Skip to main content
NCBI home page
The BMJ logoLink to The BMJ
. 2001 Oct 20;323(7318):930.

Delivery by caesarean section

Increased numbers of caesareans do not match diagnoses of fetal distress

Ashis Banerjee 1,2, James Hollinshead 1,2, Edward Williams 1,2
PMCID: PMC1121449  PMID: 11693133

Editor—MacKenzie and Cooke say that the main documented indication for intrapartum caesarean section is fetal distress.1 Tuffnell et al looked at how the interval between the decision to perform an emergency caesarean section and the time of delivery affected fetal outcome.2 They concluded that a delayed decision to delivery time was an unusual cause of neonatal problems.

To help put these papers into context, we looked at the West Sussex births database of 152 284 births between 1980 and 1999. Over this period, the diagnosis of fetal distress has doubled from 6.4% to 13% of all births, and the percentage of caesarean sections has increased from 9.7% to 20.2% of all births. The proportion of caesarean sections where a diagnosis of fetal distress is recorded has increased from 15.4% to 19.3% of all caesarean sections. This implies that increasing numbers of caesarean sections are being performed for fetal distress. The doubling in the rate of diagnosis of fetal distress is unlikely to reflect a true increase in the incidence of this condition. There is no evidence that changes in maternal health or clinical care could explain such a dramatic rise. This implies that the diagnosis of fetal distress is imprecise.

We then investigated if the diagnosis of fetal distress was a valid test in predicting poor fetal outcome, which was defined as a one minute Apgar score of less than 5. From 1980 to 1999, only 18% of cases diagnosed with fetal distress subsequently had a poor fetal outcome (table). Over the same period, 74% of babies born with an Apgar score of less than 5 did not have a prepartum diagnosis of fetal distress. Therefore, the diagnosis of fetal distress is not a sensitive test for predicting poor fetal outcome (sensitivity 26%, specificity 92%). This is true even if we examine the most recent data for 1996-8 (sensitivity 30%, specificity 88%). The resulting excess of false positive results tends to encourage unnecessary caesarean sections.

Therefore, our analysis implies three things. Firstly, increasing numbers of caesarean sections are being performed for fetal distress. Secondly, the diagnosis of fetal distress is imprecise and a poor predictor of fetal outcome. Finally, the result is a tendency for unnecessary caesarean sections. Until firm and consistent diagnostic criteria for fetal distress are agreed, it is difficult to interpret studies looking at the decision to delivery time for emergency caesarean sections.

Table.

Prepartum diagnosis of fetal distress compared with Apgar score at 1 minute: West Sussex births data, 1980-99. Values are numbers of deliveries unless stated otherwise

Apgar score <5 Apgar score >5
No of deliveries:
 Fetal distress diagnosed 2553  11 469
 Fetal distress not diagnosed 7284 130 978
Sensitivity 26.0%
Specificity 91.9%
Positive predictive value 18.2%
Negative predictive value 94.8%
Open in a new tab

References

  • 1.MacKenzie IZ, Cooke I. Prospective 12 month study of 30 minute decision to delivery intervals for “emergency” caesarean section. Observational case series. BMJ. 2001;322:1334–1335. doi: 10.1136/bmj.322.7298.1334. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Tuffnell DJ, Wilkinson K, Beresford N. Interval between decision and delivery by caesarean section—are current standards achievable? BMJ. 2001;322:1330–1333. doi: 10.1136/bmj.322.7298.1330. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Effective system of mobilisation is used in Singapore

Horng Yen Wee 1,2, Swee Chong Quek 1,2

Editor—We agree with Tuffnell et al that an emergency caesarean section is a complex multidisciplinary procedure requiring the timely assembly of staff.1-1 One factor that was not addressed in their paper was the team's response time. During our specialist registrar rotations in the United Kingdom, we observed that many units rely on paging each individual on the team. In an emergency, this process can be cumbersome, requiring a valuable member of the delivery suite team making multiple phone calls. Furthermore, the recipient may be unaware of the urgency of the page.

The KK Women's and Children's Hospital in Singapore is the largest maternity hospital in the country, delivering about 15 000 babies a year. It uses a public announcement system to mobilise the team when an emergency caesarean section is required. Each team member is required to be aware of their own specific roles in such an emergency and respond appropriately.

