UBC Faculty Research and Publications

Centralized or decentralized perinatal surgical care for rural women: a realist review of the evidence… Kornelsen, Jude; McCartney, Kevin; Williams, Kim Aug 13, 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-12913_2016_Article_1629.pdf [ 637.52kB ]
JSON: 52383-1.0308662.json
JSON-LD: 52383-1.0308662-ld.json
RDF/XML (Pretty): 52383-1.0308662-rdf.xml
RDF/JSON: 52383-1.0308662-rdf.json
Turtle: 52383-1.0308662-turtle.txt
N-Triples: 52383-1.0308662-rdf-ntriples.txt
Original Record: 52383-1.0308662-source.json
Full Text

Full Text

RESEARCH ARTICLE Open AccessCentralized or decentralized perinatalsurgical care for rural women: a realistreview of the evidence on safetyJude Kornelsen1*, Kevin McCartney1 and Kim Williams2AbstractBackground: The precipitous closure of rural maternity services in British Columbia (BC), Canada, and internationallyhas demanded a reevaluation of how to meet the perinatal surgical needs of rural women in accordance with theTriple Aim objectives of safety, cost-effectiveness, and satisfaction of all key stakeholders. There is emerginginternational evidence that General Practitioners with Enhanced Surgical Skills (GPESS) are a well-positioned healthservice solution due to their generalist nature in low-volume settings. A realist review was undertaken to evaluateinternational evidence on efficacious models of perinatal surgical care. This article presents findings of thesafety of such practice, one discrete part of the full realist review.Methods: This paper was derived from a larger review, which used a realist review methodology to guide theapproach, and adhered to the RAMESES quality standard for realist reviews. Seven academic databases weresearched in December 2013, using year (1990) and language (English) limiters in keeping with a rapid reviewapproach. Mining of bibliographies in addition to consultation with international experts led to further inclusion ofacademic and grey literature up to March 2014.Results: Two hundred fifty-four articles were originally identified; 119 articles were removed from consideration forlack of fit, resulting in the review of 191 articles from the peer reviewed and grey literature. Of these, 53 pertainedto safety and are considered herein. Evidence on the safety of GPESS was consistent in the literature cited. Clinical,case study, and qualitative evidence demonstrates that perinatal surgical care is equally safe when provided byGPESS and specialist physicians.Conclusion: Findings allow health planners to confidently build perinatal surgical services around the contributionof GPs with enhanced surgical skills and focus on educational, regulatory, and continuing professional developmentmechanisms to ensure their sustainability. Volume-to-outcomes associations are variable and inconclusive withregards to safety, suggesting the need for more evidence. These findings, and the attendant health servicesplanning directions, are reassuring as they suggest the viability of local models of care where feasible.Keywords: General practitioners with enhanced surgical skills, GP proceduralists, GP surgeons, Rural maternity care,Rural surgical care, Realist review* Correspondence: jude.kornelsen@familymed.ubc.ca1Department of Family Practice, University of British Columbia, DavidStrangway Building, 3rd Floor, 5950 University Blvd., Vancouver V6T 1Z3, BC,CanadaFull list of author information is available at the end of the article© 2016 Kornelsen et al. Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.Kornelsen et al. BMC Health Services Research  (2016) 16:381 DOI 10.1186/s12913-016-1629-6BackgroundAcross Canada and internationally, we have seen theprecipitous closure of many rural health services [1–5],leading in some instances to deteriorating populationhealth outcomes and reduced quality of care [5–8]. Cur-rently, professional trends in General Surgery and Ob-stetrics have led to a reduction in their contribution tosmaller rural services [9, 10]. This is occurring within aresearch and policy context that recognizes the benefitsof services “closer to home” [11, 12] balanced with theneed for fiscally responsible planning [4]. In small-volume centres, a generalist approach has been shownto be the most efficacious way of meeting the needs ofthe population [10, 13, 14]. In the case of cesarean sec-tion, these conditions have created a response fromGeneral Practitioners with Enhanced Surgical Skills(GPESS) training to meet the operative needs of thepopulation in many jurisdictions. The primary care focusof their work alongside their availability for limited pro-cedural work addresses the challenges of low surgicalvolume in conjunction with the primary care needs ofrural communities. Although this solution has been rec-ognized and integrated into rural health care planning injurisdictions such as Australia, the United States,Norway, Scotland, and in more ad-hoc ways in Canadaand other jurisdictions, a review of the international lit-erature on the safety and outcomes of GPESS has notbeen undertaken. In 2012, the BC Ministry of Healthheld province-wide consultations with key stakeholdersin order to establish a set of consensus-driven actionitems for a provincial primary maternity care agenda,known as the BC Primary Care Plan. These consulta-tions also resulted in a series of short-term “actionitems.” Health care decision makers recognized that anyreasoned debate about these issues demanded a rigorousreview of the international literature.In addition to the BC Primary Maternity Care Plan,perinatal planning in British Columbia (BC) has beenconceptually guided by a report authored by JusticePeter Seaton in response to the Royal Commission ofHealth Care and Costs, which recommended “[m]edicallynecessary services… be provided in, or as near to, thepatient’s place of residence as is consistent with qualityand cost-effective health care” (P. A-6). This recom-mendation was made based on two features. First, theSeaton report recognizes the challenges rural residentsface in accessing health care, including insufficient supplyof providers, inappropriate emergency services and thecost incurred by patients forced to travel for treatment[15]. These same challenges are faced by rural residents invarious international jurisdictions. Second, the Seatonreport expressed the belief that a decentralized health caresystem would better respond to many health needs withinrural and remote communities.The fundamental challenge to providing operativebackup for deliveries in rural communities internation-ally is lack of availability of surgical providers [16]. Thishas become the reality in rural British Columbia as well[17]. The solution pursued worldwide is to increase thesupply of rural generalist surgeons, including trainingmore General Practitioners with Enhanced SurgicalSkills and involving more General Surgeons in the deliv-ery of perinatal surgical services. The relatively smallprocedural volumes of these programs, however, are as-sociated with important issues regarding program sus-tainability – which deter specialist practice – includingthe challenge of maintaining competence for the profes-sional staff, lack of opportunity for intensive applicationof practitioners’ skills, restriction on the numbers ofskilled providers that can be supported by the local ser-vice demand (leading to vacation and on-call relief prob-lems), and programs associated with high unit costs.Despite this, research evidence has demonstrated theimportance of local cesarean section in sustaining ruralmaternity services [18, 19]. A BC study found local ac-cess to cesarean section increased the proportion of localdeliveries from less than 30 % (no local cesarean section)to greater than 75 % when operative deliveries were lo-cally available [20, 21]. Similarly, a study in Albertafound a local retention rate of 22.1 % for women incommunities without cesarean section compared to70.1 % in communities with local operative delivery [22].In the early 1990s, evidence began emerging whichsuggested that the profession of General Surgery wasaging and due to inevitable retirement