authority. I would hope that the consultant would escape involvement in any civil action that a patient
might bring against the hospital (and possibly the general practitioner) if it could be shown that preventable damage had occurred because neces-
from the introduction of population screening for the risk of Down's syndrome. CHARLOTTE FLEMING Bristol Maternity
withheld. havebeen sary drugs
Ho,spital, Bristol
BS2 8EG
CntaChmsr,624-7. Depi abrtmetofv
Brownlow Medical Centre, London N 11I 2BDLeiLaotr,
~~~~~~~~~~DAVID
S EJOSSE
1 Lauder I. Auditing necropsies. BMJ 1991;303:1214-5. (16
2November.) 2Department of Health. Report on confidential enquiri'es 'nto maternal
deaths in the United Kingdom 1985-7. London: HMSO, 1991. 3 Campling EA, Devino HB, Lunn JN. The report of the nati'onal deaths 1989. London: confidential into perioerative of Surgeons, 1990. Collegeenquiry 4 Rushton DI. West Midlands perinatal mortality survey 1987. An .audit of 300 perinatal autopsies. BrJOseGneol19;8
~~~~~~~~~~~Royal
GOLDIE
Southmead Hospital, Bristol BS IO 5NB
1 Wilkie
P, Sibbald B, Raftery J, Anderson 5, Freeling P. Prescribing at the hospital-general practice interface. 1. Hospital outpatient dispensing policies in England. BMJ
Risk of Do n' syndrome and amniocentesis rate
1 Wald NJ, Cockle HS, Densum JW, Nanchahal K, Royston P, Chard T. Maternal serum for Down's syndrome in 1988;297:883-7. pregnancy. 2 Lewis M, Faed MJW, Howie PW. Screening for Down's syndrome based on individual risk. BMJ7 3 Spencer K. Evaluation of an assay of the free 0l-subunit of choriogonadotrophin and its potential value in screening for
~~~early
1992;304:29-31. January.) (4
0
~~~~~~~~~~
SIR, -Several articles on assessing the risk of Down's syndrome on the basis of maternal age
BMJ7
screening
Communication of results of
nlecropsies
1991;303:551-3.-IthipaePulWityndclags SIR ,-nterpprPalWhtyndclags
Clin Chem 1991;37:809-14. ~~Down's syndrome. 4 Williams ES. for Down's Keatinge RM,
Screening
syndrome,
inadequate provision of results of necropsy to clinicians; we support their recoinmendations regarding the provision of this in-
expose the
1991;303:1063.fomtotognrlpaiines fraint eea rciinr.
BMJ 5 Murday V, Slack J. Screening for Down's syndrome in the North
East Thames region. BMJ7 1985;291: 1315-8.
and biochemical markers in maternal serum have
Necropsy reports are needed to complete the picture of a patient's death, but lack of resources in local pathology departments, the expense of pur-
emphasised the sensitivit and specificity (that is,chsncone'rprt,adide,tequly true positive and false positive rates) that may be Auditing necropsies of coroners' necropsies (which has been quesexpected by offering amniocentesis to women tioned2) mnay be a serious obstacle. General practiwhose risk of bearing a fetus with the syndrome SIR, -Lauder highlights the value of necropsy in tioners should, surely, receive a copy of all relevant exceeds a predefined cut off limit." It seems to be audits of clinical practice.I The purpose and reports free of charge. Funding needs to be assumed, firstly, that all or most obstetricians perceived value of necropsy will be different for the' identified to implement these recommendations in would be prepared to adopt the same limit for pathologist, clinician, relatives, and coroner. Thus the light of the report The Autopsy and Audit.3I offering amniocentesis and, secondly, that most the need to audit clinical practice may well be an Whitty and colleagues' paper concerns necroppatients would accept such an offer. This second insufficient reason for relatives to consent to a sies other than those done by coroners; such assumption has recently been questioned.' necropsy while an inadequate necropsy may deter necropsies are performed after only 3'% of all We examined both these factors in Bristol, clinicians from making further' requests. deaths, and the paper therefore addresses only a where assessment of the risks of Down's syndrome Though some pathologists may be reluctant to fraction of the problem faced by general practiderived from maternal serum atfetoprotein concen- admit that inadequate necropsies are performed, tioners when trying to obtain information on tration and maternal age has been available on few attempts have been made to access the quality patients of theirs who have died. Knowledge of the request for women over 30 (based on the report by of necropsies. The most notable exceptions have time, place, and cause of death is essential for Murday and Slack5) since 1985. We examined the been the confidential inquiries into maternal administrative purposes -for example, for cancelfigures for 1989 as this was the first year for which deaths and the national confidential enquiry into ling appointments, following up relatives and res ults were available on computer, thus faciliperioperative deaths, which have identified con- carers, and auditing both the quality of care and siderable deficiencies.3 Perinatal necropsies were the quality of death certification. tating the analysis. The study was in two parts. The first was a audited in the West Midlands regional perinatal We know that