whose pathological identification is often difficult, variable, and contentious. His observed difference in sensitivity would be more likely to reflect pathological sensitivity than clinical sensitivity. Thirdly, his example illustrating unchanged sensitivity and specificity in the clinical diagnosis of cancer ignores the changing incidence of various cancers and differing hospital populations of patients with cancer and fails to mention that the comparison standard-necropsy-has itself changed in the past 30 years. It has never been a standard procedure. The author's suggestions for the future may be of practical value in France, but they are not in the United Kingdom. Firstly, assessment of error in postmortem diagnosis requires a standard detailed postmortem examination-a process that could currently occur only in a centre of excellence with adequate staffing and funding. The error rate found could never be applied widely outside such a centre because of the variable quality of both consent and coroners' postmortem examinations. Secondly, current legislation does not allow postmortem examinations to be performed in a way that would allow proper sampling. Saracci's article appears in the Audit in Practice section of the journal, but its suggestions have little value in England and Wales without radical reform of the Human Tissues Act and the coroners' system together with considerable changes in the teaching and practice of necropsy. They are not likely to be part of audit in practice in the United Kingdom for many years. RYK JAMES M A GREEN
Department of Forensic Pathology, University of Sheffield, Sheffield I Saracci R. Is necropsy a valid monitor of clinical diagnosis
performance? BMJ 1991;303:898-900. (12 October.)
The rise of post-traumatic stress disorders SIR,-I wish to add a few comments to Gary Jackson's editorial on post-traumatic stress disorders.' This term arose out of work with American veterans of the Vietnam war, who first attracted attention because so many became spectacular social casualties, unable to take up ordinary roles and liable to violent and self destructive behaviour. In contrast, British service personnel who saw intense, albeit shortlived, fighting in the Falkland Islands have a high prevalence of post-traumatic stress disorders five years later (nearly one in four) but have unremarkable work and social lives.2 I studied peasants displaced by the war in Nicaragua, all survivors of atrocities, and found that features associated with post-traumatic stress disorder were common, but these people were nevertheless active and effective in maintaining their social world as best they could in the face of the continuing threat of further attacks.3 Indeed, this threat rendered a "symptom" of the disorder like hypervigilance adaptive. When these people did seek treatment it was for psychosomatic ailments, which are not included in the definition of the disorder. Studies of, for example, Cambodian war refugees, both in border refugee camps and in the United States, show similar findings.4 The diagnosis of post-traumatic stress disorder says little about ability to function. Medical models, focusing on individual psychopathology and liable to Western ethnocentrism, have inherent limitations in capturing the complex ways in which individual people, communities, and indeed whole societies abroad register overwhelming tragedy, socialise their grief, and reconstitute a meaningful existence. What seems central, and anthropological reports concur, is that
BMJ VOLUME 303
16 NOVEMBER 1991
it is in a social setting that the traumatised who need help reveal themselves and the processes that determine how victims become survivors (as most do) are played out over time. Arguably, a telling example of what happens when social networks are not supportive arose when the American veterans came home to find that their nation and, more subtly, their families were disowning their guilt for the war and blaming them instead. This rejection was surely an important factor in the subsequent genesis of their social dysfunction. At the moment the diagnosis ofpost-traumatic stress disorder does not address these issues. D SUMMERFIELD Medical Foundation For The Care of Victims of Torture, London NW5 3EJ 1 Jackson G. The rise of post-traumatic stress disorders. BMJ 1991;303:533-4. (7 September.) 2 O'Brien LS, Hughes SJ. Symptoms of post-traumatic stress disorder in Falkland veterans 5 years after the conffict. BrJ Psychiatry 1991;159:135-41. 3 Summerfield D, Toser L. "Low intensity" war and mental trauma in Nicaragua: a study in a rural community. Medicine and War 1991;7:84-99. 4 Mollica R, Wyshak G, Lavelle J. The psychosocial impact of war trauma and torture on Southeast Asian refugees. Am J Psychiatry 1987;144:1567-72.
