 national summaries collectedby the Department of Health and Social Security.8

Outpatient numbers were divided by 52 to convert them fromannual to weekly attendances. Estimates of the number andtype of surgical operations were obtained from the hospitalinpatient inquiry 10% sample of discharges, derived principallyfrom hospital activity analysis and processed by the Office ofPopulation, Censuses and Surveys on behalf of itself and theDHSS.3 The proportion of urological procedures performed bygeneral surgeons are contained in Ashley and Collingwood'sreport on urological requirements in Great Britain.4 Theincidence of different cancers was taken from the cancerregistry of the Office of Population, Censuses, and Surveys.'

The number of consultant surgeons in each region was foundin tables issued by the DHSS." Most consultant surgeons arenot contracted to work for the National Health Service for thefull working week (11 notional half days). The total number ofconsultants may be misleading, and a more appropriate measureof consultant manpower is the whole time equivalent. This iscalculated by adding up the number of contracted notionalhalf days a week and dividing by 11.

Results

All figures are given as the means for England and Walesper 100 000 population unless otherwise stated. The populationof England and Wales in 1978 was 49-1 million.

OUTPATIENTS

One hundred and thirty three surgical outpatients per 100 000population were seen each week. Thirty eight (29%) of thesepatients were new referrals, while 95 (71%) were "old," beingseen for further investigations, results of tests, or diagnosis oras a routine postoperative visit. In addition, two patients per100 000 population were seen each week in domiciliary con-sultations. 7

INPATIENTS

Thirty seven patients per 100 000 population were admittedeach week under the care of a general surgeon into beds on a

TABLE I—Top 20 general surgical operations

Operation

AppendicectomyInguinal hernia repairBenign breast disease (excision biopsy)CholecystectomyAll anal operations (including fissure, fistula,

and haemorrhoids)Cystosocopy with or without bladder

diathermy*Varicose veinsMalignant skin lesion (excluding melanoma)CircumcisionProstatectomy*MastectomyOrchidopexyColectomy, total or partialRectal carcinoma, excision or diathermyThyroidectomyVagotomyHydrocele (aspiration or excision)Femoral hernia repairAmputation of leg for vascular diseaseDefunctioning colostomy

No per Total No in100 000 England and Wales

143 5129-675 5739

71 6

62-354'751 644-635529923 621-518-817316911-711-68-78-0

704806365037 10036 31(1

35 160

30620268802533021 920174201467011 5801057092408500828057305 72042503940

*Excluding operations performed by whole time urologists.

surgical ward. Seventeen (46%) of these patients were emer-gency admissions, and the remaining 20 (54%) were admittedeither from the waiting list (13; 36%) or by diary bookedadmission and inpatient transfer from another specialty (7;18%). The mean number of patients occupying general surgicalbeds at any time was 47 per 100 000 population. Only 43 (91%)of these patients are recorded as general surgical admissions;one reason for this discrepancy may be that the remainingfour patients occupying general surgical beds were not admitted

IIg&i

HE

toUJo\cb

ato•ft00

to<Tl

-J

K)CO

§

E»Q 05 O «ar HH »ZSin in m it utl-l'os, E-.S s,a|s 1? iajiill II fir stiff*t • § isi r sllss 1 3 gifs " lii.g "3aS B

•0 j'l • •

1

| ' ' .

S"^ '' ' ' '2.o.

O OO OO ** ****t** »* N» MUM^ wOi 3-i 6 OK>» 4^ 6ww

t

to W Ol 4xu»^ 00 -J rfklJiUito *^JQv 00 to O CTi CJ* ^ *? Ui* lUI

\^^ v_^ , k ,frt^W to ffl Jjj OH ^o

C-o&n o o >n n ^ — Sf«3E: s: s: s ^3 3 i-§.§§.§ § § s S-E ge.P-o^ao, 2, S ° oS i 3

||l| || ^ 3| -j

BB» 1 I " ; " " |

ft - '. ' " . . . ' fgj- |"-1 n£.•3

51

c

1*<

i ( 1 ^ 4i > i

t*t to 00 »— UJ

s si § § o §§ i§

If I ill 1 1 IP If|l a i|| s S s | |8"« 1 a s « ||«

"* HP*

6f' . • - , . . - . 3 • - '

S., a., g ,;•

g

' ' 1 '

- r

2J S 2SS 2 5 1 s| g|* •" . * i •* / j , i

- * -

»** K) W

I— <O (— OV !*•• IO KJ -JH- H- ^JW iiift >— w 4k wife- (O O

^^ -^ ^__v_' A v - L J

| gf §e z2§l?$ * 1

iftliSiillHSri|58ii!j! -f • lilli- ' I ;-r n •-*.

