This document has been prepared by the Joint Working group of the Society of Rural Physicians of Canada (SRPC), The Maternity Care Committee of the College of Family Physicians of Canada(CFPC), and the Society of Obstetrics and Gynecologists of Canada (SOGC).

Stuart Iglesias, Stefan Grzybowski, Michael C. Klein, Guy Paul Gagne, Andre Lalonde

December 9, 1997

The Society of Rural Physicians of Canada (SRPC), The College of Family Physicians of Canada (CFPC) Committee on Maternity Care, and the Society of Obstetrics and Gynecologists of Canada (SOGC) share a commitment to providing the best maternity care possible for Canadian women. Representatives of these three organizations have formed a joint working group to develop policies and guidelines to support rural maternity care. The working group recognizes that input from rural women, nurses, midwives and physicians will be essential to the ultimate success of the implementation of these guidelines.

Objectives

Every woman in Canada who resides in a rural community should be able to obtain quality maternity care as close to home as possible. Whenever feasible she should give birth in her own community within the supportive circle of her family and friends. Respect for these women requires that public policy and clinical care guidelines support the provision of quality maternity care programs in rural Canada.

1. The Objectives

	To recognize and publicize that women should be able to give birth safely in rural Canada.
	To confirm that maternity care in rural communities is effective, appropriate, and safe with particular attention to those communities without local cesarean section capability.
	To encourage rural maternity programs to adopt evidence based best practice standards.
	To support rural physicians, nurses, and midwives to acquire the basic training, continuing professional education, and special skills required for rural maternity practice.
	To establish an evidence-based framework such that local risk management policies for rural obstetrics formulated by the licensing bodies, referral hospitals, and academic departments be consistent across the country.
	To assist women, communities, and local professional staff to gain a greater sense of ownership in the local maternity service.
	To promote the development of high quality rural perinatal databases.
	To provide a framework and criteria for audit and peer review

The Definition of Rural

The definition of rural is multi-factorial and necessarily somewhat arbitrary. It depends upon the size of the population, the size of the medical staff, the number of specialist medical staff on site, the health care facilities and level of technology available, and remoteness.

"In general terms, rural practice can be defined as practice in non-urban areas where most medical care is provided by a small number of general practitioners/family doctors with limited or distant access to specialist resources and high technology health care facilities."¹

This excludes all of the urban and suburban communities plus all the secondary care centres which enjoy reasonable access to tertiary care facilities. It is unclear whether some of the larger but very remote secondary care centres should be considered rural.

A practical definition in current Canadian application defines "rural remote" to be communities about 80-400 km and "rural isolated" to be greater than 400 km or about 4 hours transport in good weather from a major regional hospital. In some of the agricultural zones of Canada, the population is widely dispersed and served by hospitals that are rural in nature but within 80 km of small urban centres. These small hospitals function relatively independently to provide safe and adequate maternity care. Facilities that meet these characteristics can be defined as "rural close".²

The recent publication of Leduc’s General Practice Rurality Index (GPRI) provides an improved tool for the assessment of a community’s rural nature than one based on distance alone.³ The index assigns point scores to remoteness from closest advanced referral centre, remoteness from closest basic referral centre, catchment population size, number of general practitioners and specialists and presence of an acute care hospital.

Maternity care in rural Canada will always be provided with various levels of intensity. Personal attitudes, staffing and resource issues, communication and transport obstacles, and levels of training will influence more cautious risk management strategies for individual hospitals. Some patients will choose, when fully informed of the risks and benefits, to travel to a larger centre to give birth. All of these decisions should be fully supported within this position paper.

However, there are other rural maternity programs where nurses, midwives, and physicians with excellent training and involved with continuing professional education are committed to a much greater intensity of obstetrical care. They have the full support of their patients and their communities. They might wish to provide oxytocin augmentation of labor and induction of labor by various methods and/or to provide a full range of obstetrical analgesia options, and/or to acquire special skills training. This position paper should provide the framework and mechanism to ensure that conditions for safety, appropriateness, and accountability are met within a risk management strategy that belongs to the women, their communities, and their local professional staff.

This position paper should provide a platform sufficiently large for the variety of existing rural maternity programs while encouraging and validating the quality of rural maternity care.