Postoperative feeding SIR,-Nicholas D Maynard and David J Bihari highlight the advantages of enteral nutrition and the dangers of parenteral nutrition. Enteral nutition, which can be given to a much wider range of patients postoperatively than traditional teaching dictates, protects mucosal integrity and reduces bacterial translocation, whereas parenteral nutrition is associated with problems with the catheter and hepatobiliary and other complications. Against this background the authors' assertion that "the time has come for formal comparisons of enteral with parenteral nutrition in severely ill patients" is inappropriate. Unless new evidence emerges to suggest a particular advantage from specific nutrients administered intravenously such a study would be unethical. There is no doubt that patients who are unable to eat must be given nutritional support, nutritional support should be administered enterally, and parenteral nutrition is required only when intestinal function is unavailable or inadequate. Consequently most parenteral nutrition is supplemental rather than total, and the term total parenteral nutrition should be restricted to those few patients who have no intestine or no intestinal function. Finally, with reference to the authors' remarks about the dangers of Intralipid it is worth pointing out that use of this energy source in the short term permits supplemental and possibly total parenteral feeding through a peripheral vein, thus avoiding the more serious complications. Furthermore, during long term central parenteral nutrition the risk of venous thrombosis is considerably reduced when some of the energy requirements are provided by Intralipid. C R PENNINGTON Departments of Pharmacology and Clinical Pharmacology, Ninewells Hospital and Medical School, Dundee DDl 9SY I Maynard ND, Bihari DJ. Postoperative feeding. BMJ 1991;303: 1007-8. (26 October.) 2 Pithie AD, Pennington CR. The incidence, aetiology and management of central vein thrombosis during parenteral nutrition. Clin Nutr 1987;6:151-3.
Disasters in the inner city SIR,-I read David Adshead's personal view on general practice in an inner city on the same day that I learnt that an enthusiastic doctor with an
impressive research record had resigned from a partnership in inner city Leicester to take up a post in semirural practice. David Adshead says that the fires that occurred were only one factor in his decision to leave his inner city practice. I would like to illustrate from our recent experience some of the other factors that may have played a part. Workload- Demand for appointments and home visits has risen in the past year-for example, night visits during April to September this year increased by 16% over the same period in 1990. Attempts to educate patients towards using the service more responsibly are frequently met with incomprehension or hostility. Violence-Although we and our staff have not yet been physically harmed, obscene language and aggressive behaviour are common both in the reception area and in telephone conversations. Cnrme-We have reported to the police five burglaries or acts of vandalism against practice property or personal property in the past month. Income-In the first full year of the new contract the practice's net profit rose by only 6% despite our achieving higher rate targets and offering a full range of other income generating services. Premises-We work from a grossly inadequate building. The cost rent scheme will meet only 60% of the cost of bringing it to an acceptable standard. It is a daunting decision to fund the rest of the cost from a practice income that may be declining. Recruitment-We received just six applications for a partnership vacancy despite our close association with a university department of general practice. None of these problems amount to a disaster. Taken together they illustrate the morale sapping strain on inner city doctors. We feel abandoned by both the Royal College of General Practitioners and the BMA. Training seems to produce general practitioners with a vocation to practise in comfortable market towns. If the fall in applications to vocational training schemes and the steady loss of talented doctors from urban deprived areas continue I foresee the implosion of general practice in these areas as doctors are squeezed by mounting demand and dwindling resources. ADRIAN HASTINGS
Saffron Group Practice, Leicester LE2 6UL 1 Adshead D. Disasters in the inner city. BMJ7 1991;303:101. (19
October.)
