Intended for healthcare professionals

Education And Debate

Senior house officers in medicine: postal survey of training and work experience

BMJ 1997; 314 doi: (Published 08 March 1997) Cite this as: BMJ 1997;314:740
  1. Pamela J Baldwin, senior researchera,
  2. Ray W Newton, associate deanb,
  3. Graham Buckley, executive directorc,
  4. Margaret A Roberts, consultant physician in geriatric medicined,
  5. Marjory Dodd, research assistanta
  1. a Working Minds Project, Blackford Pavilion, Astley Ainslie Hospital, Grange Loan, Edinburgh EH9 2HL
  2. b Ninewells Hospital and Medical School Dundee DD1 9SY
  3. c Scottish Council for Postgraduate Medical and Dental Education, Edinburgh EH2 1JE
  4. d Victoria Geriatric Unit Glasgow G41 3DX
  1. Correspondence to: Dr Baldwin
  • Accepted 31 December 1996


Objectives: To describe working conditions for senior house officers in medicine in Scotland and to relate these to the quality of clinical training they receive.

Design: Postal questionnaire survey.

Subjects: All senior house officers in medicine and related specialties in post in Scotland in October 1995 (n=437); 252 (58%) respondents.

Main outcome measures: Questionnaires covered hours, working patterns, measures of workload, an attitudes to work scale, and experience of education and training.

Results: In the week before the questionnaire, doctors on rotas had worked a mean of 7.4 (95% confidence interval 5.8 to 9.0) hours in excess of their contracts, compared with 3.7 (2.0 to 5.5) hours for those on partial shifts. The most common reason for this was “the needs of the patients or the service.” Those on partial shifts reported significantly less continuity of care with patients than those on rotas (Mann-Whitney U test, z=-4.2, P<0.0001) or full shifts (z=-2.08, P=0.03). Doctors in general medicine reported significantly higher measures of workload (number of acute admissions, number of times called out, and fewest hours' uninterrupted sleep) than those in subspecialties. Consultants' clinical teaching and style of conducting a ward round were significantly related to factors extracted from the attitudes to work scale.

Conclusions: The quality of senior house officers' training is detrimentally affected by a variety of conditions, especially the need for closer support and supervision, the need for greater feedback, and the lack of time that consultants have to dedicate to clinical training. Efforts should be made to improve these conditions and to reinforce a close working relationship between trainee and supervising consultant.

Key messages

  • Many senior house officers continue to work long hours in excess of their contracts

  • Senior house officers see partial shifts as detrimental to their own health, patient care, and clinical training

  • Acute general medicine has a higher intensity of work than the allied specialties

  • The quality of consultant feedback has an important influence on perception of learning, ability to cope, and relationships between junior and senior staff


The “new deal,” formulated as an initiative to reduce junior doctors' hours, has necessitated changes in working patterns for senior house officers, with the introduction of full and partial shifts.1 At the same time there are indications that senior house officers in medicine in Scotland are becoming increasingly dissatisfied with their training experience.

A working group on doctors in basic medical training, with representatives from the Royal College of Physicians of Edinburgh, the Royal College of Physicians and Surgeons of Glasgow, and the Scottish Postgraduate Council for Medical and Dental Education, initiated a survey of doctors in training in medicine and engaged an independent agency to carry it out. The aim was to determine current working conditions and educational experience and how these influence the quality of clinical training. The full report, which covers careers, clinical experience, educational structures, and the specific needs of women doctors in medicine, will appear as a joint publication from these institutions.2 This paper indicates some principal findings.


The working group constructed a questionnaire that contained questions on type of post, hours, workload, training experience, and careers; an existing attitudes to work questionnaire which has been used elsewhere for junior doctors3; and a section for comments in response to three open questions. It was piloted among junior staff, and the final version was sent to all 437 medical senior house officers in post in Scotland in October 1995 with a covering letter and a stamped addressed envelope for return of the questionnaire. The questionnaires were anonymous, and respondents were asked to return them within three weeks. Two reminders were sent out, resulting in a total response of 252 (58%). Statistical analyses used spss for Windows (version 6.0).


Of the respondents, 43% (108) were working in a teaching hospital with a university department of general medicine; 40% (101) were in district general hospitals; and 17% (43) were in teaching hospitals without university departments. There were more men (55%) than women (45%), but this sex ratio does not differ significantly from that in the latest available census data (1994) for senior house officers in hospital medicine (57:43). The median age of the survey population was 26 (range 23 to 43) years.

At the time of the survey, 44% (111) of the respondents were working in general medicine; 18% (45) in geriatric medicine; 7% (18) in cardiology; 6% (15) in infectious diseases; 5% (13) each in renal medicine and gastroenterology; 4% (10) each in dermatology and neurology, and the remaining 7%% (18) in a variety of subspecialties, each less than 3%. Again, this pattern does not differ appreciably from the latest published figures for senior house officers in medicine in Scotland. In these respects, the sample can be viewed as representative of the total population.