The Evidence For the Safety of Rural Maternity Care

A Medline search of the literature (1980 to 1997) was done using key words rural and obstetrics and cross searching with the following MESH headings, maternity, perinatal, asphyxia and cesarean section. Articles which were included were well constructed retrospective or cohort studies and relevant to the questions postulated below. There were no relevant randomized controlled trials. This search was supplemented by consultations with Tom Nesbitt and Roger Rosenblatt both of whom have conducted research in this area.

The setting for the conduct of the studies that we considered valid for the Canadian setting included the requirement that the data be derived from organized perinatal systems facilitating consultation and transfer between the rural settings and the perinatal centres to which they referred.

A. Is a limited local rural obstetrical service better than no local obstetrical service?
The only studies available that address this question are from the United States. A study from Washington State showed that women who live in communities with poor local access (what Tom Nesbitt called high outflow communities) are more likely to bear infants who are premature, have prolonged hospitalizations with higher costs or both.⁴ Larimore and Davis showed a significant quantifiable increase in infant mortality due to lack of maternity caregivers in rural Florida.⁵

Lack of local maternity services leads to potential isolation and compromise of women who do not have the financial means to travel to other communities to seek their routine antenatal and intrapartum care. No one will be trained to handle emergencies. Absence of intrapartum care will lead to reduced antenatal care resources and expertise.

B. Is a small rural maternity service safer with Cesarean Section capability than without?
This comparison of similar rural services with and without Cesarean Section capability has not been done. It would be essential for communities which presently have Cesarean Section capability to maintain this service until this evidence is available. It would also be appropriate for communities which are presently successfully providing maternity services without local Cesarean section capability to continue to provide maternity service. There are 125 hospitals in Canada that provide maternity service without full time Cesarean section capability on site.⁶

C. Are the outcomes of rural hospitals as good as urban maternity services?
The research envelope is thin. A limited number of studies have compared the outcomes of care in different size hospitals, the smallest of which do not have cesarean section capability. Black and Fyfe looked at pregnancies and deliveries in Northern Ontario.⁷ They attributed all pregnancy outcomes to the place of residence of the mother and the hospital within which catchment area she lived. They showed that populations served by small Level I hospitals had perinatal loss rates similar to the rates in those served by larger secondary or tertiary care facilities, even when all adverse outcomes were attributed back to local hospitals.

In Nova Scotia, Peddle et al showed that small community hospitals with less than 100 deliveries per year had the lowest perinatal morbidity and mortality rates in the province.⁸ These small hospitals did 23% of the deliveries in Nova Scotia.

A population based study from remote British Columbia demonstrated no adverse perinatal outcomes attributable to lack of local cesarean section capability in 5 years of maternity care.⁹

International data from Australia and New Zealand show that women delivering in rural hospitals manned exclusively by GP's and midwives with an without immediate cesarean section capability have fewer premature births, and fewer hypoxic infants and lower birth weight specific mortality rates than the Level II and III centres to which they refer.^10,11

In summary, what evidence there is available suggests that rural hospitals, with limited services and, in many case, without local cesarean section capability, do offer acceptably safe maternity care. Furthermore, and perhaps more importantly, populations served by rural hospitals which do not offer maternity care seem to have worse perinatal outcomes.

Clearly the limited data supports the maintenance of rural maternity care services for women in Canada.

D. Research Agenda
The studies described above all suffer from relatively small numbers with the exception of data from the entire birth experience of New Zealand.⁹ But the New Zealand results are weakened by the fact that the results were based on the hospital in which delivery took place rather than booking based data or data attributed to the hospital by the residence of the mother as Black did in Northern Ontario.

There is an urgent need for Canadian research on the maternal and neonatal outcomes of births in these small hospitals. We need studies comparing the safety of and care for populations served by similar rural hospitals with and without local cesarean section capability and more information about the outcomes for rural communities that have lost their local maternity service.

E. Audit of Outcomes
We need to establish large coordinated databases, at provincial and national level with the opportunity to compare practices and outcomes. Ideally results should be attributed to maternity services by maternal residence within the catchment area of each hospital as Black did in his study in Northern Ontario rather than by place of birth. This will measure the outcomes for the system of care rather than the selected population that delivers locally. At the same time the population based ratio of local delivery and intrapartum transfer will provide important quality of care indicators.

Hospital based statistics will also be important in order to assess hospital and practitioner practice and allow for important feedback which may serve as the driver for quality improvement initiatives at a local level. The Northern and Central Alberta Education and Audit Program (NCAEAP) is an example of a hospital- based audit system that is already up and running. It was established in 1991, includes most level I, II, and III hospitals in the province and provides comparative hospital statistics and confidential physician statistics to participants. Similar databases exist in B.C. and Nova Scotia. What is required is a national collaboration with standardized data collection. This should be available in due course through the Canadian Perinatal Surveillance System.