The right to know SIR,-Though many would agree that the new legislation allowing patients to have access to their written medical records is on balance a positive step, the question of the need for modifying medical records to make them more comprehensible to patients is not as clear cut as Paul McLaren's editorial seems to imply. ' Medical records must serve primarily as a medium for condensing clinical information in a form that can be rapidly assimilated by other health care workers who are concerned with a patient's care. Although summary records held by patients may be helpful in some circumstances, they cannot be expected to replace conventional medical records. It is not merely the jargon inevitably used in medical records that will be incomprehensible to many outside medicine; many of the concepts of disease processes and their treatment are complex and cannot be adequately explained to those with little medical knowledge in a format constrained by the need to be concise. Before concentrating our attention on the way in which we write our medical records we must ask what motivates patients to seek access to their records. Two probable reasons are that patients wish to know more about their condition and its
1271
treatment than they do and that they suspect that important information is being withheld from them. Both are likely to arise from suboptimal communication between medical practitioner and patient. The onus is on medical practitioners to ensure that there is good communication with their patients so as to allow for a full explanation of their condition and its management and for patients to vent fears and anxieties. Building up trust between a patient and doctor and exchanging information require time. In a world where medical performance is increasingly measured by numbers of patients seen, doctors must strongly argue the case for adequate time for communicating with their patients. Perhaps if this is realised both patients and providers of care will be saved much of the time and money that might otherwise be spent in seeking access to medical records. Contrary to McLaren's suggestion, low rates of requests for access to medical records might then indeed be a cause for relief. ROBERT A SHEEHAN-DARE
DIermatology Department,
sure it's nothing serious but we need to do some tests" or "Doctor knows best." It will confuse only if it is given badly, and the doctor should be responsible for assessing the impact of that information on the patient. If you accept the ethical argument that patients have a right to know everything about their state of health then the study of how best to present that informatioon becomes an important scientific endeavour. A record for use by doctor and patient should be the most efficient way of doing this. Short has highlighted some of the issues that need to be further clarified in clinical studies before more specific recommendations can be made. PAUL McLAREN
United M\edical and Dental School, Gus's Hospital, London SE1 9RT
SIR, -Paul McLaren's topical and well referenced editorial unfortunately offers no support for his most crucial statement: "New ways of keeping records have been studied and their effects on patients found to be positive."' Many doctors will want to know what these new ways entail. Do they avoid the use of unintelligible abbreviations and symbols without taking up more of the doctor's time? Do they separate the facts about the patient's health from the conjectures involved in the process of diagnosis? Patients certainly have a right to the facts but not, in my view, to the conjectures, which might give rise to more confusion and anxiety than enlightenment. DAVID SHORT Aberdeen AB9 2P1. I McLaren P. 'Ihe right to know. BM1J 1991;303:937-8. (19
October.)
AUTHOR'S REPLY, -The summary records held by patients that I referred to in my editorial are examples of new ways of keeping medical records.' General guidance on how to present the records are reported in the studies that I referenced. Other authors, such as Tattersall,2 have produced more specific recommendations on the style and terminology of records for particular groups of patients. Where style has not been modified significant confusion has resulted for patients.' Writing a record for use by patients and doctors will take more time. This would, however, be more than compensated for by the improved communication with the patient and other professionals and improved adherence to treatment. Summary records held by patients have concentrated on facts about health rather than conjecture on diagnosis, but this need not be the case. Communicating to patients the conjectures concerned in diagnosis and treatment can assist in building a trusting therapeutic relationship. Obtaining informed consent for a surgical procedure means giving patients more than the facts about their condition, and explaining possible side effects entails sharing conjecture with patients. Is David Short suggesting that patients do not have a right to such information? Such sharing of conjecture may indeed lead to anxiety for the patients in the short term. It is, however, less likely to result in longlasting distress than giving already anxious patients bland reassurance such as "I'm 1272
D C DUNN
Director, Confidential Comparative Audit Service, Royal College of Surgeons of England, London WC2A 3PN
Seven stages of audit SIR,-I was disappointed to see the content of
Audit Views recently as most of the reports were I AlcLaren P. The right to know. BMJ 1991;303:937-8. (19 October.) 2 Tattersall R. Writing for and to patients. Diabetic Med 1990;7: 917-9. 3 Rutherford W, Gabriel R. Audit of outpatient letters. BMJ 1991;303:968. (19 October.)