would not be ableto sustain a strong rural presence without training newpractitioners [9, 14, 23–25]. However, attracting new re-cruits was difficult due to the perception of lack of inter-est in the specialty, leading to demanding call schedulesand the lack of sub-specialist support in rural environ-ments [10]. This is despite the recommendations of theBarer-Stoddart report [26], which suggested priority begiven to training generalist surgeons for practice in non-urban hospitals. The lack of General Surgeons in ruralareas is not unique to Canada but also characteristic ofrural Australia [27] and the United States [28–30].The reality in British Columbia is most rural areas arenot serviced by local specialist support: General Practi-tioners with Enhanced Surgical Skills are the primarysurgical service provider [14, 31, 32], making the GPESSmodel synonymous with “decentralized perinatal surgicalservices.” For populations of 5000–15,000, surgical ser-vices are provided locally by one or more GPESS,cesarean section often being the backbone to their pro-cedural skills repertoire. For populations of 15,000–25,000, there is usually a specialist surgeon, in some in-stances an obstetrician, supported by one or moreGPESS (a “mixed” model). In these larger communities,Kornelsen et al. BMC Health Services Research  (2016) 16:381 Page 2 of 13the GPESS provides call relief and often covers the opera-tive delivery program. For populations greater than25,000, there are usually groups of specialists withoutany GPESS [32].In 1995/96, the most recent published data, 1838 c-sections were performed by 200 rural GPs in Canada[33]. Rural intrapartum care was provided by 1704 ruralGPs, who attended 25,602 births, 8.4 % of total births inCanada that year [33]. Three-quarters of all GPs per-forming c-sections were doing so west of Ontario [33],and GPs with Enhanced Surgical Skills practiced at 60 ofthe 72 small rural hospitals (<51 beds, <15,000 personcatchment) providing surgical services in BC, Alberta,Yukon, and the Northwest Territories [34]. Forty-threeof those hospitals had GPs performing c-section proce-dures [34].Given this context, this review sought international lit-erature on models of care to meet the perinatal surgicalneeds of rural women in order to provide a broader con-text to rural health planning in British Columbia. Al-though GPESS is synonymous with decentralizedperinatal surgical care in BC, this review considered allmodels in jurisdictions with a comparable health servicescontext. Due to space limitations, this paper focuses ex-clusively on evidence of the safety and outcomes ofmodels reviewed.This realist review was commissioned by Perinatal Ser-vices of British Columbia (PSBC), a provincial policybody in BC, as part of a provincial strategic planningprocess to establish an evidence-informed primary ma-ternity care agenda. One of the action items resultingfrom the agenda was focused on resolving some of theinter-professional and regulatory tensions within themedical community regarding GPs with Enhanced Surgi-cal Skills and their role in sustaining perinatal surgicalservices for rural women. As GPESS were seen as under-scoring only one potential model of care, the reviewquestion was structured to be purposively open to evi-dence suggesting the effectiveness of a more centralizedresponse as well (i.e., moving rural women into regionalmaternity care units for labor and delivery). The final re-search question was:Can we meet the perinatal surgical needs of ruralwomen more effectively through an optimallycentralized or optimally decentralized model of care?Commissioners felt that exploring what is knownfrom other jurisdictions, as well as from BC andacross Canada, in a systematic and comprehensive waywould provide the scaffolding on which to build aframework to address conditions in British Columbia.Although the entirety of the review covered five discretethemes, the focus of this paper is on what was learnedfrom research literature on safety and outcomes. Thefrequent lack of policy and service context found inacademic literature is a considerable barrier to inter-jurisdictional learning [35], and so the contextual fea-tures of BC are made explicit with the intention of im-proving the international applicability of the findings.The findings from other themes are presented elsewhere.The full report is publically available [36].MethodsIn health service research, traditional meta- and system-atic reviews have significant limitations for stakeholdersin jurisdictions outside of the review setting. Context atevery level, including health system structures, healthprofessional relationships, historical precedent, andcommunity expectations all impact the portability of so-lutions from setting to setting. Given this, it was deter-mined that an efficacious way to look at models ofhealth care delivery and their applicability to the BritishColumbia context was through a realist review method,which brings a mandate to examine the totality of evi-dence on a research question with appropriate consider-ation for the dynamic policy and practice landscape inwhich that evidence was embedded. This method allowsresearchers to consider new questions and directions asthe literature is examined [37], particularly useful whensearching for models of care from other jurisdictions. Itis based on an approach Wong et al. [38] call “CMO”:understanding the complex relationship of Context,Mechanism, and Outcome. In addition to being context-ually located, evidence included in a realist review isbroad, reflective of the variety of influencing factorsinvolved.The RAMESES quality standard for realist reviewsguided the methodology [39] with the current studymeeting an excellent standard by most criteria (i.e., feas-ible topic, appropriately structured question, under-standing and application of realist philosophy, rigor ofappraisal process).This study emerged from a larger review initiated toaddress an evidence gap in best practices for meetingthe perinatal surgical needs of rural women guided bythe question “Can we meet the perinatal surgical needsof rural women more effectively through an optimallycentralized or optimally decentralized model of care?”Evidence was requested by Perinatal Services of BritishColumbia with a particular focus on optimal levels of(de)centralization within a planning context of budgetconstraint.Inclusion criteria for the search were research find-ings published in the English language since 1990 with atleast one search term from each of three areas (see Table 1).Placing limitations on the search parameters is consist-ent with a rapid review approach. Rapid review methodsKornelsen et al. BMC Health Services Research  (2016) 16:381 Page 3 of 13are often considered in relation to full systematic reviewsand have become common place in many health disci-plines. Rapid realist methodology, however, is still emer-ging and is not yet well defined. The procedures of thisresearch were held up to the scrutiny of rigorous realistreview guidelines, while the limitations placed on thesearch, the timeframe of the full review (under 6months), and the close involvement of an end-user arein keeping with a rapid review definition.A broad and iterative approach to the search termswas particularly important due to the consolidated na-ture of the question (evaluations of optimally centralizedor decentralized models) and the need to consider thethematic areas that would address the question (safety,outcomes, sustainability, costs, satisfaction). Wesearched MEDLINE, PubMed, EMBASE, CINAHL, EBMReviews, NHS Economic Evaluation Database, and PAISInternational for literature. The primary search wascompleted in December 2013. Grey literature was ob-tained from Perinatal Services of British Columbia andthe SAX Institute of Australia, and the review teammined bibliographies for further academic and greyresources, through March 2014.The further inclusion criteria were applied at the reviewstage of the full body articles that account for a relativelyhigh rate (30 %; 57 of 192) of exclusion upon full articlereview. Articles were included only if they