Defining Principles

A. Regionalization
Women at higher risk for adverse maternal and/or perinatal outcomes should deliver in centres with the facilities to manage the complications of labour and delivery. The regionalization of maternity care, as a general principle, similar to the regionalization of all medical and surgical care in Canada, appears to serve rural Canada very well. Regionalization is widely accepted by patients and health professionals, especially in rural Canada, and perceived to be an excellent organizing principle of maternity care.

B. Risk Management
Risk can never be completely avoided. As long as communities include women of childbearing age, obstetrical risk will occur. Although some risk can be anticipated, a substantial portion of adverse outcome is unexpected.¹² For example, using Manitoba data, in a general hospital, 10 percent of infants with a low risk score prior to delivery require a resuscitation at delivery.¹³

Some patients at increased risk can be identified antepartum and transferred although transfer itself is associated with risk. Intrapartum events require frequent assessment of risk, continuous disclosure and informed consent.

The responsibility for the management of complications and risks in maternity care rests with the local care unit. Local professional staff, hospital boards, and the local community need to develop and maintain a comprehensive system to deal with complications that may develop. Practice and procedures should be evidence and guideline-based.

In addition a formal risk management process should be in place. Risk management is a continuous process.¹⁴ It starts with identification and analysis of risk, proceeds to the establishment of actions to manage risk, and then evaluates the results which leads to further identification and analysis in a cyclical fashion of continuous quality improvement.

C. Guidelines
The SOGC has developed a number of guidelines and policy statements which provide a basic strategy for managing common maternity care issues. (Appendix 1) The SRPC and the CFPC Committee on Maternity Care also endorse them as appropriate and applicable for rural practice except for minor concerns related to the Fetal Health Surveillance guidelines. In principle, guidelines and policy statements should be applied uniformly to the care of all low risk maternity care in Canada. Future maternity care guidelines issued by any of the three organizations should be subject to an expeditious and effective process of joint consultation and approval.

The SOGC has stated that "Clinical Practice Guidelines do not define the standard of care nor are they intended to dictate an exclusive course of treatment to be followed."¹⁵ And the organization has further asserted that:

"Variations of practice, taking into account the needs of individuals, patient resources, and the limitations unique to the institutions or type of practice may be appropriate. A guideline can, and will, be modified according to local conditions. If so, it should be documented in individual departments and/or hospitals."¹²

This tolerance of flexibility in local applications of SOGC guidelines should not be construed as an acceptance of a lower standard of care in rural Canada. The standard of care for a low risk maternity patient should be the same in the smallest Level I hospital as it is in tertiary care centres. This requires that:

1. Rural Canada sustains a commitment to the human and financial resources to meet national maternity care standards and, most importantly,
2. It is critical to appreciate that the loss of local maternity services for rural communities may well be associated with worse perinatal outcomes for the population served, even when patients travel to maternity centres with an excellent standard of care.

D. Evidence Based Medicine
We should all strive to practice according to the best evidence based information available. Critical appraisal of relevant literature should guide policy and practice guideline development. A summary and meta analysis of randomized controlled trials is presented in the Cochrane Library.¹⁶ Where the information does not yet exist we should encourage appropriate research to be done.

E. Informed Choice
Women and their maternity care providers should be partners in choice. Informed consent requires full disclosure to prospective mothers of the advantages and limitations of the local maternity care service consistent with guidelines and audit of local outcomes. This should include issues of anticipated obstetrical risk, as well as time-frames and transport modalities. This should include the conditions for, and risks of, transport to a secondary or tertiary care centre. Each woman should have the opportunity to choose where she will seek her maternity care. This process of disclosure and consent must continue intrapartum as risk is periodically re-evaluated during the progress of labour.

F. Characteristics of a Safe Rural Maternity Service:
A rural risk management strategy should include as a minimum:

	A qualified, competent, and committed professional staff.
	Sufficient financial and technical resources to meet National standards of care.
	Detailed written transport protocols.
	Open lines of communication and collaboration with regional referral centers.
	Continuing audit and quality improvement programs.