general practitioners would like to survey of stated policies on offering amniocentesis mortality survey4; only half attained what was a receive this information routinely,' but presently among the 15 consultant obstetricians practising in relatively low minimum standard. This is a matter no mechanism exists for its provision. The existing Avon. The second was an analysis of the numbers of concern as the findings may influence the health service information systems do not permit of estimates of risk in various numerical risk reproductive behaviour of the parents or the the combination of general practitioner's name, groups and the numbers of patients actually pro- management of future pregnancies'. ,patient's name, and cause of death. In Newcastle The coroner's necropsy raises further problems upon Tyne, with the help of the district health ceeding to amniocentesis within each risk group. Three obstetricians stated that they would offer as there is rarely a perceived need for ancillary authority (supplying the patient's name and cause amniocentesis if the risk was ~, 1: 100,. five if it was investigations. This is not helped by the respons-e of death) and family health services authority ~_>1:200, one if it was ~_ 1:250, and two if it was of coroners to requests for payments for investi- (adding the general practitioner's name), we are 7s' 1: 300, while four did not use a fixed figure. Thus gations that may have to be met from laboratorY preparing to create lists that include over 90% of most chose a level of risk ~_- 1:200. The table shows budgets. These vary from no response (because the deaths. We shall shortly circulate such lists to that only 14 (39%) of 36 patients with a reported coroner believes that they are an integral part of the general practitioners and evaluate th'e benefits of risk ~, 1:200 actually underwent amniocentesis and examination) to an arbitrary fee (irrespective of the providing them. that only 34 (3 1%) of 110 with a reported risk work entailed) to acceptance of the Home Office ANITA BERLIN 1: 300 underwent amniocentesis. The reasons for rates as published by the BMA. There is also a lack JOHN SPENCER of uniformity in the use of specialist pathologists. Division of Primary Health Care, Rates of amni'ocentesi s associated with different risks of The Royal College of Pathologists has notified the Medical School, Dowzn's syndrome Home Office of specialist- paediatric pathologists Newcastle upon Tyne NE2 4HH Dvso fEieilg n ulcHat, on whom a coroner can call when appropriate. The RIBOA No of No (%) of use of this facility varies. The British PaediatriC Newcastle upon Tyne Medical School amniocenteses Reported risk reports Pathology Association is surveying its members to assess the extent to which this facility is being used 1 Whitty P, Parker C, Prieto-Ramos F, Al-Kharusi S. Communica1:100 8 4 (50) tion of results of necropsies in North East Thames region. < 1: 100-sy 1: 200 28 10 (36) by individual coroners. BMJ 199 1;303:1244-6. (16 November.) 74 20 (27) If we are to promote necropsy we must ensure
might bring against the hospital (and possibly the general practitioner) if it could be shown that preventable damage had occurred because neces-
from the introduction of population screening for the risk of Down's syndrome. CHARLOTTE FLEMING Bristol Maternity
withheld. havebeen sary drugs
Ho,spital, Bristol
BS2 8EG
CntaChmsr,624-7. Depi abrtmetofv
Brownlow Medical Centre, London N 11I 2BDLeiLaotr,
~~~~~~~~~~DAVID
S EJOSSE
1 Lauder I. Auditing necropsies. BMJ 1991;303:1214-5. (16
2November.) 2Department of Health. Report on confidential enquiri'es 'nto maternal
deaths in the United Kingdom 1985-7. London: HMSO, 1991. 3 Campling EA, Devino HB, Lunn JN. The report of the nati'onal deaths 1989. London: confidential into perioerative of Surgeons, 1990. Collegeenquiry 4 Rushton DI. West Midlands perinatal mortality survey 1987. An .audit of 300 perinatal autopsies. BrJOseGneol19;8
~~~~~~~~~~~Royal
GOLDIE
Southmead Hospital, Bristol BS IO 5NB
1 Wilkie
P, Sibbald B, Raftery J, Anderson 5, Freeling P. Prescribing at the hospital-general practice interface. 1. Hospital outpatient dispensing policies in England. BMJ
Risk of Do n' syndrome and amniocentesis rate
1 Wald NJ, Cockle HS, Densum JW, Nanchahal K, Royston P, Chard T. Maternal serum for Down's syndrome in 1988;297:883-7. pregnancy. 2 Lewis M, Faed MJW, Howie PW. Screening for Down's syndrome based on individual risk. BMJ7 3 Spencer K. Evaluation of an assay of the free 0l-subunit of choriogonadotrophin and its potential value in screening for
~~~early
1992;304:29-31. January.) (4
0
~~~~~~~~~~
SIR, -Several articles on assessing the risk of Down's syndrome on the basis of maternal age
BMJ7
screening
Communication of results of
nlecropsies
1991;303:551-3.-IthipaePulWityndclags SIR ,-nterpprPalWhtyndclags
Clin Chem 1991;37:809-14. ~~Down's syndrome. 4 Williams ES. for Down's Keatinge RM,
Screening
syndrome,
inadequate provision of results of necropsy to clinicians; we support their recoinmendations regarding the provision of this in-
expose the
1991;303:1063.fomtotognrlpaiines fraint eea rciinr.