General Infirmary at Leeds, Leeds LS I 3EX 1 MilcLaren P. The right to know. BMIJ 1991;303:937-8. (19 October.)
resources required for policing would be at the expense of care. We believe that this voluntary activity should be encouraged, and this will be achieved only by maintaining confidentiality. Comparative audit, which is the natural extension of the desire to audit oneself, would be crushed in its infancy by heavy handed external compulsion.
Voluntary confidential audit of outcome of surgery SIR,-In June a confidential comparative audit meeting was held at the Royal College of Surgeons of England to discuss findings of a questionnaire survey. Questionnaires had been sent to 1025 general surgical consultants in England and Wales, asking them for details of their workload, throughput, mortality, and complication rates. Participants were given a confidential number and returned their data anonymously. So far 149 (71%) of the 209 consultants who applied for numbers have returned data. At the meeting we discussed 137 270 inpatient admissions with a mortality of 1-95% (range 0-5%) and a morbidity of 7% (0-22%) and 113930 operations with a mortality of 1 1% (0-5%) and a morbidity of 6% (1-23%). Low complication rates could be explained by poor audit, low risk surgery, or brilliant surgical technique. High complication rates could be explained by fastidious audit, high risk surgery, or poor technique. High risk groups included patients having complex surgery, elderly patients, and patients undergoing emergency surgery or complex procedures such as arterial or colonic surgery. Individual surgeons had sufficient information to understand which of these factors was relevant to their results. They reported 2585 colonic resections with a mortality of 6%, a complication rate of 28%, and a rate of anastomotic leak of 1-4%. The commonest complications were related to the abdominal wound (6 4%). The meeting analysed 3043 open and 1653 laparoscopic cholecystectomies. Of the laparoscopic cholecystectomies, 86 were converted to open operations. Sixteen deaths were reported after the open operation (0-52%). Although this mortality was low, it was four times that reported after laparoscopic surgery (0 12%; two deaths). Morbidity after these two procedures was 13-9% and 8 0% respectively. The morbidity after the conversion operation was 9-8%. This exercise and the confidential inquiry into perioperative deaths have shown that surgeons will collect and compare details of complications and are anxious to take part in such an exercise in the pursuit of excellence. It has been suggested that comparisons should not be confidential and that surgeons should be judged by their complication rates. A review of surgical results, however, requires adequate surgical training, and an open review would require extensive and expensive policing. Without this surgeons would be under pressure to minimise their reported complication rates. This is not in the public interest. The
not true audits.' A working group of the World Health Organisation defined audit as a seven stage procedure, the most vital element being to ensure an improvement (indicating change) in care .by reassessing results after change had come about based on the initial data gathering.2 Of the 15 articles summarised, only two seemed to reflect a change in the quality of care. Ironically, one of the papers summarised was bewailing the failure to apply basic principles of correct audit.3 Most of the articles summarised are not articles on audit but articles on data gathering. The guidelines issued by the working group of WHO merit reiteration. They are (1) problem identification; (2) setting priorities; (3) determining methodology; (4) setting criteria and standards; (5) comparing performance with standards; (6) designing and implementing remedial action; and (7) re-evaluating the quality of care. As can be seen from these seven stages, the studies that were summarised either did not perform the full audit cycle or were summarised in such a way that this did not seem apparent to the reader. Audit has been around for long enough now that we should at least start to see articles in print that describe a complete audit cycle, including the re-evaluation of care after changes have been made to protocols, criteria, and standards. A P PRESLEY
Glevum Way Surgery,
Abbeydale, Gloucester GL4 9BL I Audit views. BMJ 1991;303:904. (12 October.) 2 WHO Working Group. The organisation of quality assurance. Quality Assurance in Health Care 1989;1: 111-23. 3 Ewart HE. The audit of cervical cytology screening programmes: discussion paper. Journal of the Royal Society of Medicine
1991;84:488-90.
