Why does obstetrics merit such attention in an Occasional. Paper about rural practice? Obstetrics is an integral and unavoidable part of many rural practices and ...
23
CHAPTER 7
Obstetrics in rural practice: problem or solution? A Gordon Baird MRCOG, MRCGP and John CM Gillies FRCPEd, MRCGP
DIVERSITY in approach and style has always been a strength of UK general practice. This diversity is extreme in the provision of maternity services, where some practices are only minimally involved whilst others provide complete maternity care. For many doctors in the latter group this is not optional but an integral part of the rural life they have chosen. The provision of maternity services in rural areas in the UK has received little attention.
The rural perspective Why does obstetrics merit such attention in an Occasional Paper about rural practice? Obstetrics is an integral and unavoidable part of many rural practices and demands specific facilities. Political and managerial changes, such as proposed by the Cumberlege Report (1993), and the restructuring of hospital services threaten both facilities and the maintenance of expertise, skill and education. Some of the evidence presented in this paper and much of what was proposed in the Cumberlege Report might suggest that the need for doctors to be involved is unnecessary and perhaps even harmful. Why should rural general practitioners be involved in obstetrics? First, the general practitioner is the person most identified by women with antenatal care in pregnancy (Taylor, 1986; Tucker et al., 1994). Many women express a wish for the general practitioner to be involved. Continuity of care and the presence of a trusted professional is highly valued by patients. Secondly, immediate medical cover for unexpected emergencies will centre around the general practitioner, and midwives are required to summon medical assistance in the event of abnormalities or emergencies. Rural women with uncomplicated pregnancy should not and would not accept a journey of many miles to a regional centre for all antenatal and intrapartum care. The options of induction, transfer when labour becomes established with the risk of delivery occurring in transit, awaiting the onset of labour in a remote district general hospital, or home delivery are unacceptable for normal pregnancy. A community unit managed by general practitioners and midwives together is a sensible alternative to any of the above. How big is the problem? As Rousseau discusses in Chapter 1, problems of definition of rurality muddle any discussion. In a recent policy document the Scottish Office (1993) reviewed deliveries performed in Scottish general practitioner hospitals. There were 170 beds in 25 hospitals in which 1750 women were delivered annually. The general practitioner hospitals were
between 14 and 120 miles from district general hospitals. Five hospitals required ferry, lifeboat or air transfer of patients in an emergency. In our own unit we deliver half of our maternity patients. Assuming that other Scottish rural practitioners do the same (Marsh and Channing, 1989), the total number of babies delivered by Scottish rural general practitioners is equivalent to the workload of a large regional maternity unit. A telephone enquiry to all Scottish general practitioner hospitals revealed that 229 general practitioners are directly involved in intrapartum care, with a further 70 referring to them but too far away to be directly involved. These 300 general practitioners are serving a population of 600 000. The Scottish policy document highlighted low bed occupancy as a problem in general practitioner maternity units, without accepting its inevitability in small communities, and ignoring the resource implications of women boarding in a district general hospital awaiting labour, and ambulance, midwife and general practitioner time associated with transfer of women to regional centres (Scottish Office, 1993). Transport itself carries a risk, not only for mother and baby, but also for their attendants. It is obvious that frequent transfers over long distances in remote rural areas with limited ambulance resources and poor roads increase risk enormously. Helicopter and lifeboat transfer add yet another dimension to this complex problem. The death of a midwife, an ambulance paramedic, a term pregnancy loss and a near miss maternal death during a recent transfer changed a low level of individual risk to a highland community disaster. It also emphasized that transfer of patients puts professionals at risk as well as their patients. These risks are obviously increased in more isolated areas and in bad weather.