Once a decision for an emergency caesarean section is made, a “code green” is activated via the telephone operator through a public announcement system. All team members receive the urgent message simultaneously. This ensures the rapid assembly of all the necessary staff at the appropriate locations including the mobilisation of support midwives and personnel involved in the transfer of the patient to the operating theatre. In addition the theatre staff can begin the process of receiving the patient and preparing for surgery.

Yeap et al examined 113 consecutive emergency caesarean sections over one year, accounting for 0.7% of the total 16 267 deliveries (sixteenth FIGO world congress of gynaecology and obstetrics, Washington, DC, September 2000). The rate of caesarean sections for this period was 20.7%. The mean time interval from decision to delivery was 14.9 minutes. The arbitrary time standard of 30 minutes was achieved in 112 out of 113 cases. In the remaining case, delivery was achieved in 32 minutes. Fetal distress on cardiotocography and fetal bradycardia accounted for 16% and 39% of cases, respectively. The code green model is simple and reliable, eliciting a swift and coordinated response, thereby reducing the time interval to delivery. Although such a system may not be directly applicable to units that do not use a public announcement system, the principle may readily be modified to existing paging facilities.

References

  • 1-1.Tuffnell DJ, Wilkinson K, Beresford N. Interval between decision and delivery by caesarean section—are current standards achievable? Observational case series. BMJ. 2001;322:1330–1333. doi: 10.1136/bmj.322.7298.1330. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Caesarean section urgency—use clinical definitions

S Michael Kinsella 1,2,3,4,5,6, D Nuala Lucas 1,2,3,4,5,6, Steven M Yentis 1,2,3,4,5,6, Anita Holdcroft 1,2,3,4,5,6, Anne E May 1,2,3,4,5,6, Michael Wee 1,2,3,4,5,6, P Neville Robinson 1,2,3,4,5,6

Editor—The papers by Tuffnell et al and MacKenzie and Cooke, together with the accompanying editorial by James, reinforce the concept that the definition of urgency of caesarean section should not incorporate the intended interval between decision and delivery.2-1–2-4 If, on the one hand, the actual interval is greater than the intended interval there may be medicolegal consequences; on the other hand, there could be a danger to patients if undue haste is used to achieve a preselected interval that is not clinically warranted. The degree of urgency may even decrease after the decision—for example, after successful use of intrauterine resuscitation.

The Royal Colleges of Obstetrics and Gynaecology and Anaesthetists and the Obstetric Anaesthetists Association support the use of a new four point urgency category for caesarean section that is based on clinical definitions,2-5 which produces more consistent interobserver rating than the intended interval.2-4

The 30 minute standard for the interval between decision and delivery, which originated in the United States as a compromise that would allow senior staff not resident in hospital to attend, is too long to allow delivery of a live infant in cases of complete placental failure—for example, abruption. We disagree with James that this arbitrary figure is a useful audit standard. We do, however, support the use of data on the interval for audit. In a draft version of standard 10.3.1, from the clinical negligence scheme for trusts, it is stated that a standard is set for the interval which should be annually reviewed and remedial measures instituted if it was not met. If data in the urgency category are collected that use the agreed definitions2-4 as well as data on the interval (as was done for last year's national sentinel audit of caesarean section), individual units may produce evidence based local standards for interval according to urgency category. These will account for individual circumstances—for example, the distance of the operating theatre from the labour rooms, staffing levels, etc—and avoid the problem of trying (and failing) to achieve a 30 minute standard for all caesarean sections for fetal distress.

Footnotes

The authors are the same as in reference 4 below and in reference 14 in the paper by Tuffnell et al.

References

  • 2-1.Tuffnell DJ, Wilkinson K, Beresford N. Interval between decision and delivery by caesarean section—are current standards achievable? Observational case series. BMJ. 2001;322:1330–1333. doi: 10.1136/bmj.322.7298.1330. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-2.MacKenzie IZ, Cooke I. Prospective 12 month study of 30 minute decision to delivery intervals for “emergency” caesarean section. BMJ. 2001;322:1334–1335. doi: 10.1136/bmj.322.7298.1334. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-3.James D. Caesarean section for fetal distress. BMJ. 2001;322:1316–1317. doi: 10.1136/bmj.322.7298.1316. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-4.Lucas DN, Yentis SM, Kinsella SM, Holdcroft A, May AE, Wee M, et al. Urgency of caesarean section: a new classification. J R Soc Med. 2000;93:346–350. doi: 10.1177/014107680009300703. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2-5.Anonymous. Classification of urgency of caesarean section. RCOG News. 2000;7:64. [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Don't throw the baby out with the bath water