First words SIR,-Minerva may not be correct in saying that Lord Macaulay's first words were, "Thank you, madam, the agony is abated."' These were perhaps his first recorded words, but Lord Bertrand Russell recollected a great aunt of his who knew Lord Macaulay's nurse, who claimed that he said the following at the age of 3: "In the ideal constitution there should be a just balance of forces; while on the one hand the will of the people should be adequately represented in a popular house, on the other hand the wisdom of those who through their stake in the country have acquired a certain stability of political judgement, should not remain unheard."2 PAUL WENTWORTH Brantford General Hospital,
Brantford, Ontario N3R lG9, Canada 1 Minerva. BMJ 1991;303:934. (12 October.) 2 Feinberg B, ed. The collected stories of Bertrand Russell. London: George Allen and Unwin, 1972:282-3.
BMJ
VOLUME
303
16 NOVEMBER 1991
Department of Forensic Pathology, University of Sheffield, Sheffield I Saracci R. Is necropsy a valid monitor of clinical diagnosis
performance? BMJ 1991;303:898-900. (12 October.)
The rise of post-traumatic stress disorders SIR,-I wish to add a few comments to Gary Jackson's editorial on post-traumatic stress disorders.' This term arose out of work with American veterans of the Vietnam war, who first attracted attention because so many became spectacular social casualties, unable to take up ordinary roles and liable to violent and self destructive behaviour. In contrast, British service personnel who saw intense, albeit shortlived, fighting in the Falkland Islands have a high prevalence of post-traumatic stress disorders five years later (nearly one in four) but have unremarkable work and social lives.2 I studied peasants displaced by the war in Nicaragua, all survivors of atrocities, and found that features associated with post-traumatic stress disorder were common, but these people were nevertheless active and effective in maintaining their social world as best they could in the face of the continuing threat of further attacks.3 Indeed, this threat rendered a "symptom" of the disorder like hypervigilance adaptive. When these people did seek treatment it was for psychosomatic ailments, which are not included in the definition of the disorder. Studies of, for example, Cambodian war refugees, both in border refugee camps and in the United States, show similar findings.4 The diagnosis of post-traumatic stress disorder says little about ability to function. Medical models, focusing on individual psychopathology and liable to Western ethnocentrism, have inherent limitations in capturing the complex ways in which individual people, communities, and indeed whole societies abroad register overwhelming tragedy, socialise their grief, and reconstitute a meaningful existence. What seems central, and anthropological reports concur, is that
BMJ VOLUME 303
16 NOVEMBER 1991
it is in a social setting that the traumatised who need help reveal themselves and the processes that determine how victims become survivors (as most do) are played out over time. Arguably, a telling example of what happens when social networks are not supportive arose when the American veterans came home to find that their nation and, more subtly, their families were disowning their guilt for the war and blaming them instead. This rejection was surely an important factor in the subsequent genesis of their social dysfunction. At the moment the diagnosis ofpost-traumatic stress disorder does not address these issues. D SUMMERFIELD Medical Foundation For The Care of Victims of Torture, London NW5 3EJ 1 Jackson G. The rise of post-traumatic stress disorders. BMJ 1991;303:533-4. (7 September.) 2 O'Brien LS, Hughes SJ. Symptoms of post-traumatic stress disorder in Falkland veterans 5 years after the conffict. BrJ Psychiatry 1991;159:135-41. 3 Summerfield D, Toser L. "Low intensity" war and mental trauma in Nicaragua: a study in a rural community. Medicine and War 1991;7:84-99. 4 Mollica R, Wyshak G, Lavelle J. The psychosocial impact of war trauma and torture on Southeast Asian refugees. Am J Psychiatry 1987;144:1567-72.