focused ondirect discussions of maternal surgical care, includingbut not limited to safety of practice models, governanceof care models, and sustainability of service delivery.Articles were also included on the centralization of deci-sion making, ways of incorporating specialist care intoservice models, and optimal geography and/or level ofservice delivery. Much of the literature excluded at thefull article review stage was focused on internist, general,or other non-obstetric surgery for rural patients.Literature from low- and middle-resource settings wasexcluded manually due to lack of fit with BC’s healthservice delivery context. While BC and the world havemuch to learn from such health settings, there aremeaningful validity problems to synthesizing across dis-tinctly different health contexts around the world, espe-cially when applied to a question regarding medicalized,surgical intervention. The expectation of a realist reviewincludes answering which interventions work for whomunder what circumstances, such that material and cul-tural differences in health service settings challenge theappropriateness of a single review from all jurisdictions.Search parameters did include all settings, however.Consequently, literature was included from internationalcontexts deemed relevant to the context in BC: Scotland,the United Kingdom, Norway, Finland, Sweden, Holland,Germany, New Zealand, the United States, Australia,and the rest of Canada.The lead reviewer reviewed articles selected for inclu-sion and extracted appropriate data that was thenreviewed by the lead author. A sub-set of articles (8) wasreviewed by additional reviewers and compared forconsistency of extracted data. There was a high degreeof consistency between reviewers.Although the types of evidence found in the searchwere varied, the majority of studies were descriptive innature. Case studies, service reviews and chart audits in-cluding retrospective chart reviews were common, oftenusing population level data at the national or regionallevel. Additionally, several studies used chart reviews tocompare outcomes from specialist obstetric surgical careto generalist care. Program or intervention research wasa smaller portion of the research than expected, perhapsbecause of our focus on models of care rather thansmaller units of health service delivery. Still, a handful ofarticles detailed trials of new models of care, includingspecialist outreach and telehealth. Finally, editorials andgrey literature reports that were found with the help ofpolicy and service programming experts in both Canadaand Australia were also included. There were no existingsystematic reviews or randomized controlled trials in thebody of evidence. When considered together and in thebroader context of international rural maternity care,however, consistency of findings indicates reliable evi-dence. The review team approached the research litera-ture mindful of the importance of the role of context inthe outcomes of the intervention (GPESS), in contrast tothe more traditional cause-and-effect perspective. TheTable 1 Search Terms and Keywords. Search terms and areasfor inclusion criteria of literature searchSearch area Keywords ReasoningMaternal /PerinatalHealthobstetric*matern*perinatalreproduct*(birth or birthing)parturi*This review focuses on maternaland obstetric care, and soappropriate terms were furnishedto limit the search to that singulararea of care.PerinatalSurgical Caresurgerysurgical(cesarean orcaesarean orc-section*)We aimed for a broad surgicalrequirement, rather than anexhaustive list of obstetric surgeries.Rural andRemoteHealthServices(decentral* orde-central*)rural health*rural hospitalsrural communit*remote health*remote communit*“hub and spoke”ruralremoteThe review seeks to comparemodels of centralized anddecentralized care. Increasinglysince 1990, centralization of carehas been the backdrop of studiesregarding decentralized models.Moreover, this review seeks tocompare models of care in theirability to provide safe, high quality,cost effective perinatal surgical careto rural women specifically, and sorural health was a required searchsubject.Kornelsen et al. BMC Health Services Research  (2016) 16:381 Page 4 of 13focus on context was embedded in recommendationsthat were made based on the literature.The commissioners convened two province-widemeetings of policy makers, practitioners, and administra-tors working in and with GPESS at which findings ofthis study were considered. The second meeting focusedalmost exclusively on the findings from the review.These meetings served as an “expert panel” for the re-view team and allowed a high degree of confidence thatall relevant literature was included in the review.ResultsTwo hundred fifty-four articles were originally identi-fied as relevant from database searching. Upon con-sultation with the commissioners, one hundrednineteen articles were removed from consideration forlack of fit. These included articles highlighting clinicalevidence on the relative safety of particular morbid-ities for parturient women (e.g. eclampsia, diabetes,HIV) (n = 27) and articles regarding defensive medi-cine and litigation concerns (n = 35). In total, 191 ar-ticles were subjected to in-depth review. Fifty-threepertained directly to safety and outcomes (see Fig. 1).A reference table is attached (see Additional file 1). Asupplementary bibliography is also provided of thosearticles included in the full realist review but not spe-cifically relevant to safety (see Additional file 2).General practitioners with enhanced procedural skillshave historically provided cesarean section support forrural maternity units in many jurisdictions internation-ally, including the United States [27, 40–48], Australia[49–52], Scotland [19, 53, 54], New Zealand [55], andacross Canada [17, 22, 56–59]. Internationally, concernover volume thresholds and competency has ultimatelyled to a series of studies on the safety and service qualityof GPs with Enhanced Surgical Skills. This research evi-dence can be categorized into five sub-categories: bestpractice standards, community expectations of safety,outcome comparisons by service provider, relationshipof volume to outcomes, and consequences of small ser-vice closures. Each theme will be explored below.Best practice standardsStudies on best practices focus on either comparisons topublished standards of practice or comparisons to specialistoutcomes. One of the earliest contributions to this field isthe retrospective chart audit at two rural hospitals in theUS states of Washington and Oregon from, 1978–1992 byDeutchman et al. [60]. The authors found that GPs per-formed 79 % of cesarean section procedures at thoseFig. 1 Exclusion Tree. The process of literature search exclusionKornelsen et al. BMC Health Services Research  (2016) 16:381 Page 5 of 13hospitals. Reviewing the data from these deliveries, theauthors concluded that GPs met or exceeded all stan-dards of surgical outcomes in the published medical lit-erature. An Australian study [61] reported on data on 5950deliveries performed by GPs in rural New South Wales,Australia, between 1990 and 1991, and concluded that“[t]here is no evidence that obstetric care in NSW ruralhospitals with accredited obstetric units is below stan-dards acceptable to the community” (p242) whencompared against all 88,275 deliveries in New SouthWales in the same period.International descriptive studies found similar re-sults from GPESS-supported units. Kirke [62] lookedat 195 births at a remote hospital with GPESS care600 km east of Perth, Australia. Though complex andhigh-risk pregnancies were referred early, manywomen still in care went on to develop antenatal riskfactors including hypertension, obesity, and pre-eclampsia, and the catchment population reported ahigh level of gestational diabetes. Intrapartum andpost-partum complications such as maternal sepsis,antepartum hemorrhage, shoulder dystocia, failure toprogress, and fetal distress occurred at rates similarto regional averages. No perinatal or maternal mortal-ity was experienced in the study period, and healthoutcomes reported were