Malcolm Griffiths 1,2, B Sharma 1,2

Editor—We congratulate Tuffnell et al and MacKenzie and Cooke on their papers concerning the audit standard of 30 minutes from decision to delivery.3-1,3-2 We agree that there is no published scientific evidence in favour of a 30 minute limit, although this standard has been widely accepted. But, as pointed out, in many pressing cases more urgent delivery will be necessary to provide a good outcome and in others delivery beyond 30 minutes will have little impact on outcome.

Similar audits in our own unit have produced comparable results. Most recently, we have shown a mean time of 45 minutes with only 20% achieved in less than 30 minutes. But what we, Tuffnell et al, and MacKenzie and Cooke have shown is that it is possible safely to achieve delivery within 30 minutes of making the decision—provided that inertia is removed and organisational issues are overcome. Accordingly, we would advise against throwing the baby out with the bath water by dismissing the 30 minute standard as unrealistic or not achievable. We believe that Tuffnell et al and MacKenzie and Cooke may have introduced some confusion—clinical negligence schemes for trusts do not require audit to the 30 minute standard, but rather audit to a standard.

Like Tuffnell and MacKenzie, one of us (MG) regularly acts as an expert witness. His experience has been that in 40 cases of litigation about cerebral palsy he has not come across a case where the 30 minute standard was crucial in favour of the claimant. On the contrary, substandard care is most frequently determined by failure to decide that caesarean section is needed. Conversely, cases of cerebral palsy arising from unexpected and catastrophic events (such as major abruption)—where a decision to delivery interval of 20-30 minutes is achieved, but poor outcome cannot be avoided—frequently thwart claimants and provide a defence.

Similar to Tuffnell et al and MacKenzie and Cooke, we found that babies born quickest often have the poorest condition at birth. This is not cause and effect, but rather a reflection of the fact that staff proceed with the greatest haste in the most dire of cases.

Finally, we applaud Tuffnell for his conflict of interest declaration and wonder whether one by MacKenzie was not even more appropriate. Not only does he provide expert reports, but he does so largely on behalf of defendants and is also a counsel member of one of the three medical defence organisations in the United Kingdom.

Footnotes

MG regularly provides expert reports for claimants and defendants and is a member of the doctors' support group of the Association for Victims of Medical Accidents.

References

  • 3-1.Tuffnell DJ, Wilkinson K, Beresford N. Interval between decision and delivery by caesarean section—are current standards achievable? Observational case series. BMJ. 2001;322:1330–1333. doi: 10.1136/bmj.322.7298.1330. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3-2.MacKenzie IZ, Cooke I. Prospective 12 month study of 30 minute decision to delivery intervals for “emergency” caesarean section. BMJ. 2001;322:1334–1335. doi: 10.1136/bmj.322.7298.1334. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Interval between decision and delivery can be shortened

Alvaro Aguillera 1, Patricia Barlow 1, Serge Rozenberg 1

Editor—With reference to the paper by Tuffnell et al on the interval between decision and delivery by caesarean section,4-1 in 1983 we established a protocol that we called “catastrophic caesarean section” for extremely urgent sections in our institution. The procedures are conducted in suboptimal conditions of sterility to reduce extraction time to a strict minimum (no shaving of the patient, no scrubbing of the obstetrician, chlorhexidine disinfectant sprayed on the patient's abdomen, prophylaxis with broad spectrum antibiotics, general anaesthesia unless a locoregional anaesthetic is already in place).

Retrospectively, we analysed the caesarean sections performed between 1995 and 1997 to evaluate their frequency and neonatal and maternal mortality and morbidity. In a series of 818 sections that were performed during that period, 55 sections (0.8% of all deliveries) were conducted in extreme urgency (24 for acute fetal distress, 19 for haemorrhage, 9 for prolapsed cord, 2 for breech presentation at complete dilatation, 1 for uterine rupture). We matched every catastrophic caesarean section to the next elective procedure and to the next procedure that was conducted in a semi-emergency, which was defined as a caesarean section that was neither elective nor in the catastrophic protocol (most of these were performed for lack of progression of labour or fetal distress). The three groups were comparable in terms of maternal age, gestational age, neonatal weight, and the proportion of primiparous women.