Postoperative feeding SIR,-Nicholas D Maynard and David J Bihari highlight the advantages of enteral nutrition and the dangers of parenteral nutrition. Enteral nutition, which can be given to a much wider range of patients postoperatively than traditional teaching dictates, protects mucosal integrity and reduces bacterial translocation, whereas parenteral nutrition is associated with problems with the catheter and hepatobiliary and other complications. Against this background the authors' assertion that "the time has come for formal comparisons of enteral with parenteral nutrition in severely ill patients" is inappropriate. Unless new evidence emerges to suggest a particular advantage from specific nutrients administered intravenously such a study would be unethical. There is no doubt that patients who are unable to eat must be given nutritional support, nutritional support should be administered enterally, and parenteral nutrition is required only when intestinal function is unavailable or inadequate. Consequently most parenteral nutrition is supplemental rather than total, and the term total parenteral nutrition should be restricted to those few patients who have no intestine or no intestinal function. Finally, with reference to the authors' remarks about the dangers of Intralipid it is worth pointing out that use of this energy source in the short term permits supplemental and possibly total parenteral feeding through a peripheral vein, thus avoiding the more serious complications. Furthermore, during long term central parenteral nutrition the risk of venous thrombosis is considerably reduced when some of the energy requirements are provided by Intralipid. C R PENNINGTON Departments of Pharmacology and Clinical Pharmacology, Ninewells Hospital and Medical School, Dundee DDl 9SY I Maynard ND, Bihari DJ. Postoperative feeding. BMJ 1991;303: 1007-8. (26 October.) 2 Pithie AD, Pennington CR. The incidence, aetiology and management of central vein thrombosis during parenteral nutrition. Clin Nutr 1987;6:151-3.
Disasters in the inner city SIR,-I read David Adshead's personal view on general practice in an inner city on the same day that I learnt that an enthusiastic doctor with an
impressive research record had resigned from a partnership in inner city Leicester to take up a post in semirural practice. David Adshead says that the fires that occurred were only one factor in his decision to leave his inner city practice. I would like to illustrate from our recent experience some of the other factors that may have played a part. Workload- Demand for appointments and home visits has risen in the past year-for example, night visits during April to September this year increased by 16% over the same period in 1990. Attempts to educate patients towards using the service more responsibly are frequently met with incomprehension or hostility. Violence-Although we and our staff have not yet been physically harmed, obscene language and aggressive behaviour are common both in the reception area and in telephone conversations. Cnrme-We have reported to the police five burglaries or acts of vandalism against practice property or personal property in the past month. Income-In the first full year of the new contract the practice's net profit rose by only 6% despite our achieving higher rate targets and offering a full range of other income generating services. Premises-We work from a grossly inadequate building. The cost rent scheme will meet only 60% of the cost of bringing it to an acceptable standard. It is a daunting decision to fund the rest of the cost from a practice income that may be declining. Recruitment-We received just six applications for a partnership vacancy despite our close association with a university department of general practice. None of these problems amount to a disaster. Taken together they illustrate the morale sapping strain on inner city doctors. We feel abandoned by both the Royal College of General Practitioners and the BMA. Training seems to produce general practitioners with a vocation to practise in comfortable market towns. If the fall in applications to vocational training schemes and the steady loss of talented doctors from urban deprived areas continue I foresee the implosion of general practice in these areas as doctors are squeezed by mounting demand and dwindling resources. ADRIAN HASTINGS
Saffron Group Practice, Leicester LE2 6UL 1 Adshead D. Disasters in the inner city. BMJ7 1991;303:101. (19
October.)