as safe for mothers and babiesas the specialist-led units. Cameron and Cameron [63]used obstetrical audit data from 1991–2000 at theGPESS-led rural Atherton hospital near Cairns,Australia, to show that perinatal mortality (stillbirthplus neonate death within 28 days) was substantiallylower than the state average (5.3 per 1000 vs 11.8 forQueensland State or 11.8 for the Far North Queens-land county). This unit was run by GPs, some ofwhom held an obstetrics diploma, with specialist sup-port 96 kms away and access to outreach and evacu-ation services for only part of the study period. Thecommunity received four to six visits per year fromspecialist obstetrician-gynaecologists provided by theFar Northern Region Obstetrics and Gynaecology Ser-vice (FROGS).In another Australian study, Scherman, Smith, andDavidson [64] studied the outcomes of a midwife-ledunit with GP surgical support and OB specialist consult-ation in its first year (n = 164 births). The unit had lowantenatal (10 %) and intrapartum (4 %) transfer, and92 % spontaneous vertex delivery (i.e. 8 % intervention,including c-section, instrumental delivery, and breechbirth). No Apgar scores below 7 were recorded at 5 min,and 89 % of neonates required no resuscitation. The rateof perinatal injury was half the state average at just 27 %.Though midwives led the unit, the authors contend thatthe low transfer rate was possible because of GP surgicalsupport in the event of emergency.Comparison between levels of providersA sub-set of the research reviewed compared GPESS-ledservices to specialist-led models. Aubrey-Bassler et al.[65] studied outcomes in four Canadian provinces (BC,Alberta, Saskatchewan, and Ontario), considering 1448c-sections by 15 rural GPs and 4344 by specialists. Datawas collected from Discharge Abstracts between 1991and 2000, and showed that rates of iatrogenic morbiditywere higher among GPs (OR 1.6; CI 1.1–2.3; 2.5 % vs.1.6 % for specialists). However, this was accounted forby the difference in rate of puerperal infection (1.6 % vs.0.8 % for specialists). Surgical error was the same be-tween groups. GP proceduralists did, however, havehigher rates of referral to acute care and their patientshad longer post-surgical hospital stays (by 5.5 h onaverage).These findings were echoed by Homan, Olson, andJohnson [16] in a smaller study between two comparablehospitals in New England. Using 125 consecutive c-sections from each hospital – one with GP-led maternalsurgical care and the other with specialist-led surgery –this study found no difference in intraoperative or infec-tious complications, and no difference in neonate out-comes. Demographics of delivering mothers, prenatalrisk factors, and indications for c-section were found tobe similar between the two samples. The GP-led unit ex-perienced fewer post-operative complications in contrastto the findings of Aubry-Bassler et al. [65], but theobstetrician-led unit did have a shorter post-operativestay.Lynch et al. [66] compared two hospitals in BritishColumbia, one with c-section capability (Bella Coola)and one without (Haida Gwaii). In both communities,transfer or referral required considerable travel time andcould be delayed by inclement weather. Between the twohospitals, there were no differences in adverse outcomesand no maternal deaths were reported in the studyperiod (1986 to 2000) for either unit. The primary differ-ence was in referral rates. Almost 20 % more localwomen were able to deliver in a c-section capable mater-nity unit than in the unit without surgical support dueto the higher risk tolerance local operative serviceallows.In the studies noted above, GPESS cases were pre-selected to include only low-risk courses of care withknown complications referred to specialist obstetriciansprior to delivery, diminishing the strength of findings.Using population level data addresses this methodo-logical shortcoming, as demonstrated in the studiesbelow.The largest study of this kind in British Columbia ex-amined 87,294 singleton births between 2000 and 2007.Grzybowski, Stoll, and Kornelsen [20] compared birthsfrom catchment areas with GPESS surgical support (n =Kornelsen et al. BMC Health Services Research  (2016) 16:381 Page 6 of 139,174) to the outcomes from obstetrician serviced catch-ments (n = 54,714). Using two-step logistic regressionanalysis to predict rates of adverse perinatal outcomes,the authors showed that health outcomes were compar-able between GPESS-led surgical units, mixed-modelunits with both GPESS and specialists, and obstetriciansurgical units. The authors found that 80 % of womendelivered locally with GPESS support, while only 25 %could do so in communities without any surgical capability.Iglesias et al. [22] used population data is their studyof births in Alberta in 1999–2000, which examined pa-tient outflow (the rate of patients leaving the communityfor care) and maternal-newborn outcomes based onlevel of local maternity services. The study illustratesthat areas with limited maternity services are likely tohave an increased rate of induction, and that in commu-nities without local c-section capability there is largeoutflow. Communities that offered intrapartum carewithout local c-section capability delivered 22.1 % of thematernity population and this number increased to70.1 % in communities with local c-section capabilities(level 1C).Tucker et al. [19] found very similar rates in Europe’smost centralized health care system in Scotland. Compar-ing 1400 deliveries from eight of the twelve rural maternitycatchments of Scotland, the authors demonstrated thatroughly the same percentage of women remained “low-risk”throughout their pregnancy, and similarly, the rate of spon-taneous vaginal delivery was stable when measured bycatchment area rather than birth unit. Though low-riskcases were managed well by low-resource units, greateroutflow from catchments with 1A equivalent servicesthreatened sustainability. As with the Iglesias et al. [22]study above, midwife-only units (no surgical capability)were only able to perform 31 % of local deliveries, whilemidwife-led units with GP surgical support managed 70 %of local cases, and OB-led units performed 86 % of thebirths from their local catchments. Thus, the low interven-tion rates found in midwife-only and midwife-led units inother studies are shown to be reliant on referral and surgi-cal support, as to be expected in a tiered service model witha risk management mandate.Similar referral numbers appear in all population leveldata found for this review. Kornelsen, Grzybowski, andIglesias [21] found that with GPESS support in a com-munity, between 78 % and 85 % of births take place lo-cally in BC and Alberta. Without c-section capability,that rate falls to between 24 % and 35 %. Humber andDickinson [18] reported the most optimistic numbers,finding rates of 85 % and 40 % respectively.Service size and outcomes: is there a relationship?Considerable attention is paid in the literature linkingthe size of maternity units with procedural outcomes,with some of the research evidence showing that theoutcomes of small units are comparable to larger ser-vices. However, three studies indicate an outcomes dis-advantage for small units, specifically among neonates.A controversial study from Moster, Lie, and Markestad[67] found that Norwegian maternity units with 2000–3000 births per year had better outcomes than smallerunits. This study looked at 700,000 low risk singletonbirths between 1972–1995 and found that units with<100 annual deliveries were almost twice as likely (OR1.8; 1.1–3.1) to experience a late neonatal death (within28 days of birth) than a unit with 2000-3000 births peryear. However, the methodology of this study has limita-tions and several other studies undermine the power ofmany of the central claims by Moster, Lie, and Markes-tad [67].Norum et al. [68] studied births from the scattered,northern, remote population of Norway and concludedthat a very decentralized model of care that gave rise tosmaller maternity units was necessary for a countrywhere inclement weather and seasonal darkness makestransfer and even referral