1 •' 8 &e g 2.3 1 s3 °o3•<

\ * i, '

'

IO t— to "-M •- •- -J » ^ » ^

S -Jo ww 9 4ka<Q -j \p a-i-nS woo in ^5<£ « o -Jo O o o o o o o o o o o

III

c cancercancert)ow

el cant

»

ssl-

,*

sii

Cole

ctom

yE

xcision

01S

mall-b

ow

Z'3Ov;2,3oq

*

,

-5£uioo tn

*o

ooo

I

under the care of a general surgeon. The mean hospital stayfor patients occupying general surgical beds was 8-2 days.

Twenty nine (62%) of the 47 patients occupying generalsurgical beds underwent an operation while in hospital. Thecommonest reason for a patient occupying a general surgicalbed without undergoing an operation was observation after ahead injury (six patients a week per 100 000 population), whichaccounted for 39% of the non-operative cases in a generalsurgical ward.

SURGICAL OPERATIONS

Roughly 645 000 general surgical operations were performedin England and Wales during 1978. The 20 most commonlyperformed operative procedures accounted for 68% of allgeneral surgical operations (table I). An allowance has beenmade, in accordance with Ashley and Collingwood's report,1

for the urological procedures performed by urologists.Details of peripheral vascular procedures are not included in

the hospital inpatient inquiry tables. The incidence of legamputation for vascular disease was 8-7 per 100 000; many ofthese patients would have undergone one or more revascularisa-tion procedures. The incidence of abdominal aortic aneurysmgrafting was 2-2 per 100 000 population.

Half of the procedures in table I have been classified asmajor surgical operations.8 To provide a better perspective ofthe serious diseases managed by general surgeons, the incidenceof cancers dealt with by general surgeons is summarised intable II. Eight patients per 100000 population underwentoutpatient day case surgery each week.2

GENERAL SURGICAL CONSULTANTS

Of 945 consultant general surgeons in England and Wales in1978 (849 whole time equivalent), 124 (13%) held whole timecontracts with the NHS, the remainder being employed onpart time contracts. The variation between regions in the sizeof population served by a whole time equivalent general surgicalconsultant for the period between 1976 and 1980 is shown inthe figure. During this period there was an increase in thenumber of surgeons employed in the less well staffed regions,which resulted in a small reduction in the median of the ratioof population to surgeon for all regions from one whole timeequivalent consultant surgeon per 59 000 population to one per55 000 population.

The variation in staffing between regions shown in the figurerefers to the average or mean staffing level in each region.

100-

80-

Population(Thousands)

40-

20.

[

.

! I 1 1 1:

:

1979 1980

The variation in the size of the population (thousands), servedby one whole time equivalent general surgeon, between regionsin England and Wales from 1976 to 1980 is presented in a "boxand whisker" format. The "box" represents a quartile above andbelow the median (horizontal line), and encloses half the variationin regional mean staffing levels. The "whiskers" at each endrepresent the upper and lower extremes of regional mean staffinglevels.

This hides an inequality in staff distribution between districtswithin a region. In the North East Thames Regional HealthAuthority, for example, there was a fivefold difference in staffinglevels between districts, varying roughly from one whole timeequivalent consultant surgeon per 19 000 population in onedistrict to one per 100000 in another.9 Thus there is a muchgreater disparity in starring levels between districts within aregion than is the case between regions. This is not apparentfrom the figure, where only the mean for each region is shown.

Discussion

An attempt has been made to provide a quantitative estimateof the workload in general surgery in England and Wales basedon the figures that are available. There are some importantreservations.

(1) Figures recording the number of operations are collectedby clerical staff from details contained in the hospital case notes,in particular from discharge summaries. Owing to a combinationof factors, such as inaccurate discharge summaries, complicatedillnesses or operations, multiple operations during one ad-mission, anomalies between regions in the recording of day casesurgical admissions, and inexperienced clerical staff, this may

underestimate the work done,10 perhaps by as much as 20%.u

The number of operations done is certainly not overestimated.(2) The figures obtained refer to 1978, and the management

of some conditions has changed since then. For example, thenumber of hospital admissions for duodenal ulcer has fallenwith the introduction of H2 histamine blockers,12 and operationsfor inguinal hernia and varicose veins may have become morecommon." In addition, the figures may have altered becauseof changes in the incidence of disease since 1978.

(3) The operative workload does not necessarily correlatewith the amount of work in the postoperative period. Forexample, anterior resection of the rectum would be followedby more complications than would the four inguinal herniarepairs that could be done in the same operating time. Thenumber of leg amputations for peripheral vascular disease is apoor guide to the workload in vascular surgery since many ofthese patients undergo one or more reconstructive proceduresbefore amputation, and in many cases amputation is avoided.

