Article Text

Download PDFPDF

Senior house officer withdrawals from hospital posts: a questionnaire survey
  1. D S Vickery1,
  2. G G Bodiwala2
  1. 1Department of Accident and emergency Medicine, Gloucestershire Royal Hospital, Gloucester, UK
  2. 2Accident and Emergency Department, Leicestershire Royal Infirmary, University Hospitals of Leicester NHS Trust, Leicester, UK
  1. Correspondence to:
 Mr D Vickery, Department of Accident and Emergency Medicine, Gloucestershire Royal Hospital, Great Western Road, Gloucester GL1 3NN, UK;


Objectives: To assess the nature and number of unexpected withdrawal by senior house officers (SHOs) after acceptance of a hospital post at interview, and to investigate the reasons and possible solutions.

Design: Postal questionnaire based study.

Subjects: Medical staffing departments, accident and emergency (A&E) consultants, and withdrawing SHOs in England.

Main outcome measures: Unexpected withdrawals by hospital specialty, three year trend in A&E departments, notice and reasons given, and action taken.

Results: 39% of medical staffing departments reported unexpected withdrawals in a broad spectrum of specialties for February 1998. In the specialty of A&E medicine this occurred in 34% of departments. Overall 72% of A&E departments had experienced this problem over a three year period, and the trend is increasing. The majority of A&E consultants (70%) took no action, and there was a lack of consensus among all respondents on the appropriate course of action to prevent this escalating problem.

Conclusions: Unexpected SHO withdrawal is a substantial issue in hospital medicine and has been increasing in A&E medicine. Measures to prevent this national problem are urgently needed.

  • medical staffing
  • senior house officer

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Shortfalls in medical manpower may have profound effects on the quality of care delivered to patients as well as increasing the burden on other medical staff and healthcare professionals.1 Senior house officers (SHOs) make up the second largest number of doctors in hospital medicine, and as a result are a significant and essential component of manpower planning for the NHS.2 Unexpected withdrawal of an SHO after the process of advertisement, shortlisting, interview, and appointment has substantial implications. The post needs to be readvertised, and the interview panel reconvened. Other candidates available to fill the vacancy at the time of interview may well have accepted a subsequent position elsewhere.

Previous studies on shortfalls in medical manpower numbers have focused on recruitment issues, difficulties with specific specialties, the impact of the New Deal, and the balance between training and service commitments.3,4

This study investigated the unexpected and often unpredictable shortfalls that occur after successful recruitment. Firstly, we aimed to quantify the number of unexpected withdrawals in all hospital specialties and examine the trend in the specialty of accident and emergency (A&E) medicine. Secondly, we explored reasons for unexpected withdrawal and discussed potential preventative measures.


In the spring of 1998 a postal questionnaire was sent to the medical staffing departments of all hospitals in England with A&E departments that employed SHOs.5 Details were requested on the specialties that had experienced unexpected SHO withdrawal prior to their contracted start date of 4 February 1998.

A separate questionnaire was sent to the A&E consultant in the same hospital, requesting the number of unexpected withdrawals by SHOs in that specialty. Where known, the reason for each SHO withdrawing, the period of notice given and the subsequent action taken with that individual doctor was detailed. The number of unexpected SHO withdrawals during the preceding three years was also requested.

A questionnaire for the SHOs that withdrew from the February 1998 A&E posts was enclosed in the letters to the medical staffing departments. Anonymity for the individual SHO and the hospital was guaranteed.

All groups were asked their opinion on six statements regarding unexpected SHO withdrawal and could answer: agree; disagree; or don't know. Additional comments and suggestions were invited.

The six statements were:

  1. Doctors are waiting until after having accepted a job offer, before making a decision as to whether it is in fact their preferred post.

  2. Doctors are using their first job offer as security against unemployment enabling them to continue to apply elsewhere.

  3. The conduct of these doctors reflects an unprofessional attitude.

  4. Clear advice on professional behaviour regarding job applications should be made available to all junior doctors.

  5. The referees of each successful applicant should be contacted in writing informing them that the doctor has been appointed.

  6. Regional centralisation of all junior doctor appointments should be considered.

Repeat questionnaires were sent to all A&E and medical staffing non-responders.


A total of 140 out of 203 (69%) replies were received from the medical staffing departments, and 142 (70%) replies from the A&E consultants. Three replies were received from the SHOs themselves.

Thirty nine per cent of the medical staffing departments reported unexpected SHO withdrawals for the February 1998 start date. They occurred in a total of 14 different specialties, of which five constituted 90% of the total (fig 1).

Figure 1

Unexpected SHO withdrawals by specialty. Values are percentages of the total withdrawals.

Thirty four per cent of the surveyed A&E departments experienced unexpected SHO withdrawals for February 1998. In 27% of these (13 of 48) more than one SHO had withdrawn. Table 1 shows the reasons for the unexpected withdrawal of 56 out of the total of 65 SHOs that withdrew. The majority of A&E consultants took no specific action with the SHOs that unexpectedly withdrew (table 2). Length of notice was available for 55 of the SHOs, of which 29% withdrew less than one month before the start date. In one department there were three withdrawals within this period. By the 4 February only four A&E departments had a residual SHO vacancy.