The mean time (from decision to birth) required to perform a catastrophic section was 13 minutes (range 4-27). Although the localisation of the operating theatre is ideal —in the delivery room—and well trained staff is always present, in about 20% of the catastrophic sections more than 15 minutes were necessary to deliver the baby. The results in term of neonatal mortality and morbidity and maternal morbidity are summarised in the table.

Table.

Neonatal morbidity and mortality and maternal data by type of caesarean section. Values are numbers unless stated otherwise

Catastrophic (n=55) Semi-urgent (n=55) Elective (n=55) P value
Neonatal data
 Mean neonatal weight (g) 2983 3199 3097 NS
Admitted to neonatal intensive care unit 37 21 10 <0.05
Mean Apgar score (1 min):
 All 5.8 7.9 8.6
 >37 weeks 6.4 8.27 8.6 <0.05
 <37 weeks 3.7 5.75 8.3
 <32 weeks  1 4.7  7
Apgar score (1 min):
 ⩽3 17  4  1 <0.05
 ⩽7 21  8  2 <0.05
Mean Apgar score (5 min):
 All 8.4 9.5 9.8
 >37 weeks 8.5 9.8 9.6 <0.05
 <37 weeks 6.9 8.6 10
 <32 weeks 3.75 8.3  9
Apgar score at 5 min ⩽7 15  2  0
Intubated 13  4  0 <0.05
Born dead  1  0  0
Postnatal mortality  2  0  0
Maternal data
 Antibioprophylaxis 54 51  40* <0.05
All infections 16 13  6   0.058
Haemorrhage >1.5 l  6  2  3 <0.05
Temperature >38°C  5  5  3 NS
Open in a new tab
4-150

n=51. 

In this small series of patients there was no increase in the frequency of endometritis, urinary tract infections, or wound infections, but the cumulative number of infections tended to be slightly higher in this group. The number of haemorrhages was also higher in this group, but five out of six patients already had a haemorrhage before the start of the section. We believe, therefore, that it is often possible to conduct a catastrophic caesarean section in less than 15 minutes with acceptable maternal morbidity, provided that such a protocol is developed. We advocate its use in all maternity departments.

References

  • 4-1.Tuffnell DJ, Wilkinson K, Beresford N. Interval between decision and delivery by caesarean section—are current standards achievable? Observational case series. BMJ. 2001;322:1330–1333. doi: 10.1136/bmj.322.7298.1330. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Explanation for absence of adverse outcomes is needed

Vincent Richman 1

Editor—The absence of adverse outcomes from delaying urgent deliveries in the papers by Tuffnell et al and MacKenzie and Cooke is anomalous and requires explanation.5-1,5-2 The first possible explanation is that the urgent deliveries might not actually have been necessary. This would be consistent with an extensive literature that maintains that current caesarean rates are unnecessarily high.5-3

The second possible explanation concerns the timing metric, the time from decision to caesarean delivery. Patient outcomes depend on the condition of the patient, not on the specific decision time of the obstetrician. Tuffnell et al assume that the time of decision corresponds to a specific and comparable health status among the patients. The absence of an outcome effect arising from 20 minute delays could reflect 20 minute variations in the obstetricians' decision time relative to the patient's condition.

The third possible explanation concerns the measure of adverse infant outcomes. Adverse infant outcomes were measured in terms of medical decisions, the admissions to special care. Perhaps a direct measure of the infant medical status, the Apgar score, would have produced different results.

The fourth possible explanation concerns the impact of nosocomial factors. Tuffnell et al ignored hospital factors and assumed that the decisions both to perform caesarean sections and to admit infants to special care are standardised and consistent at all times throughout the week and throughout the year. Decisions about caesarean sections are complex, and the rates of caesarean section are characterised by hospital variations in practice patterns.5-4,5-5