challenging. The pressingquestion is not whether the births that happened inhigher level units were safer, but whether intrapartumcare to women living in rural and remote areas would besafer and achieve better outcomes under centralizedconditions. That is, when taking into account real-world,geographic constraints, what is the health cost of nolocal care? By excluding all out-of-hospital deliveries intheir analysis, namely those that occurred during trans-fer, and by not considering the attendant challenges andhealth impacts of greater (or total) referral to centralizedmaternity units, Moster, Lie, and Markestad [67] avoid acritical geographic reality.On the other hand, Viisainen et al. [69] examined acci-dental, out-of-hospitals births in Finland between 1962–1973, and compared them to data from 1992/93 (thisdata was not tracked in Finland between 1973 and1992). Between 1962 and 1973, the rate of accidental,out-of-hospital birth fell from 1.3 per 1000 to 0.4 per1000 whereas in 1992/93 it had reached 1.0 per 1000 livebirths. Viisainen et al. [69] argued there was a connec-tion between the closure of small units and the rise inaccidental, out-of-hospital births, events known to haveexceptionally poor outcomes relative to delivery in hos-pitals. In fact, the crude risk factor for perinatal deathwas six times higher among babies born accidentally outof hospital, and over three times higher when birthweight is controlled [69, 70].Despite increased concern over accidental, out-of-hospital births in Finland, the rate continued to increaseduring the 2000s according to Hemminki, Heino, andGissler [70]. Their study of all births in Finland from1991–2008 found that among children born weighingKornelsen et al. BMC Health Services Research  (2016) 16:381 Page 7 of 13>2500 g (the same low-risk cut-off used by Mosler, Lie,and Markestad, [67] above), mortality was similar acrossall hospital types, sizes, and locations. However, thenumber of maternity units in Finland decreased 31 %over that span while births declined just 9 %, and acci-dental, out-of- hospital births increased. Of note, therate normalized across regions during the study period,indicating that not just rural and remote women sufferedthis care deficit, but that urban-adjacent women alsobegan to experience unplanned, out-of-hospital births inincreasing numbers. This fits with data reported byGrzybowski, Stoll, and Kornelsen [6] from BC,Canada, that women between one and two hoursfrom services were more than six times (OR = 6.41;CI 3.69–11.28) more likely to have an unplanned,out-of-hospital birth. Hemminki, Heino, and Gissler[70] provide a strong case for the need for smaller,local-to-mothers birthing units, concluding, “[t]heanalysis suggests that in a regionalized system with afunctioning referral system, there is no need to closedown small hospitals for reasons related to health orhealthcare procedures” (p1191).Their conclusion echoes that of another Finnish studyby Viisainen, Gissler, Hartikainen, and Hemminki [71].Population birth data from 1987/88 was analyzed by ser-vice level of delivery hospital and catchment, selectedfor low-risk deliveries (n = 123,065). Their study showedgood outcomes for all levels of service when low-weightand premature neonates and those requiring surveillancewere cared for in hospitals providing the highest level ofcare (level 3). In a population catchment analysis [71],women determined to be low-risk had similar outcomesregardless of the hospital type at which they delivered;“[T]his study… indicates that ‘safety’ cannot be used as abasis for centralizing birth care in large level 3 facilities”(p404).In a study done by Heller et al. [72], however, au-thors found a gradient of worsening outcomes fromthe largest and best resourced to the smallest birthunits in Hesse, Germany. Looking at 582,655 birthsbetween 1990–1999, they reported that in units with<500 births per year, early neonatal death (within7 days of birth) is three times more likely than inunits with >1500 births annually. However, the au-thors note that without information on staffing, skill,training, levels of collaborative practice, and other in-dicators of quality of care within the delivery units,the influence of size of hospital in rates of highermortality is unknown. Interestingly, this study usesthe most inclusive definition of “low-risk,” calling allbabies born of normal weight (2500 g–4200 g) with-out death by congenital abnormality a low-risk preg-nancy and birth. Analysis that controlled for time ofbirth and gestational age and included late neonataldeath (within 28 days) yielded similar results. In theseanalyses, however, maternal confounders were notcontrolled for.Merlo et al. [73] also found a small unit outcome dis-advantage, this time in Sweden, and attempted to definethe percentage of proportional change in risk of neonatalmortality by birthing unit size. Using a multilevel logisticregression in which the outcomes of all births between1990–1995 (n = 691,742) were nested in hospital leveloutcomes (n = 66), a confounder to hospital size was dis-covered. Just 4 % of Sweden’s institutionalized birthstake place in units with <500 annual births and withouta pediatrics department, and this group showed the lar-gest risk for neonatal mortality. The authors note, how-ever, that the absolute survival rate in these relativelyhigher-risk birthing environments was 99.9 %, and theabsolute survival difference compared to large regionalhospitals was 0.06 % (or 0.6 deaths per 1000 births).In response to these earlier studies, Tracy et al. [74]examined over 750,000 births over three years inAustralia to compare outcomes by birthing unit annualvolume. The study was limited to low-risk women.Among women without pre-existing or antenatal onsetof hypertension or diabetes, and whose babies were bornat >2500 g, rates of mortality were comparable in unitswith fewer than 100 deliveries and those with 2000 ormore. Units of all sizes were found to have very similaroutcomes, while smaller units tended to have less inter-vention, including lower rates of c-section [74]. Import-antly, Tracy et al’s [74] categories for unit size andchosen sample size are in direct reference to Moster etal.’s [67] study, noted above.Taken together, the differences in outcomes found byHeller et al. [72], Merlo et al [73], and Moster et al. [67]must be interpreted through a lens of clinical as well asstatistical significance with attention paid also to poten-tial iatrogenic costs due to lack of local access and travel.Further, the larger context of acceptable outcomes is im-portant. Norum et al. [68] report a neonatal mortalityrate of 2.2 per 1000 for all births in Northern Norway,and a national rate of 2.3 per 1000. For context, as of2011, Germany also achieved a neonatal mortality rateof just over 2 deaths per 1000 births, roughly half ofCanada’s rate of 4.7 [75]. Exceptional outcomes havealready been achieved in small units from an inter-national perspective, and the attendant health costs ofgreater centralization remain unknown in these threeEuropean studies.Finally, there is a potential confound in the data ofboth Heller et al. [72] and Moster et al. [67]: the relativehealth of the adult population. Rural Canadians suffer aknown health disadvantage compared to urban popula-tions [76]. A study from Sweden by Finnstrom et al. [77]found lower rates of neonatal death, respiratoryKornelsen et al. BMC Health Services Research  (2016) 16:381 Page 8 of 13disturbance, cerebral palsy, and 5-min Apgar scores of<4 in smaller delivery units when controlling for mater-nal age, parity, gestational age, smoking during preg-nancy, maternal body mass index, and parentcohabitation. Their massive study of 1.5 million single-ton births between 1985 and 1999 found that in unitswith <500 annual births, the odds of neonatal death wasjust 0.84 (CI 0.63–1.11) compared to the reference cat-egory of units with 1000–2499 annual births [77], due inpart to appropriate referral. Those units with 500–999births did slightly better with an odds ratio of 0.82 (CI0.73–0.92) of neonatal death. The authors found, as didMerlo et