The Recommendations

1. Women in Canada who reside in a rural community should receive high quality maternity care.
2. Rural hospitals should, within a regionalized risk management system, offer maternity care to a low risk population. While anesthetic and surgical services are desirable, the available evidence suggests that good outcomes can be sustained within an integrated risk management system without local access to operative delivery.
3. There should be a single standard of care for the provision of maternity care services to low risk women.
4. Rural maternity care services should develop a formally documented risk management strategy that includes issues of management of obstetrical risk, regionalized care, local resources and transfer options.
5. Maternity care providers should be skilled in a recognized emergency skills and risk management course such as ALSO / ALARM and NRP.* These should be consolidated in a single program and delivered to and in rural Canada.
6. The SOGC, CFPC, and SRPC should promote the SOGC Guidelines as appropriate for general application to rural Canada. Where appropriate these will be amended to recognize the realities of rural practice in order to reflect a single standard of perinatal care for Canada. The pertinent obstetrical guidelines and policies are listed in Appendix 1.
7. Regional perinatal databases should be population based. Linkages between these databases should be created to advance the rural maternity care research agenda and allow for the effective comparison of outcomes.
8. Regional perinatal databases should provide to hospitals and maternity care providers the information required for audit and Continuous Quality Improvement (CQI).
9. Future Maternity Care Guidelines issued by any of our three organizations should be subject to an expeditious and effective process of joint consultation and approval.

* Advanced Life Support in Obstetrics, Advanced Labour and Risk Management, Neonatal Resusitation Program

Appendix 1: SOGC Guidelines for Obstetrical Care

Fetal Health Surveillance in Labour - Parts 1 to 3 and Conclusion (October 1995 to January 1996) is generally appropriate for rural practice. There needs to be further discussion on two points. Firstly the recommendation to submit cord gasses is primarily a risk management strategy and not based on evidence demonstrating an improvement in perinatal outcomes. Secondly, if evidence does become available to support cord gas sampling our insistence on a single standard of care precludes special exemption for rural Canada.

References

1. ROURKE, J., "In search of a definition of ‘rural’", Can J Rural Med, vol.2(3) p.113-115.
2. Canadian Association of Emergency Physicians, Recommendations for the management of rural remote and rural isolated emergency health care facilities in Canada. Ottawa, 1997: vol.6.
3. LEDUC, E., "Defining rurality: a general practice rurality index for Canada", Can J Rural Med. 1997; ,2(3), p.125-131.
4. T.S. NESBITT, F.A.CONNELL, L.G.HART, R.A.ROSENBLATT, "Access to obstetric care in rural areas: effects on birth outcomes". American Journal of Public Health, 1990; vol 80(7) p.814-818.
5. W.L.LARIMORE, A.DAVIS, "Relationship of infant mortality to availability of care in rural Florida". Journal of the American Board of Family Practice, 1995; 8: 392-399.
6. C.LEVITT, L.HANVEY, D.AVARD, G.CHANCE, J.KACZOROWSKI, Survey of Routine Maternity Care and Practices in Canadian Hospitals. Ottawa: Health Canada and Canadian Institute of Child Health, 1995.
7. Black,D.P., FYFE, I.M., "The safety of obstetric services in small communities in Northern Ontario", Can Med Assoc J , 1984; 130: 571-576.
8. PEDDLE, L.J., BROWN, H., BUCKLEY,J., DIXON, W. et al "Voluntary regionalization and associated trends in perinatal care: the Nova Scotia reproductive care program". American J Obstet Gynecol, 1983; 145(2): 170-176.
9. GRZYBOWSKI,S.C., CADESKI, A.S, HOGG, W.E., "Rural Obstetrics: a 5 year prospective study of the outcome of all pregnancies in a remote northern community". Can Med Assoc J 1991; 144(8): 987-94 (10 year results available from authors)
10. WOOLLARD, LA., HAY, R.B. "Rural Obstetrics in NSW. Aust NZ Obstet Gynaecol 1993; 33(3): 240-28.
11. ROSENBLATT, R.A., REINKEN,J., SHOWMACK, P. "Is obstetrics safe in small hospitals", Lancet, August 24, 1985; 429-431.
12. SOGC Guideline on Risk Management
13. HALL, P.F., HARRISON, M., BROWN, R., "Risks of Risk Scoring " (Abstract), Int. J Obst Gynecol 1994; 46:100
14. SOGC Guideline on Cerebral Palsy
15. SOGC annotation on all guidelines
16. The Cochrane Library (Database on disc and CD ROM). The Cochrane Collaboration Oxford: update software; 1996, updated quarterly.