References

  • 5-1.Tuffnell DJ, Wilkinson K, Beresford N. Interval between decision and delivery by caesarean section—are current standards achievable? Observational case series. BMJ. 2001;322:1330–1333. doi: 10.1136/bmj.322.7298.1330. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5-2.MacKenzie IZ, Cooke I. Prospective 12 month study of 30 minute decision to delivery intervals for “emergency” caesarean section. BMJ. 2001;322:1334–1335. doi: 10.1136/bmj.322.7298.1334. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5-3.Flamm BL. Caesarean section: a worldwide epidemic? Birth. 2000;27:139–140. doi: 10.1046/j.1523-536x.2000.00139.x. [DOI] [PubMed] [Google Scholar]
  • 5-4.Richman VV. Lack of local reflection of national changes in caesarean delivery rates: the Canadian experience. Am J Obstet Gynecol. 1999;180:393–395. doi: 10.1016/s0002-9378(99)70220-6. [DOI] [PubMed] [Google Scholar]
  • 5-5.Richman VV. Setting goals for reductions in Canadian caesarean rates: Benchmarking medical practice patterns. Am J Obstet Gynecol. 1999;181:635–637. doi: 10.1016/s0002-9378(99)70505-3. [DOI] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Longer term follow up of surviving babies is required

P O D Pharoah 1,2, M J Platt 1,2

Editor—James, in his editorial on caesarean section for fetal distress, says that intrapartum hypoxia results in death in about 0.5 in 1000 pregnancies and cerebral palsy in one in 1000 pregnancies.6-1 As the birth prevalence of cerebral palsy is 2.0-2.5/1000, it is implicit that intrapartum hypoxia is the cause of 40-50% of all cerebral palsy; an estimate that is higher than generally accepted. In the case-control studies from Western Australia and Oxford, only 9% and 10% of cases of cerebral palsy of gestational age of 37 weeks were attributable to intrapartum hypoxia.6-2,6-3

The main rationale for emergency caesarean section is to prevent hypoxic ischaemic cerebral impairment that becomes clinically manifest as cerebral palsy. In some cases a rapid delivery of the distressed infant will limit the cerebral impairment, but it would be prudent not to be overoptimistic about what emergency caesarean section may achieve. In most cases of cerebral palsy the cerebral impairment occurs prepartum, and the fetal distress exhibited may be the result and not the cause of the impairment. It is a reasonable assumption that a fetus who has had a cerebral insult prepartum is less able to tolerate the stresses of labour than an intact one and, therefore, is more likely to show signs of distress. In such cases, emergency caesarean section may increase rather than decrease the birth prevalence of cerebral palsy by enabling infants impaired prepartum to survive labour.

James notes some of the outcome measures that have been examined in auditing the time taken to caesarean section. These audit measures, such as pH value of the umbilical artery and Apgar scores, are all made immediately after birth. What is required as part of the audit is a longer term follow up of the surviving infants. This will be time consuming and will require a large sample to show an effect.

References

  • 6-1.James D. Caesarean section for fetal distress. BMJ. 2001;322:1316–1317. doi: 10.1136/bmj.322.7298.1316. . (2 June.) [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6-2.Blair E, Stanley F. When can cerebral palsy be prevented? The general causal hypotheses by multivariate analysis of a case control study. Paediatr Perinat Epidemiol. 1993;7:272–301. doi: 10.1111/j.1365-3016.1993.tb00405.x. [DOI] [PubMed] [Google Scholar]
  • 6-3.Gaffiley G, Flavell V, Johnson A, Squier M, Sellers S. Cerebral palsy and neonatal encephalopathy. Arch Dis Child. 1994;70:F195–F200. doi: 10.1136/fn.70.3.f195. [DOI] [PMC free article] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Author's reply

I Z Mackenzie 1

Editor—I am delighted at the response to the study Cooke and I reported on decision to delivery intervals for emergency caesarean sections. Several respondents have illustrated the lack of precision in defining fetal distress and emergency caesarean section.

Banerjee et al reported an increase in caesarean section for fetal distress from 15.4% to 19.3% between 1980 and 1999; the proportion of caesarean sections for intrapartum fetal distress in Oxford remained constant at approximately 14% in 1976 and 1996, although this represents an increase from 1.1% to 2.2% of all women in labour in the two years quoted.

The definition of “emergency” caesarean section is even more indistinct, as illustrated by the letters from Wee, citing a study reporting a rate of emergency caesarean sections of only 0.7%, and Kinsella et al, of 0.8%. These data most probably refer to “crash” caesarean section, which in our study represented 0.4% of women delivered; of the 22 when the interval was recorded 17 (77%) were delivered within 30 minutes. This was achieved with a system that had been in place for more than 25 years of obstetric and anaesthetic staff being resident on or near the delivery unit 24 hours of the day, without the need for a public address system.