The right to know SIR,-Though many would agree that the new legislation allowing patients to have access to their written medical records is on balance a positive step, the question of the need for modifying medical records to make them more comprehensible to patients is not as clear cut as Paul McLaren's editorial seems to imply. ' Medical records must serve primarily as a medium for condensing clinical information in a form that can be rapidly assimilated by other health care workers who are concerned with a patient's care. Although summary records held by patients may be helpful in some circumstances, they cannot be expected to replace conventional medical records. It is not merely the jargon inevitably used in medical records that will be incomprehensible to many outside medicine; many of the concepts of disease processes and their treatment are complex and cannot be adequately explained to those with little medical knowledge in a format constrained by the need to be concise. Before concentrating our attention on the way in which we write our medical records we must ask what motivates patients to seek access to their records. Two probable reasons are that patients wish to know more about their condition and its
1271
treatment than they do and that they suspect that important information is being withheld from them. Both are likely to arise from suboptimal communication between medical practitioner and patient. The onus is on medical practitioners to ensure that there is good communication with their patients so as to allow for a full explanation of their condition and its management and for patients to vent fears and anxieties. Building up trust between a patient and doctor and exchanging information require time. In a world where medical performance is increasingly measured by numbers of patients seen, doctors must strongly argue the case for adequate time for communicating with their patients. Perhaps if this is realised both patients and providers of care will be saved much of the time and money that might otherwise be spent in seeking access to medical records. Contrary to McLaren's suggestion, low rates of requests for access to medical records might then indeed be a cause for relief. ROBERT A SHEEHAN-DARE
DIermatology Department,
sure it's nothing serious but we need to do some tests" or "Doctor knows best." It will confuse only if it is given badly, and the doctor should be responsible for assessing the impact of that information on the patient. If you accept the ethical argument that patients have a right to know everything about their state of health then the study of how best to present that informatioon becomes an important scientific endeavour. A record for use by doctor and patient should be the most efficient way of doing this. Short has highlighted some of the issues that need to be further clarified in clinical studies before more specific recommendations can be made. PAUL McLAREN
United M\edical and Dental School, Gus's Hospital, London SE1 9RT
SIR, -Paul McLaren's topical and well referenced editorial unfortunately offers no support for his most crucial statement: "New ways of keeping records have been studied and their effects on patients found to be positive."' Many doctors will want to know what these new ways entail. Do they avoid the use of unintelligible abbreviations and symbols without taking up more of the doctor's time? Do they separate the facts about the patient's health from the conjectures involved in the process of diagnosis? Patients certainly have a right to the facts but not, in my view, to the conjectures, which might give rise to more confusion and anxiety than enlightenment. DAVID SHORT Aberdeen AB9 2P1. I McLaren P. 'Ihe right to know. BM1J 1991;303:937-8. (19
October.)
AUTHOR'S REPLY, -The summary records held by patients that I referred to in my editorial are examples of new ways of keeping medical records.' General guidance on how to present the records are reported in the studies that I referenced. Other authors, such as Tattersall,2 have produced more specific recommendations on the style and terminology of records for particular groups of patients. Where style has not been modified significant confusion has resulted for patients.' Writing a record for use by patients and doctors will take more time. This would, however, be more than compensated for by the improved communication with the patient and other professionals and improved adherence to treatment. Summary records held by patients have concentrated on facts about health rather than conjecture on diagnosis, but this need not be the case. Communicating to patients the conjectures concerned in diagnosis and treatment can assist in building a trusting therapeutic relationship. Obtaining informed consent for a surgical procedure means giving patients more than the facts about their condition, and explaining possible side effects entails sharing conjecture with patients. Is David Short suggesting that patients do not have a right to such information? Such sharing of conjecture may indeed lead to anxiety for the patients in the short term. It is, however, less likely to result in longlasting distress than giving already anxious patients bland reassurance such as "I'm 1272
D C DUNN
Director, Confidential Comparative Audit Service, Royal College of Surgeons of England, London WC2A 3PN
Seven stages of audit SIR,-I was disappointed to see the content of
Audit Views recently as most of the reports were I AlcLaren P. The right to know. BMJ 1991;303:937-8. (19 October.) 2 Tattersall R. Writing for and to patients. Diabetic Med 1990;7: 917-9. 3 Rutherford W, Gabriel R. Audit of outpatient letters. BMJ 1991;303:968. (19 October.)
General Infirmary at Leeds, Leeds LS I 3EX 1 MilcLaren P. The right to know. BMIJ 1991;303:937-8. (19 October.)
resources required for policing would be at the expense of care. We believe that this voluntary activity should be encouraged, and this will be achieved only by maintaining confidentiality. Comparative audit, which is the natural extension of the desire to audit oneself, would be crushed in its infancy by heavy handed external compulsion.