Table 1

Reasons given for unexpected SHO withdrawal from A&E posts in February 1998

Table 2

Action taken by the A&E consultant

Three year data on unexpected SHO withdrawal from A&E were available for 137 departments. Overall 72% (98 of 137) had experienced the problem at some stage during this time. Figure 2 shows that an upward trend is apparent, particularly from February 1997 onwards.

Figure 2

SHO withdrawals from A&E departments over a 3.5 year period.

The responses to the structured questions are shown in tables 3 and 4. Only three SHOs responded and therefore their replies cannot be taken as representative, and have not been included.

Table 3

Why is unexpected SHO withdrawal happening?

Table 4

What can be done about unexpected SHO withdrawal?


Our study shows that unexpected withdrawal by SHOs from hospital posts in England is a significant, widespread, and increasing manpower problem.

A shortfall in SHO numbers at any stage, particularly at the August and February changeover dates, can have substantial impact on patient care.6 In the increasingly common context of shift work patterns for junior doctors, covering vacant posts will impact on working hours, training issues, and overall morale. There may well be financial costs of providing locum cover, as well as the process of readvertising the post and convening interview panels.

If sufficient notice is given of withdrawal then there may be time to readvertise and appoint before the start date, or alternatively appoint candidates unsuccessful at the original interview. In this study although 48 A&E departments had SHOs unexpectedly withdrawing after interview, only four departments had an unfilled vacancy on 4 February.

The GMC recognises that doctors who fail to honour agreements to take up posts may cause a considerable disruption to the running of a hospital. Guidance in Good Medical Practice makes clear that doctors must take up posts they have accepted, unless there is time for alternative arrangements to be made.7 Contractually one months' notice needs to be given. Our study showed that in only 71% of A&E withdrawals did this occur more than one month before 4 February 1998.

Fourteen different hospital specialties experienced the problem, however the impact was greatest in the four specialties of A&E, orthopaedics, paediatrics, and obstetrics and gynaecology. Our study shows that there is an increasing trend in unexpected withdrawals from A&E posts, and there is similar anecdotal evidence for some of the other specialties. It seems likely therefore that this problem will continue to increase and have a great impact on hospital medicine.

The second part of the study was designed to examine the reasons behind this issue and explore possible solutions. There was broad agreement among the A&E consultants and medical staffing departments that junior doctors are deferring the decision about a specific post until after having accepted the offer (47% and 48%). In addition 60% and 54% thought that junior doctors were using the first job offer as security while continuing to apply elsewhere. This may reflect a lack of preparation about certain jobs, lack of career guidance and, as was felt by 80% of respondents, to reflect an unprofessional attitude.

Among the SHOs who withdrew from A&E posts before 4 February only 23% took up another job in A&E elsewhere. Conversely, 36% of those withdrawing did so because they were taking a post in another specialty altogether. This would correlate with the consensus of opinion as expressed above, and also suggest that some SHOs are securing A&E jobs while in fact pursuing alternative specialty positions. Perhaps a common set of circumstances would be to take up a surgical rotation for three years in preference to a six month A&E post.

The traditional system of medical staff selection is partly to blame. The timing of job advertisements is often unpredictable, uncoordinated, and competitive between hospitals. Unlike most other professions the successful candidate is required to accept or refuse a post at the time of interview.8 None the less the individual doctor has a professional responsibility and must acknowledge that verbal acceptance of a job is legally binding. There was almost unanimous agreement (87% and 92%) among respondents that clear advice on this must be made available to all junior doctors. A number of interview panels make this explicit with each candidate. Other methods could include advice in the BMJ classified section, and in job descriptions and application forms.

Among the A&E consultants the majority (70%) took no action with an individual SHO after withdrawal. Fourteen per cent wrote to their referees, 8% to the SHO themselves, and 3% to the GMC. There was a lack of consensus regarding the notification of referees that a candidate had been appointed. In view of the lack of coordination of the whole medical appointment process, it would seem logical to us that automatic notification of referees and a request not to provide further references would help prevent the problem. There has been suggestion that a centralised appointments system on a regional or national basis with the UCAS system as a model, might provide a solution.9 However, a clear majority of respondents (61% and 70%) were against this.

Methodological considerations

There was an approximately 30% non-response rate to the questionnaire from medical staffing departments and A&E consultants. As such it is possible that the withdrawal rates for SHOs, and the three year trend for A&E departments differs from that reported. In determining potential bias we found no obvious difference between responders and non-responders in terms of nature of department (district general hospital or teaching hospital), staffing numbers, or geographical location.

A structured set of statements was used to enable comparison of answers and a consensus response. Through wide discussion we felt that the six statements were valid and relevant to the nature of the study. In addition opportunity was given for comment and suggestions. A broad range of replies were received.

Clearly in this type of study it is important to gain the opinions and experience of SHOs. The response rate from those SHOs that withdrew from the February 1998 posts was poor. None the less we feel that the findings are significant and merit further discussion. For this to be meaningful there should be involvement of all doctors both junior and senior.


This study highlights the scale of unexpected SHO withdrawal from hospital posts and has canvassed opinion among A&E consultants and medical staffing departments. We believe that broad discussion on the whole subject is needed with strategic planning to resolve the issues as soon as possible.


DSV conceived the study, analysed the data, and wrote the first draft, DSV and GGB designed the questionnaire and modified the final draft. DSV is guarantor of the paper.



  • Funding: none.

  • Conflicts of interest: none.

Linked Articles

  • Primary Survey
    Peter Driscoll Jim Wardrope