BMJ 5 Murday V, Slack J. Screening for Down's syndrome in the North
East Thames region. BMJ7 1985;291: 1315-8.
and biochemical markers in maternal serum have
Necropsy reports are needed to complete the picture of a patient's death, but lack of resources in local pathology departments, the expense of pur-
emphasised the sensitivit and specificity (that is,chsncone'rprt,adide,tequly true positive and false positive rates) that may be Auditing necropsies of coroners' necropsies (which has been quesexpected by offering amniocentesis to women tioned2) mnay be a serious obstacle. General practiwhose risk of bearing a fetus with the syndrome SIR, -Lauder highlights the value of necropsy in tioners should, surely, receive a copy of all relevant exceeds a predefined cut off limit." It seems to be audits of clinical practice.I The purpose and reports free of charge. Funding needs to be assumed, firstly, that all or most obstetricians perceived value of necropsy will be different for the' identified to implement these recommendations in would be prepared to adopt the same limit for pathologist, clinician, relatives, and coroner. Thus the light of the report The Autopsy and Audit.3I offering amniocentesis and, secondly, that most the need to audit clinical practice may well be an Whitty and colleagues' paper concerns necroppatients would accept such an offer. This second insufficient reason for relatives to consent to a sies other than those done by coroners; such assumption has recently been questioned.' necropsy while an inadequate necropsy may deter necropsies are performed after only 3'% of all We examined both these factors in Bristol, clinicians from making further' requests. deaths, and the paper therefore addresses only a where assessment of the risks of Down's syndrome Though some pathologists may be reluctant to fraction of the problem faced by general practiderived from maternal serum atfetoprotein concen- admit that inadequate necropsies are performed, tioners when trying to obtain information on tration and maternal age has been available on few attempts have been made to access the quality patients of theirs who have died. Knowledge of the request for women over 30 (based on the report by of necropsies. The most notable exceptions have time, place, and cause of death is essential for Murday and Slack5) since 1985. We examined the been the confidential inquiries into maternal administrative purposes -for example, for cancelfigures for 1989 as this was the first year for which deaths and the national confidential enquiry into ling appointments, following up relatives and res ults were available on computer, thus faciliperioperative deaths, which have identified con- carers, and auditing both the quality of care and siderable deficiencies.3 Perinatal necropsies were the quality of death certification. tating the analysis. The study was in two parts. The first was a audited in the West Midlands regional perinatal We know that general practitioners would like to survey of stated policies on offering amniocentesis mortality survey4; only half attained what was a receive this information routinely,' but presently among the 15 consultant obstetricians practising in relatively low minimum standard. This is a matter no mechanism exists for its provision. The existing Avon. The second was an analysis of the numbers of concern as the findings may influence the health service information systems do not permit of estimates of risk in various numerical risk reproductive behaviour of the parents or the the combination of general practitioner's name, groups and the numbers of patients actually pro- management of future pregnancies'. ,patient's name, and cause of death. In Newcastle The coroner's necropsy raises further problems upon Tyne, with the help of the district health ceeding to amniocentesis within each risk group. Three obstetricians stated that they would offer as there is rarely a perceived need for ancillary authority (supplying the patient's name and cause amniocentesis if the risk was ~, 1: 100,. five if it was investigations. This is not helped by the respons-e of death) and family health services authority ~_>1:200, one if it was ~_ 1:250, and two if it was of coroners to requests for payments for investi- (adding the general practitioner's name), we are 7s' 1: 300, while four did not use a fixed figure. Thus gations that may have to be met from laboratorY preparing to create lists that include over 90% of most chose a level of risk ~_- 1:200. The table shows budgets. These vary from no response (because the deaths. We shall shortly circulate such lists to that only 14 (39%) of 36 patients with a reported coroner believes that they are an integral part of the general practitioners and evaluate th'e benefits of risk ~, 1:200 actually underwent amniocentesis and examination) to an arbitrary fee (irrespective of the providing them. that only 34 (3 1%) of 110 with a reported risk work entailed) to acceptance of the Home Office ANITA BERLIN 1: 300 underwent amniocentesis. The reasons for rates as published by the BMA. There is also a lack JOHN SPENCER of uniformity in the use of specialist pathologists. Division of Primary Health Care, Rates of amni'ocentesi s associated with different risks of The Royal College of Pathologists has notified the Medical School, Dowzn's syndrome Home Office of specialist- paediatric pathologists Newcastle upon Tyne NE2 4HH Dvso fEieilg n ulcHat, on whom a coroner can call when appropriate. The RIBOA No of No (%) of use of this facility varies. The British PaediatriC Newcastle upon Tyne Medical School amniocenteses Reported risk reports Pathology Association is surveying its members to assess the extent to which this facility is being used 1 Whitty P, Parker C, Prieto-Ramos F, Al-Kharusi S. Communica1:100 8 4 (50) tion of results of necropsies in North East Thames region. < 1: 100-sy 1: 200 28 10 (36) by individual coroners. BMJ 199 1;303:1244-6. (16 November.) 74 20 (27) If we are to promote necropsy we must ensure