al. [73] above, that the existence of a pediatricsdepartment played a significant role in lowering the neo-natal mortality rate in smaller units, but the absolutenumbers were too small to be statistically significant.They conclude that regionalized referral is functioningand that care is of a relatively homogeneous qualityacross unit size. These findings were validated in Swedenby Serenius et al. [78] when they examined the causeand context of all 9785 stillbirths and neonatal deaths inSweden between 1983–1995. Again, data was controlledfor maternal age, parity, and smoking during pregnancy,and again, the smallest units were found to be less likelyto experience a death (OR = 0.65; CI 0.61–0.70). Efficientreferral ensured that high-risk pregnancies were central-ized to high-resource settings, while lower risk pregnan-cies showed strong outcomes when controlled for basicindicators of maternal health.Volume in relation to outcomesThe challenge of providing local access to cesarean sec-tion in rural settings rests in the low volume of proce-dures likely to be required among a low-risk population(assuming prior referral of parturient women with riskfactors). The attendant concerns are regarding the main-tenance of provider competency. However, volume-to-outcome associations are under-studied in Canada, andassociations specific to maternal surgery are under-studied worldwide. In a review of volume-to-outcomeassociation studies in the United States and Canada,Urbach et al. [79] found that Canada’s public health sys-tem considerably reduced the effect of volume on out-comes. Of 278 separate analyses reported in 142 articlesreviewed by Urbach et al. [79], 206 (74 %) found a statis-tically significant association. Canadian studies weremuch less likely to find any association (OR = 0.24; CI0.08–0.74). Though obstetrical specific data was col-lapsed into an “Other” category in Urbach et al.’s [79]analysis, even surgeries known to have a volume-to-outcome association (such as complex heart procedures)were shown to have a lesser effect intensity in Canadacompared to the United States. The authors concludedthat a single-payer, globally financed care system withregionalized organization reduces volume concerns, ascomplex procedures are already referred to high-levelcare facilities without inter-facility competition. How-ever, only 14 of the 142 studies found by Urbach et al.[79] reported on Canadian data and just four of thestudies included data on obstetrical procedures.Using all births attended by family physicians at BCWomen’s Hospital and Health Centre from 1997–1998(n = 4,444 births), Klein et al. [80] analyzed outcomes ac-cording the personal volume of attending family physi-cians (n = 152 physicians). Thresholds of <12, 12–24,and >25 were used to explore whether attending morebirths led to better birth outcomes, but no differenceswere found in the volume cohorts in maternal complica-tions, 5-min Apgar scores <7, or adverse admissions tointensive or special care. Low-volume GPs were morelikely to consult with an obstetrician and more likely totransfer care to a specialist, but outcomes were not af-fected by attending a lesser volume of births.Distance mattersExamining 49,402 births to women from rural catch-ments between 2000–2004, Grzybowski, Stoll, and Kor-nelsen [6] found that neonatal mortality was three timesmore likely for births in which the women had to travelfour or more hours to services (OR = 3.17; CI 1.45–6.95). As well, induction was found to be 1.3 times morelikely in women who had to travel to services, mostly forlogistical reasons [81].Even in the relatively more dense Netherlands, lon-ger travel times are associated with worsened out-comes [82]. Travel of more than 20 min to careresulted in higher total mortality (OR = 1.17; CI1.002–1.36), higher neonatal mortality within 24 h(OR = 1.51; CI 1.13–2.02), and greater rates of adverseoutcomes (OR = 1.27; CI 1.17–1.38) in Ravelli et al’s[82] study of 751,926 births in Holland between2000–2006. Few women in the Netherlands travelmore than 30 min (as measured by driving time with-out delays) to birthing services, which contrasts withthe geographic realities of BC. However, their findingof an odds ratio of additional risk of 1.01 (CI 1.00–1.01) per minute of travel time corroborates the find-ings of Grzybowski, Stoll, and Kornelsen [6] above.Though no one in the Netherlands would have totravel four hours (240 min) to service, by extrapolat-ing Ravelli et al’s [82] per-minute findings, the in-creased risk of neonatal mortality for such a longtravel time would be OR = 3.40 – just slightly higherthan the 3.17 number found here in BC. Such a find-ing from a very different health context is evocativewhen considering the centralization of services as amethod of improving outcomes.Kornelsen et al. BMC Health Services Research  (2016) 16:381 Page 9 of 13DiscussionResearch literature has shown that local access tocesarean section increases the proportion of womensafely able to deliver in their local community to at least70 % from 30 % in services not offering local cesareansection capacity. Finding and supporting the health hu-man resource compliment in communities with enoughvolume to sustain such services, however, has been diffi-cult. The very nature of rural services is defined by lowvolume, making specialist practice in the smaller com-munities unfeasible. The solution in the international ju-risdictions covered in this review has been a reliance onGPs with Enhanced Surgical Skills. Due to the numberof sites supporting GP procedural practice and the num-ber of evaluative studies that have resulted, research evi-dence on the safety and efficacy of this practice isstrong. Perhaps equally as importantly, there is no exist-ing clinical, case study or qualitative evidence to suggestthat cesarean section is less safe when provided by aGPESS than when provided by a specialist obstetrician.Supporting and sustaining local maternity services iscrucially important to achieving good perinatal healthoutcomes. Although the proportion of outflow from thecommunity is reduced with local operative delivery, re-search evidence also tells us that the lack of any localmaternity service is worse than services withoutcesarean section. This is due in part to the unintendedmorbidities incurred when women present to an unpre-pared service fully dilated, or physiological and psycho-social morbidities, caused by travelling to access care.Additionally, health service realities – including thelack of continuity of care when women leave their com-munities – must be accounted for in a comprehensivereview of safety of the evidence on small local surgicalservices.Enmeshed in concerns over the safety of the practiceof GPESS, there has also been the ongoing debate onpractice thresholds; that is, the number of proceduresperformed, both by individual clinicians or in facilities,in relation to outcomes. The literature in this reviewsuggests that volume-to-outcomes associations are ex-tremely variable across procedure and context, but as awhole greater birth volume does not improve birth out-comes. This does not speak to greater procedural vol-ume, however, specifically in regards to cesareansections. Although we do know that greater volume in-creases confidence (particularly greater volume in resi-dency) [36], careful consideration of the relationshipbetween GPESS volume of cesarean sections and out-comes is a crucial gap in our evidence and in need offurther investigation.Although a context-mechanism-outcomes (CMO) the-ory was not postulated at the onset of this review due tothe pragmatic intent of the commissioners, it is clearthat an a posteriori understanding of CMO can beunderstood from the reviewed literature and applied tothe creation of evidence-based models of care.The context for these models must include a statementof support from a governance level recognizing the im-portance of meeting the perinatal surgical needs of ruralwomen as close to home as possible, respecting com-plexity of procedure, risk status of patient, and healthconditions in the community. Additionally, surgical careshould