Risk assessment There is overwhelming evidence that general practitioner deliveries are safe in women categorized as low risk. Analysis in this country (Black, 1982; Tew, 1985) and in New Zealand (Rosenblatt et al., 1985) concluded that small units are safer than large units even accounting for selection bias. The Association for Community Based Maternity Care has provided an excellent review of the evidence (Jewell et al., 1992). Concerns about risk in intrapartum care centre around the detection of fetal distress, the occurrence of cerebral palsy, the immediate requirement to perform operative vaginal or abdominal delivery and the difficulties of performing adult or neonatal resuscitation. Worries are magnified by rural isolation and are exacerbated by our hospital training, which emphasizes the importance of identifying and preventing
24
problems which are uncommon even in a high risk environment. Although electronic fetal monitoring has proved to be ineffective in reducing mortality in normal pregnancy (McDonald et al., 1985; Levenko et al., 1986), many women are still subjected to caesarean section (Prentice and Lind, 1987; Boylan, 1987; Albers and Savitz, 1991) and their infants to bleeding and sepsis from the use of scalp electrodes rather than equally effective intermittent auscultation (Leatherman et al., 1992). Cerebral palsy is a rare problem, affecting between 1.5 and 2 per 1000 live bom infants (Weatherall et al., 1985) and only about 8% of cases can be attributed to preventable intrapartum causes (Nelson and Ellenberg, 1986; Blair and Stanley, 1988; Nelson, 1988). The overall risk is less than 2 in 10 000. Thus a general practitioner would practise for over a thousand years before encountering such an event. The perception of doctors that huge lawsuits for missed intrapartum fetal distress leading to cerebral palsy are commonplace has no basis in fact. Birth should be thought of as a normal process requiring keen observation, though not necessarily supervision or intervention. The incidence of cerebral palsy has increased with the increasing rate of caesarean section. The most potent factor involved is prematurity and the mischievous might infer that intervention is a precipitating rather than alleviating factor. Working and living in a small community hampers objective analysis of risk; the results of emergency procedures of nightmarish proportions may literally follow the poor doctor down the street, even when the outcome is good. The only Erb's palsy I have seen was pointed out to me in the street by my father, who had delivered the child as a second twin in a difficult domiciliary delivery 25 years before. These constant reminders are a peril for doctors in small communities. Meeting medical disasters, real or perceived, is a risk of any public appearance, and serves as a reminder to doctors of their commitment to undertaking all sorts of tasks, some of which are theirs only by default. To ignore these personal aspects of rural practice when discussing risk assessment would be quite wrong, even if it is difficult to produce scientific evidence.
What problems are likely? In our practice, the caesarean section rate was only 4% of 857 women in active labour (cervical dilatation more than 4 cm). Many of these caesareans were in high risk pregnancies, and others in prolonged first stage unsuitable for general practitioner care. The requirement for urgent caesarean in labour has been overemphasized. In the last few years the forceps rate at our unit has fallen from 4% to around 1%. This may be due to more flexibility about the duration of the second stage, the introduction of oxytocin for augmentation of labour, education about the use of the partogram (Philpott and Castle, 1972) or perhaps decreasing operator confidence. Despite a transfer joumey of 75 miles, the intervention rate at the general practitioner unit has been safely and effectively reduced. The risk of low Apgar score (4 or less) occurring after delivering a low risk pregnancy is ever present, although the scale and severity of the problem is not comparable with specialized units, where highly compromised premature or growth retarded fetuses are delivered. In low risk pregnancies the incidence of low Apgar scores is reported as
1.3% (Permeyal et al., 1987) and this is similar to our own experience. Third stage anomalies such as post partum haemorrhage and/or retained placenta may occur unexpectedly in around 3%-4% of low risk pregnancies (Permeyel et al., 1987) as opposed to 4%-6% overall (Hall et al., 1985). Sudden and unexpected post or intrapartum collapse from eclampsia, post partum haemorrhage or amniotic fluid embolus is so catastrophic that in many cases death is unavoidable. The most recent confidential enquiry into matemal deaths suggests the use of protocols (DoH et al., 1994). Deaths may occur in any pregnancy and so community emergency services (this means general practitioners in rural areas) should be prepared to deal with them in any case. Who should lead obstetric care? The Cumberlege Report (1993) proposals for midwifery-led care threaten existing models of care which have carefully and thoughtfully evolved over years in response to local needs. General practitioner hospitals