(4) Few general surgeons will have a practice identical withthat described because the interdistrict variation in size ofpopulation served by a general surgeon may distort the natureof the workload. For example, if appreciable numbers of non-urgent cases, such as patients with varicose veins, travel to lessbusy districts to receive quicker treatment the pattern of workin both districts will be distorted. In addition, most generalsurgeons develop a special interest that affects the pattern ofwork referred to them.

Despite these reservations, the figures presented provide aquantitative estimate of the demand placed on a general surgicalfirm (perhaps two part time consultants and junior staff)serving a population of 100 000 in England and Wales. This isa useful background against which to plan curricula for under-graduate and postgraduate surgical teaching, as well as providinga guide to the experience that a surgical trainee may be expectedto acquire in a suitable training post. For example, in a threemonth attachment to this surgical firm a medical studentcould be expected to have seen in the wards about 18 patientsbeing treated for anal conditions (piles, fissure, or fistula) andfive cases of rectal carcinoma. In one year working on this firma junior would help in the management of, and perhaps operateon, some of the 78 patients with a benign breast lump and15 undergoing mastectomy.

The results of this study allow the effect of changes in surgicalstarring on the number of operations done by a surgeon to beassessed. A consultant surgeon serving a population of 70 000might expect to perform one thyroidectomy a month. If thesize of the population served was reduced to about 10 000 asadvocated by Bengmark11 this surgeon would do under twothyroidectomies a year, and the incidence with which a patient

8

with thyroid cancer was seen would be reduced to one patientevery six years.

Decisions about future numbers of consultant generalsurgeons and junior staff, as well as about the training of regis-trars and undergraduates, depend on estimates of the nationalworkload in general surgery. So it is important that surgeonsmake sure that the information being collected in their hospitalsis correct. We have attempted to provide some idea of the sizeof the cake; only when this has been determined can there beinformed discussion about the size of each individual slice.

We thank Dr J S A Ashley and his staff at the Office of Population,Censuses and Surveys, particularly Mrs J Pritchard, for their helpin providing much of the information used in this study, and StephenEvans, of the department of biostatistics, London Hospital MedicalCollege, and John Yates, of the health service management centre,University of Birmingham, for their helpful advice and criticism.

References1 Shipman JJ. Consultants and their future. Br MedJ 1982;284:747.2 Department of Health and Social Security. Form SH3 regional and

national summaries for 1978. London: DHSS, 1979.3 Department of Health and Social Security; Office of Population, Censuses,

and Surveys; Welsh Office. Hospital in-patient enquiry main tables1978. Series MB4 No 12. London: HMSO, 1981.

4 Ashley JSA, Collingwood J. An investigation into the urological require-ments for Great Britain. Report to the British Association of UrologicalSurgeons. London: London School of Hygiene and Tropical Medicine,1975.

5 Department of Health and Social Security; Office of Population, Censuses,and Surveys; Welsh Office. Cancer statistics registrations 1978. SeriesMB1 No 10. London: HMSO, 1982.

6 Department of Health and Social Security. Hospital medical staff, Englandand Wales, regional tables Rl-3. Statistics and Research Division 1978.London: DHSS, 1982.

7 Dowie R. National trends in domiciliary consultations. Br MedJ 1983;286:819-22.

8 Private Patients Plan. Schedule of surgical operations and procedures.London: Private Patients Plan, 1982.

" North East Thames Regional Advisory Subcommittee in General Surgery.Consultants in general surgery and urology. London : North East ThamesRegional Health Authority, 1983.

10 Butts MS, Williams DRR. Accuracy of hospital activity analysis data.Br MedJ 1982;285:506-7.

11 Whates PD, Birzgalis AR, Irving M. Accuracy of hospital activity analysiscodes. Br MedJ 1982;284:1857-8.

12 Wyllie JH, Clark CG, Alexander-Williams J, et al. Effect of cimetidineon surgery for duodenal ulcer. Lancet 1981 ;i: 1307-8.

13 Quill DS, Devlin HB, Plant JA, Denham KR, McNay RA, Morris D.Surgical operation rates: a twelve year experience in Stockton on Tees.Ann R Coll Surg Eng 1983;65:248-53.

H Bengmark S. Die Ausbildung von Chirurgen in Schweden. In: HebeierG, Feifel G, eds. Klinischer Untericht and Weiterbildung in der Chirurgie.Berlin: Springer-Verlag, 1978:123-41.

(Accepted 22 July 1983)

Printed in Great Bri tain by Bourne Offset Ltd., Iver, Bucks.