be viewed as a regional, rather than institutional,phenomenon. Consequently, the scope of practice andresources needed to implement surgical programsshould be organized regionally. General Practitionerswith Enhanced Surgical Skills (GPESS) can be safelyallowed to practice to the fullest extent of their abilitywithin the context of a regionalized and inter-professional system of referral, consultation, and emer-gency transfer support. Small ORs should become out-reach extensions of core referral hospital surgicalprograms, and the organization of services should re-spect the sustainability of the regional referral servicesand the smaller services.The mechanism needed to enact the vision of continu-ous perinatal surgical services (24/7 c-section backup)includes services provided through a well-integrated andbalanced surgical team, which includes outreach sur-geons and local generalist surgical providers. Surgicalcompetency could be enhanced by regular rotation ofteam members through a larger referral centre. Trainingprograms for rural nurses need to be strengthened, rec-ognizing the broad skillset and multifaceted nature ofrural nursing. Small service surgical team skills andcompetencies should be built and maintained throughan integrated educational program with local referralhospitals. This can be accomplished both through out-reach and by rotating small service surgical team mem-bers through the referral community’s surgical program.Additionally, inter-professional outreach surgical educa-tional and mentorship activities extended from the re-gional referral hospital to the small surgical sites on aregular basis. This model requires timely and regularfeedback within a quality improvement framework.Anticipated outcomes of the context and mechanismdescribed include a robust and sustainable networkedmodel of rural surgical services equipped to meet basicperinatal surgical needs of rural populations.LimitationsFollowing the indications of realist reviews to privilegecontext in the structure of the review and the interpret-ation of findings, this review is directly applicable to thehistory, context and political challenges in BritishColumbia, Canada. Although similar conditions may befound in other jurisdictions, the exact constellation ofKornelsen et al. BMC Health Services Research  (2016) 16:381 Page 10 of 13conditions will not be the same, thus limiting transfer-ability to other settings.This review was commissioned in response to a real-time planning challenge that arose out of a provincialpriority-setting initiative informed by external time-lines.Because of this, a rapid review approach was used. Thisdemanded attention to the balance of comprehensivenessand timeliness. Although methodological rigor was ap-plied throughout the process, the exhaustiveness of thesearch could be potentially limited due to time con-straints. This was addressed through the involvement ofthe expert panel, made up of key clinically, politically,and administratively engaged stakeholders in the prov-ince. This scrutiny of the review allowed room for theinclusion of grey literature references not capturedthrough the search strategy.ConclusionClinical, case study, and qualitative evidence demon-strates that perinatal surgical care is equally safe whenprovided by GPESS and specialist physicians. This find-ing allows health planners to confidently build perinatalsurgical services around the contribution of GPs withEnhanced Surgical Skills and focus on educational, regu-latory, and continuing professional development mecha-nisms to ensure their sustainability. Volume-to-outcomes associations are variable and inconclusive withregards to safety, suggesting both the need for more evi-dence and also the viability of low-volume services par-ticularly suited to generalists who can take on otherroles in the community. These findings, and the attend-ant health services planning directions, are reassuring asthey suggest the viability of local models of care wherefeasible. This policy direction addresses the social andhealth risks to women from communities without localaccess to maternity care, leading to improved healthoutcomes.Additional filesAdditional file 1: Article Reference Chart. Details of each of the 53articles from the full realist review considered for its contribution tosafety and outcomes and included in this manuscript. (PDF 305 kb)Additional file 2: Supplementary Bibliography. All the articles includedin the full realist review are detailed here. (PDF 192 kb)AcknowledgementsAppreciation is extended to members of the Centre for Rural HealthResearch who contributed to the reviews including Meagan McKeen, CaitlinFrame, Kelly Garton, Penny Yang, and Mary O’Sullivan. Additional thanks toTaylor Flemming who reviewed articles and prepared the manuscript forpublication, and to Renee Turner for editorial comments. Appreciation is alsoextended to Dr. Stefan Grzybowski for comments on the draft and supportthrough the review process.FundingResearch was made possible with funding provided by the CanadianInstitutes of Health Research (CIHR) and the Michael Smith Foundation forHealth Research (MSFHR).Availability of data and materialData sharing not applicable to this article as no datasets were generated oranalysed during the current study.Authors’ contributionsJK carried out the study, co-articulated the research question, and co-wrotethe manuscript. KM led the review of articles and wrote up findings from thedata extraction. KW co-articulated the research question, provided the polit-ical context for the review, organized the expert panel meeting, and editedthe manuscript. All authors read and approved the final manuscript.Competing interestsThe authors declare that they have no competing interests.Consent for publicationNot applicable.Ethics approval and consent to participateNot applicable.Author details1Department of Family Practice, University of British Columbia, DavidStrangway Building, 3rd Floor, 5950 University Blvd., Vancouver V6T 1Z3, BC,Canada. 2Perinatal Services BC, Provincial Health Services Authority, WestTower, Suite 350, West 12th Ave., Vancouver V5Z 3X7, BC, Canada.Received: 29 July 2014 Accepted: 4 August 2016References1. Hutten-Czapski P. Decline of obstetrical services in northern Ontario. CJRM.1999;4(2):72–6.2. Hutten-Czapski P. The state of rural healthcare. Presentation to the StandingSenate Committee on Social Affairs, Science and Technology; 2001 May 31;Ottawa, Canada.3. Nesbitt TS, Connell FA, Hart LG, Rosenblatt RA. Access to obstetric care inrural areas: effect on birth outcomes. Am J Public Health. 1990;80(7):814–8.4. Church J, Barker P. Regionalization of health services in Canada: a criticalperspective. Int J Health Serv. 1998;28(3):467–86.5. James AM. Closing rural hospitals in Saskatchewan: on the road to wellness?Soc Sci Med. 1999;49(8):1021–34.6. Grzybowski S, Stoll K, Kornelsen J. Distance matters: a population basedstudy examining access to maternity services for rural women. BMC HealthServ Res. 2011;11:147–53.7. Biem HJ, Hadjistavropoulous H, Morgan D, Biem HB, Pong RW. Breaks incontinuity of care and rural senior transferred for medical care underregionalisation. Int J Integr Care. 2003;3:e03.8. Gregory D, Way C, Barrett B, Parfrey P. Health care quality from theperspective of health care providers and patients during and shortly afteracute care restructuring in Newfoundland and Labrador. J Health Serv ResPolicy. 2005;10(S2):S48–57.9. Pollett WG, Harris KA. The future of rural surgical care in Canada: a time foraction. Can J Surg. 2002;45(2):88–9.10. Rebbeck PM. General surgeons: more are needed. BCMJ. 2005;47(7):379–81.11. SOGC. Returning birth to Aboriginal, rural, and remote communities [PolicyStatement]. Ottawa: SOGC; 2010. 