The conclusion reached by Sharma and Griffiths about maternal safety cannot be drawn either from the report of Tuffnell et al or from our study. Similarly, neither study sought to dismiss the concept of establishing a standard for setting a decision to delivery interval: both have challenged how the decision of 30 minutes had been arrived at. Sharma and Griffiths suggest that the fetal condition at birth was a reflection of the severity of the fetal distress that had prompted the caesarean section; they have overlooked the fact that the trend for improving fetal condition, as assessed by cord gas values, occurred with emergency caesarean when fetal distress was not a factor dictating the need for delivery.

Their concerns about expert witness report and clinical research warrants a comment. Expert witnesses provide their views for the benefit of the court and should be unbiased. To suggest that those who conduct clinical research have a potential conflict of interests is a cynical proposition. All of my consultant colleagues in post for more than 12 months prepare reports for medicolegal purposes. This implies that most consultants publishing studies on clinical issues should be admitting a conflict of interest; surely this cannot be correct.

BMJ. 2001 Oct 20;323(7318):930.

Authors' reply

D J Tuffnell 1, K Wilkinson 1, N Beresford 1

Editor—We are pleased that our audit has created so much interest. Some of the correspondents have looked at the indication for delivery and criticised our rate of caesarean section. This assumes that all caesarean sections should only be performed with clear evidence of fetal compromise. This misses the point that in clinical practice we have to deal with risk of fetal compromise and not certainty. If one considers the term breech trial, at least 14 caesarean sections are performed to avoid one case of neonatal morbidity, yet no rational observer would disagree that caesarean section is now the route of choice for the term breech.8-1

When we have audited appropriateness of caesarean section alongside timeliness we have over 95% peer agreement. As the tests we use in labour to try to identify fetal compromise are neither specific nor sensitive when they identify a case at risk, a decision needs to be made about delivery. When this decision is made then clinicians have always tried to deliver promptly. Standards have been introduced without clear evidence, and our paper considered the issue. Increasing concern about litigation has placed an emphasis on fetal outcome over maternal safety and the need to provide information to the parents.

Firstly, the lack of adverse outcome reflects, we hope at least in part, the fact that our unit makes decisions at a time before clinically significant fetal compromise occurs. We make no apology for that.

Secondly, it reflects a type two error, as much greater numbers of cases would need to be looked at before any difference was clear. Rates of cerebral palsy attributable to labour are very low and a huge series would be needed to see a difference. The small risk of adverse fetal outcome needs to be considered alongside maternal safety. When either rare outcome occurs criticism of the professionals follows. The debate about this balance needs to take place.

Where does this leave us? Although we should consider ways of speeding up, as have been described, we are more concerned with avoiding prolonged delays and perhaps finding ways of classifying cases more accurately, as also has been suggested. But with increasing litigation and also clinical governance requiring adherence to standards the presence of a national standard is hard to ignore. The concept of individual units creating their own standards may be attractive but is unlikely to be acceptable if there is any major difference from a national standard. It may well be that the sentinel audit of caesarean section can give some national data about what is achieved and could even set targets for what may be reasonable improvements.

References

  • 8-1.Hannah ME, Hannah WJ, Hewson SA, Hodnett ED, Saigal S, et al. Planned Caesarean section versus planned vaginal birth for breech presentation at term: a randomised multicentre trial. Lancet. 2000;356:1375–1383. doi: 10.1016/s0140-6736(00)02840-3. [DOI] [PubMed] [Google Scholar]
BMJ. 2001 Oct 20;323(7318):930.

Some words on conflict of interest

Richard Smith 1

Editor—Medical journals—and doctors generally—are in a transition phase in relation to conflicts of interest (or competing interests, as the BMJ calls them). We are moving from a phase where it has been normal not to declare them to one where it will be strange not to do so. In this transition phase we are confused about what to declare. My line is “If in doubt, declare,” which is why I even declared the death of my pet rabbit in an editorial on animal research.9-1 There are rarely problems when competing interests are declared but often anxieties when they seem to have been hidden.

Expert witnesses are paid—and so do have a financial interest when they are publishing on topics on which they give expert opinions. They should thus declare that they have been expert witnesses. The fact that this applies to many doctors makes it more, not less, important that such competing interests are declared.

References

Articles from BMJ : British Medical Journal are provided here courtesy of BMJ Publishing Group

RESOURCES