rely on a close team approach. Separating general practitioner and midwife in proposed midwife-led units would fragment teams, and jeopardize care of the unexpected emergencies that occur in approximately 5% of otherwise uncomplicated pregnancies (Permeyel et al., 1987). The report has been criticized for failing to take representation from the Royal College of Obstetricians (Dunlop, 1993; RCOG, 1993) and for not having RCGP representation. It suggested that, despite the difficulty of identifying low risk cases, midwives could manage uncomplicated low risk cases in the community (Reynolds et al., 1988; Hall, 1990). Twenty-five per cent of midwife-only schemes have been disbanded between 1990 and 1991, often because of discontent among the midwives themselves (Institute of Manpower Studies, 1993). In the UK, midwives-only schemes are an unproven method of providing care. Prediction of increased risk is easy but low risk patients are not risk free. Poor understanding of what 'low risk' pregnancy means ridicules the concept of informed choice in the report. The current model of consultant and general practitioner led care is flexible and has proved successful in providing community-based care (Reid et al., 1983; Mckee, 1984). The proposed changes threaten choice for women in many rural areas. The way ahead? Medical training emphasizes that there is no such thing as a risk-free pregnancy. While this is true, the question of appropriate response to risk is important. For many women, the resources of a district general hospital are not required and any disadvantages of a 'low tech' community unit are offset by their advantages. When problems occur in labour, for the most part they require relatively simple interventions. Third stage anomalies (post partum haemorrhage and retained placenta) in fit young women usually require simple resuscitative techniques in conjunction with manual removal of the placenta. Low Apgar scores in otherwise fit term babies does not require great technical expertise. These problems are usually easily dealt with by anyone familiar with acute medicine in general practice and resuscitation techniques are easily maintained by, for example, using a manikin to train for
25
neonatal intubation. Of course there must be 24-hour availability at short notice, but this is true of many other aspects of rural general practice. Training for obstetrics in rural areas has suffered from the discontinuation of travel and subsistence reimbursement for attendance at courses. It is impossible to acquire practical techniques such as manual removal of placenta, rupture of membranes, assisted deliveries and intubation by distance learning. Attendance at courses involves absence from the practice and the cost of locum cover. The Royal Colleges have a role to play in education by encouraging an appreciation of normal pregnancy and labour. The strengthening of the team approach and the inclusion of all professional groups has already been acknowledged (RCOG et al., 1992; RCOG, 1993). The specific needs for training in practical procedures and risk assessment for doctors undertaking intrapartum care must be addressed, as should the factors which have reduced incentives to maintain skills. Clear guidance on protocols for obstetric emergencies is essential. Clinical research and audit should be encouraged. The concept of choice envisaged in the Cumberlege Report (1993) is irrelevant and distracting to the provision of care in rural areas. For many women and their general practitioners their rural environment limits choice. Therefore an integrated service with a choice of key worker as suggested by the Royal Colleges (RCOG et al., 1992) should be pursued. Rural practices are major providers of training in intrapartum' care in low-risk pregnancies. Rural practice offers patients a combination of continuity of care, choice and safety. This combination within a single integrated system could usefully be copied elsewhere. References Albers LL and Savitz DA (1991) Hospital setting for birth and use of medical procedures in low-risk women. Journal of NurseMidwifery 36, 327-33. Black N (1982) Do general practitioner deliveries constitute a perinatal mortality risk? British Medical Journal 284, 488-90. Blair E and Stanley FJ (1988) Intrapartum asphyxia: a rare cause of cerebral palsy Journal of Pediatrics 112, 515-9. Boylan P (1987) Intrapartum fetal monitoring. Review. Baillietre's Clinical Obstetrics and Gynaecology 1, 73-95. Cumberlege Report (1993) Changing Childbirth: The Report of the Expert Maternity Group. London, HMSO. Department of Health et al. (1994) Report on Confidential Inquiries into Maternal Deaths in the United Kingdom 1988-1990. London, HMSO. Dunlop W (1993) Changing childbirth. Commentary II. British Journal of Obstetrics and Gynaecology 100, 1072-4. Hall M, Halliwell R and Carr-Hill R (1985) Concomitant and repeated happenings of complications of the 3rd stage of labour. British Journal of Obstetrics and Gynaecology 92, 732-8. Hall MH (1990) Identification of high risk and low risk. Bailliere's Clinical Obstetrics and Gynaecology 4, 65-76. Institute of Manpower Studies Report (1993) Mapping Team Midwifery. A Report to the Department of Health. IMS Report No 252, 107-16.