Contracted hours and patterns of working

Table 1) shows the hours which the respondents were contracted to work and respondents' actual working pattern. For assessment of hours worked in practice, subjects were also asked to report how long they had actually been on their feet working during the previous week. The responses ranged from 0 to 100 hours (mean 56.9 (54.8 to 59.1) hours).

Table 1

Working pattern and contracted hours of senior house officers in medicine in Scotland, 1995. Values are numbers (percentages)

View this table:

Working in excess of contracted hours

Senior house officers were asked about hours worked in excess of their contract: 48% (121) had worked at least some excess hours in the preceding week. There were differences among the different working patterns, with those on rotas working the greatest number of excess hours (mean 7.4 (5.8 to 9.0) v 3.7 (2.0 to 5.5) for those on partial shifts).

Those regularly working excess hours were asked to identify to what extent several possible reasons applied to them; table 2) shows the results. The most strongly endorsed reason was “the needs of the patients or service,” which 83% (209) reported as applying quite a bit or strongly. Across specialties, those working in general medicine and geriatric medicine most often cited the need to cover for an absent colleague (49% (45/92) of those in general medicine and 46% (13/28) in geriatric medicine compared with 31% (23/75) of those in the other specialties; F=3.97, P=0.02).

Table 2

Reasons for senior house officers working in excess of contracted hours. Values are numbers (percentages)

View this table:

When subjects were asked for open comments on what might deter them from a career in medicine, long hours of work emerged as the most important factor in putting them off. Their comments showed concern that this will not diminish with increased promotion—for example, “My consultants are at present working until 9 pm most nights, therefore not an attractive prospect.”

On call work

Subjects were asked to record several measures of workload during their most recent night on call. Table 3) shows that these differed among the specialties. Those in general medicine reported the highest numbers of calls both to a clinical area and to the telephone; highest number of admissions; and fewest hours' uninterrupted sleep. They were also least satisfied with their choice of medicine as a career. When all the measures of workload were examined, the number of acute admissions seemed to be a key factor. Data from the attitudes to work questionnaire for all subjects showed that the number of acute admissions was significantly negatively correlated with scores on the item “I am very satisfied with my choice of medicine as a career” (r=-0.13, P=0.03) and positively correlated with the item “I am worried about my career in this specialty” (r=0.17, P=0.007).

Table 3

Measures of workload (mean numbers during most recent night on call) and of agreement with item on attitudes to work questionnaire for senior house officers

View this table:

Continuity of care

Senior house officers were asked about their perception of continuity of care with inpatients. The majority of those on full shifts, rotas, or mixed shifts reported high continuity, with most cases being seen right through. However, those on partial shifts, of whom only 46% (16/35) said that most or all cases were seen right through, differed significantly from those on rotas (Mann-Whitney U test, z=-4.2, P=0.000) and those on full shifts (z=-2.08, P=0.03) (table 4). Comments about this included: “There is now a lack of a ‘team’ with partial shift systems, as well as loss of continuity. This has reduced job satisfaction to a minimum.”

Table 4

Continuity of care of inpatients reported by senior house officers. Values are numbers (percentages)

View this table:

Working patterns and continuity of learning

Since the type of working pattern affected perceived continuity of care for patients, we thought this might affect the learning experience. This hypothesis was tested by looking at the item on the attitudes to work questionnaire that relates to learning: “I am developing new skills.” Doctors on partial shifts agreed significantly less strongly with this item than those on rotas (z=-1.8, P=0.05).

Preferred working patterns

The doctors were asked to choose which working pattern they thought was best from three points of view: their health and personal life, patient care, and education and training.

On all three counts, most preferred some kind of rota (table 5). The data were subsequently analysed by the type of working pattern that the senior house officer was on at the time of the survey, to see if those on partial shifts (14%) differed from the rest in these preferences. They gave the same order of preference: most preferred a rota, though the response was less emphatic.

Table 5

Working pattern preferred by senior house officers. Values are numbers (percentages)

View this table:

Clinical teaching

Senior staff have different styles of teaching which may be more or less effective in training senior house officers. Among our subjects, 1% (2) reported that consultants conducted a separate teaching round; 59% (147) of consultants spontaneously discussed cases on a working round; 33% (82) only responded to questions on a working round; and 7% (18) did not discuss cases.

A simple scale was used to measure the amount of feedback that senior house officers considered they received from their current consultant on their inpatient work (table 6). Although most felt that the feedback they received was helpful, they also said that it was not enough. More than a quarter reported that feedback was both inadequate and unhelpful.