Voluntary confidential audit of outcome of surgery SIR,-In June a confidential comparative audit meeting was held at the Royal College of Surgeons of England to discuss findings of a questionnaire survey. Questionnaires had been sent to 1025 general surgical consultants in England and Wales, asking them for details of their workload, throughput, mortality, and complication rates. Participants were given a confidential number and returned their data anonymously. So far 149 (71%) of the 209 consultants who applied for numbers have returned data. At the meeting we discussed 137 270 inpatient admissions with a mortality of 1-95% (range 0-5%) and a morbidity of 7% (0-22%) and 113930 operations with a mortality of 1 1% (0-5%) and a morbidity of 6% (1-23%). Low complication rates could be explained by poor audit, low risk surgery, or brilliant surgical technique. High complication rates could be explained by fastidious audit, high risk surgery, or poor technique. High risk groups included patients having complex surgery, elderly patients, and patients undergoing emergency surgery or complex procedures such as arterial or colonic surgery. Individual surgeons had sufficient information to understand which of these factors was relevant to their results. They reported 2585 colonic resections with a mortality of 6%, a complication rate of 28%, and a rate of anastomotic leak of 1-4%. The commonest complications were related to the abdominal wound (6 4%). The meeting analysed 3043 open and 1653 laparoscopic cholecystectomies. Of the laparoscopic cholecystectomies, 86 were converted to open operations. Sixteen deaths were reported after the open operation (0-52%). Although this mortality was low, it was four times that reported after laparoscopic surgery (0 12%; two deaths). Morbidity after these two procedures was 13-9% and 8 0% respectively. The morbidity after the conversion operation was 9-8%. This exercise and the confidential inquiry into perioperative deaths have shown that surgeons will collect and compare details of complications and are anxious to take part in such an exercise in the pursuit of excellence. It has been suggested that comparisons should not be confidential and that surgeons should be judged by their complication rates. A review of surgical results, however, requires adequate surgical training, and an open review would require extensive and expensive policing. Without this surgeons would be under pressure to minimise their reported complication rates. This is not in the public interest. The
not true audits.' A working group of the World Health Organisation defined audit as a seven stage procedure, the most vital element being to ensure an improvement (indicating change) in care .by reassessing results after change had come about based on the initial data gathering.2 Of the 15 articles summarised, only two seemed to reflect a change in the quality of care. Ironically, one of the papers summarised was bewailing the failure to apply basic principles of correct audit.3 Most of the articles summarised are not articles on audit but articles on data gathering. The guidelines issued by the working group of WHO merit reiteration. They are (1) problem identification; (2) setting priorities; (3) determining methodology; (4) setting criteria and standards; (5) comparing performance with standards; (6) designing and implementing remedial action; and (7) re-evaluating the quality of care. As can be seen from these seven stages, the studies that were summarised either did not perform the full audit cycle or were summarised in such a way that this did not seem apparent to the reader. Audit has been around for long enough now that we should at least start to see articles in print that describe a complete audit cycle, including the re-evaluation of care after changes have been made to protocols, criteria, and standards. A P PRESLEY
Glevum Way Surgery,
Abbeydale, Gloucester GL4 9BL I Audit views. BMJ 1991;303:904. (12 October.) 2 WHO Working Group. The organisation of quality assurance. Quality Assurance in Health Care 1989;1: 111-23. 3 Ewart HE. The audit of cervical cytology screening programmes: discussion paper. Journal of the Royal Society of Medicine
1991;84:488-90.
First words SIR,-Minerva may not be correct in saying that Lord Macaulay's first words were, "Thank you, madam, the agony is abated."' These were perhaps his first recorded words, but Lord Bertrand Russell recollected a great aunt of his who knew Lord Macaulay's nurse, who claimed that he said the following at the age of 3: "In the ideal constitution there should be a just balance of forces; while on the one hand the will of the people should be adequately represented in a popular house, on the other hand the wisdom of those who through their stake in the country have acquired a certain stability of political judgement, should not remain unheard."2 PAUL WENTWORTH Brantford General Hospital,
Brantford, Ontario N3R lG9, Canada 1 Minerva. BMJ 1991;303:934. (12 October.) 2 Feinberg B, ed. The collected stories of Bertrand Russell. London: George Allen and Unwin, 1972:282-3.
BMJ
VOLUME
303
16 NOVEMBER 1991