Jewell D, Young G and Zander L (1992) The Case for Community Based Maternity Care. Bristol, Association for CommunityBased Maternity Care. Leatherman J, Parchman ML and Lawler FH (1992) Infection of fetal scalp electrode monitoring sites. American Family Physician 45, 579-82. Levenko KJ, Cunningham FG, Nelson S et al. (1986) A prospective trial of selective and universal electronic fetal monitoring in 34 995 pregnancies. New England Journal of Medicine 315, 615-9. McDonald D, Grant A, Sheridan Pereira M et al. (1985) The Dublin trial of intrapartum fetal monitoring. American Journal of Obstetrics and Gynecology 152, 524-39. Mckee H (1984) Community antenatal care: The Sighthill project. In Zander L and Chamberlain G (Eds) Pregnancy Care for the 1980s. London, Royal Society of Medicine and Macmillan Press. Marsh GN and Channing DM (1989) An audit of 26 years' general practitioner maternity care. British Medical Journal 298, 107780. Nelson KB (1988) What proportion of cerebral palsy is related to birth asphyxia? Journal of Pediatrics 112, 572-4. Nelson KB and Ellenberg JH (1986) Antecedents of cerebral palsy; Multivariate analysis of risk. New England Journal of Medicine 315, 81-86. Permeyel JM, Peperell RJ and Kloss M (1987) Unexpected problems in patients selected for birthing unit deliveries. Australian and New Zealand Journal of Obstetrics and Gynaecology 27, 21-23. Philpott RH and Castle WM (1972) Cervicographs in the management of labour in primigravidae. Journal of Obstetrics and Gynaecology of the British Commonwealth 79, 592-8. Prentice A and Lind T (1987) Fetal heart monitoring during labour - too frequent intervention, too little benefit? Lancet 2, 1375-7. Reid ME, Gutteridge S and Mcilwaine GM (1983) A Comparison of the Delivery of Antenatal Care between a Hospital and Peripheral Clinic. Social Paediatric and Obstetric Research Unit, University of Glasgow. Reynolds JL, Yudkin PL and Bull MJV (1988) General practitioner obstetrics: does risk prediction work? Journal of the Royal College of General Practitioners 38, 307-10. Rosenblatt RA, Reinken J and Shoemack P (1985) Is obstetrics safe in small hospitals? -Lancet 1, 429-33. Royal College of Obstetricians and Gynaecologists (1993) Response to the Report of the Expert Maternity Group: Changing Childbirth. London, RCOG. Royal College of Obstericians and Gynaecologists, Royal College of Medicine and Royal College of General Practitioners (1992) Maternity Care in the New NHS. A Joint Approach. London, RCOG, RCM and RCGP. Scottish Office (1993) Provision of Maternity Services in Scotland - A Policy Review. London, HMSO. Taylor A (1986) Maternity services: the consumer's view. Journal of the Royal College of General Practitioners 36, 157-60. Tew M (1985) Place of birth and perinatal mortality. Journal of the Royal College of General Practitioners 35, 390-4. Tucker J, Florey CduV, Howie P et al. (1994) Is antenatal care apportioned according to obstetric risk? The Scottish Antenatal Care Study. Journal of Public Health Medicine 16, 60-70. Weatherall DJ, Ledingham JGG and Warrell DA (Eds) (1985) Oxford Textbook of Medicine. Oxford, Oxford Medical Press. Vol II 21, 85-87.