Table 6

Feedback to senior house officers on inpatient work

View this table:

Influences on feelings of competence, coping, and relationships with senior staff

To examine the factors that affect senior house officers' perception of competency, the attitudes to work inventory was subjected to a factor analysis by varimax rotation. Seven factors emerged, and the first three accounted for 38% of the variance. Factor 1 was the feeling of effective learning and competency, with high loadings for such items as “I am useful most of the time,” “I am developing new skills,” and “I use my skills to the full.” Factor 2 was the perception of not coping: “The responsibilities of my job are overwhelming,” “I regularly feel I am working beyond my capabilities,” and “I am under great pressure at work,” and factor 3 represented relationships with senior staff: “Senior doctors let me know how well I am doing,” “I can discuss work problems with senior staff,” and “I can discuss personal problems with senior staff.” Subjects were then assigned the factor scores, and the results were correlated with hypothesised influences (Spearman rank correlations). There were no significant sex differences in the factor scores.

Feedback from consultants–The reported feedback on inpatient work was significantly related to all three factors. The better the perceived feedback, the more competent the senior house officers felt themselves to be (factor 1, r=0.25, P<0.0001), the less overwhelmed they were by responsibility (factor 2, r=-0.15, P=0.01), and, understandably, the better their relationship with senior staff (factor 3, r=0.45, P<0.0001).

Ward round style–The analysis was taken one step further with ward round style, to see if this concrete measure of teaching influenced the attitudes to work in the senior house officers. Using the hierarchical scale on the style of teaching, this was correlated with the same three factors, with similar results: the less detailed the ward round, the less competent the senior house officer felt (factor 1, r=-0.26, P=0.000), the more overwhelmed by the pressure (factor 2, r=0.16, P=0.013), and the worse the relationship with the senior member of staff (factor 3, r=-0.2, P=0.002).


Response rate

The response rate in this survey (58%), although limited, seems to be representative of senior house officers in terms of sex, specialty, and working pattern. We encountered some difficulties in distributing the questionnaires: one hospital returned the questionnaires because the envelopes did not have ward numbers, even though they had names and departments; others returned the envelopes, denying that the senior house officer was there (although we later found them at the same location); and some doctors reported that they did not receive questionnaires addressed to them. This reveals something about the status of senior house officers (and the distribution of their mail).

Working patterns

There are two main concerns with the “new deal.” Firstly, despite the introduction of new contracts, many senior house officers in medicine in Scotland were still working in excess of contracted hours. Secondly, though only 14% of the sample were working partial shifts, the evidence and opinion were that partial shifts are detrimental to patient care, training, health, and personal life.

The difficulties in devising a successful reduction in junior doctors' hours are apparent. Despite the advocacy of the Junior Doctors Committee of the BMA in Shifting Work Practices,4 the partial shift has not been widely implemented in Scotland. We know that in some hospitals it has been tried and rejected by mutual agreement. Neither the high intensity of work and long hours experienced on the rotas nor the lack of continuity with patients and staff reported by those on partial shifts is conducive to high quality training. It is unlikely that a standard solution will work in all medical settings. A flexible approach to working patterns is needed, but the need to solve the problem is urgent.

Special difficulties in acute general medicine

The data highlighted the particular problem of acute general medicine, where workload on a range of measures was shown to be significantly higher than in the medical subspecialties and satisfaction was lowest. In all medical specialties, the principal difficulties in the nature of training were identified by the senior house officers as being the need for closer support and supervision; the need for greater feedback; and the lack of time that consultants have to dedicate to clinical training.

Senior house officers are aware of the increasing pressure on senior staff in terms of clinical workload and administration, but the data revealed the overwhelming importance of effective, constructive feedback for doctors in training. It was shown to have a major influence on the senior house officers' perceived competence and ability to cope on the wards. While many consultants may give such feedback regularly and spontaneously, the data show that most senior house officers consider that they receive inadequate feedback. Similarly, the more specific the teaching associated with a ward round in hospital, the more effective the learning is perceived to be, and the better able the doctor is to cope with clinical duties. This finding highlights the importance of focused teaching in the clinical setting and supports the need for a closer working relationship between the senior house officer and supervising consultant.


Members of the working group were Ray W Newton, Michael Lambert, Caroline E Whitworth (Royal College of Physicians, Edinburgh); Margaret A Roberts, Stephen Gallacher, (Royal College of Physicians and Surgeons, Glasgow); Graham Buckley (Scottish Council for Postgraduate Medical and Dental Education); and Pamela J Baldwin (Working Minds Project). We are grateful to all medical senior house officers who completed the questionnaire for this survey and also thank Mr Francis Brewis (Management Executive, NHS, Scottish Office) for providing national data on senior house officers in Scotland.

Funding: Scottish Council for Postgraduate Medical and Dental Education.

Conflict of interest: None.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
View Abstract