3 p. No. 251.12. Carstairs S, Keon WJ. Canada’s aging population: seizing the opportunity.Ottawa: The Senate (Special Senate Committee on Aging); 2009. p. 247.13. Roos N, Black C, Wade J, Decker K. How many general surgeons do youneed in rural areas? Three approaches to physician resource planning insouthern Manitoba. CMAJ. 1996;155(4):395–401.14. Chiasson PM, Henshaw JD, Roy PD. General practice patterns in Nova Scotia:the role of the “generalist” general surgeon. Can J Surg. 1994;37(4):285–8.15. British Columbia Royal Commission on Health Care and Costs. Closer tohome: the report of the British Columbia Royal Commission on Health Careand Costs (Vol. 2). Victoria: Crown Publications; 1991.Kornelsen et al. BMC Health Services Research  (2016) 16:381 Page 11 of 1316. Homan FF, Olson AL, Johnson DJ. A comparison of cesarean deliveryoutcomes for rural family physicians and obstetricians. J Am Board FamMed. 2013;26(4):366–72.17. Iglesias A, Iglesias S, Arnold D. Birth in Bella Bella: emergence and demise of arural family medicine birthing service. Can Fam Physician. 2010;56(6):e233–40.18. Humber N, Dickinson P. Rural patients’ experiences accessing surgery inBritish Columbia. Can J Surg. 2010;53(6):373–8.19. Tucker J, McVicar A, Pitchforth E, Farmer J, Bryers H. Maternity care modelsin a remote and rural network: assessing clinical appropriateness andoutcome indicators. Qual Saf Health Care. 2010;19(2):83–9.20. Grzybowski S, Stoll K, Kornelsen J. The outcomes of perinatal surgicalservices in rural British Columbia: a population-based study. Can J RuralMed. 2013;18(4):123–9.21. Kornelsen J, Grzybowski S, Iglesias S. Is rural maternity care sustainablewithout general practitioner surgeons? Can J Rural Med. 2006;11(3):218–20.22. Iglesias S, Bott N, Ellehoj E, Yee J, Jennissen B, Bunnah T, Schopflocher D.Outcomes of maternity care services in Alberta, 1999 and 2000: apopulation-based analysis. J Obstet Gynaecol Can. 2005;27(9):855–63.23. Blair L. Are general surgeons a dying breed? CMAJ. 1991;145(1):46–8.24. Burke L. Projections of physician supply in Canada. Ottawa: CanadianCollaborative Centre for Physician Resources, Canadian Medical Association;2007. p. 11.25. Inglis FG. Surgical care in rural Canada: training and planning for the future[Editorial]. CMAJ. 1995;153(10):1453–4.26. Barer ML, Stoddart GL. Towards integrated medical resource policies forCanada. CMAJ. 1992;147(7):999–1005.27. Campbell NA, Kitchen G, Campbell IA. Operative experience of generalsurgeons in a rural hospital. ANZ J Surg. 2011;81(9):601–3.28. Lynge DC, Larson EH, Thompson MJ, Rosenblatt RA, Hart LG. A longitudinalanalysis of the general surgery workforce in the United States, 1981-2005.Arch Surg. 2008;143(4):345–50.29. Lynge DC, Larson EH. Workforce issues in rural surgery. Surg Clin North Am.2009;89(6):1285–91.30. Williams Jr TE, Ellison EC. Population analysis predicts a future criticalshortage of general surgeons. Surgery. 2008;144(4):548–54.31. Humber N, Frecker T. Delivery models of rural surgical services in BritishColumbia (1996-2005): are general practitioner-surgeons still part of thepicture? Can J Surg. 2008;51(3):173–8.32. Iglesias S, Jones L. Rural Surgical Programs in Western Canada. CJRM.2002;7(2):103–7.33. Iglesias S, Strachan J, Ko G, Jones LC. Advanced Skills by Canada’s RuralPhysicians. CJRM. 1999;4(4):227–31.34. Chiasson PM, Roy PD. Role of the general practitioner in the delivery ofsurgical and anesthesia services in rural western Canada. CMAJ. 1995;153(10):1447–52.35. Kildea S, Van Wagner V. ‘Birthing on country’ maternity service deliverymodels: a rapid review. Haymarket, NSW: Sax Institute; 2012. p. 66.36. Kornelsen J, McCartney K. Optimal perinatal surgical services for ruralwomen: a realist review. APRU: Vancouver; 2014. p. 119.37. Pawson R, Greenhalgh T, Harvey G, Walshe K. Realist review–a new methodof systematic review designed for complex policy interventions. J Heal ServRes Policy. 2005;10(1):21–34.38. Wong G, Greenhalgh T, Westhorp G, Buckingham J, Pawson R. RAMESESpublication standards: realist syntheses. BMC Med. 2013;11(1):21.39. Wong G, Greenhalgh T, Westhrop G, Pawson R. Quality standards for realistsyntheses and meta-narrative reviews. London: RAMESES;2014. p. 24.40. Baker E, Schmitz D, Epperly T, Nukui A, Miller CM. Rural Idaho familyphysicians’ scope of practice. J Rural Health. 2010;26(1):85–9.41. Barclay AM, Knapp DP, Kallail KJ. The provision of labor and delivery servicesby graduates of four Kansas family practice residencies. Kans Med. 1996;97(1):19–23.42. Breon TA, Scott-Conner CE, Tracy RD. Spectrum of general surgery in ruralIowa. Curr Surg. 2003;60(1):94–9.43. Callaghan J. A twenty-five year survey of a solo practice in rural surgicalcare. J Am Coll Surg. 1994;178(5):459–65.44. Hueston WJ, Murry M. A three-tier model for the delivery of rural obstetricalcare using a nurse midwife and family physician copractice. J Rural Health.1992;8(4):283–90.45. Rosenthal TC, Holden DM, Woodward W. Primary care obstetrics in rural WesternNew York: a multi-center case review. NY State J Med. 1990;90(11):537–40.46. Smith DW, Murphy NJ. Alaska’s obstetrical delivery systems: a descriptiveepidemiologic study. Alaska Med. 2000;42(3):78–84.47. Wadland WC, Havron AF, Garr D, Schneeweiss R, Smith M. National surveyon hospital-based privileges in family practice obstetrics. Arch Fam Med.1994;3(9):793–800.48. Young RA, Byrd AN. Practice patterns of rural Texas physicians trained in afull-service family practice residency program. Tex Med. 1999;95(2):64–8.49. Homer CS, Biggs J, Vaughan G, Sullivan EA. Mapping maternity services inAustralia: location, classification and services. Aust Heal Rev. 2011;35(2):222–9.50. Robinson M, Slaney GM, Jones GI, Robinson JB. GP proceduralists: ‘thehidden heart’ of rural and regional health in Australia. Rural Remote Health.2010;10(3):1402.51. Swayne A, Eley DS. Synergy and sustainability in rural procedural medicine:views from the coalface. Aust J Rural Health. 2010;18(1):38–42.52. Welch R, Power R. General practitioner obstetric practice in rural andremote Western Australia. Aust NZ J Obstet Gynaecol. 1995;35(3):241–4.53. Godden DJ. Rural health care in the U.K.: a rapidly changing scene. J AgricSaf Health. 2005;11(2):205–10.54. Tucker J, Hundley V, Kiger A, Bryers H, Caldow J, Farmer J, Harris F, Ireland J,van Teijlingen E. Sustainable maternity services in remote and ruralScotland? A qualitative survey of staff views on required skills, competenciesand training. Qual Saf Health Care. 2005;14(1):34–40.55. Simmers D. The few: New Zealand’s diminishing number of rural GPsproviding maternity services. NZ Med J. 2006;119(1241):U2151.56. Baker DK. Rural surgery in Canada. World J Surg. 2006;30(9):1632–3.57. Dooley J, Kelly L, St Pierre-Hansen N, Antone I, Guilfoyle J, O’Driscoll T. Ruraland remote obstetric care close to home: program description, evaluationand discussion of Sioux Lookout Meno Ya Win Health Centre obstetrics. CanJ Rural Med. 2009;14(2):75–9.58. Johnson D, Jin Y. Low-volume obstetrics. Characteristics of family physicians’practices in Alberta. Can Fam Physician. 2002;48:1208–15.59. Rourke JT. Trends in small hospital medical services in Ontario. Can FamPhysician. 1998;44:2107–12.60. Deutchman M, Connor P, Gobbo R, FitzSimmons R. Outcomes of cesareansections performed by family physicians and the training they received: a15-year retrospective study. J Am Board Fam Pract. 1995;8(2):81–90.61. Woollard LA, Hays RB. Rural obstetrics in NSW. ANZJOG. 1993;33(3):240–2.62. Kirke AB. How safe is GP obstetrics? An assessment of antenatal risk factorsand perinatal outcomes in one rural practice. Rural Remote Health. 2010;10(3):1545.63. Cameron B, Cameron S. Outcomes in rural obstetrics, Atherton Hospital1991-2000. Aust J Rural Health. 2001;9(S1):S39–42.64. Scherman S, Smith J, Davidson M. The first year of a midwifery-led model ofcare in Far North Queensland. Med J Aust. 2008;188(2):85–8.65. Aubrey-Bassler K, Newbery S, Kelly L, Weaver B, Wilson S. Maternaloutcomes of cesarean sections: do generalists’ patients have differentoutcomes than specialists’ patients? Can Fam Physician. 2007;53(12):2132–8.66. Lynch N, Thommasen H, Anderson N, Grzybowski S. Does having cesareansection capability make a difference to a small rural maternity service? CanFam Physician. 2005;51:1238–9.67. Moster D, Lie RT, Markestad T. Relation between size of delivery unit andneonatal death in low risk deliveries: population based study. Arch Dis ChildFetal Neonatal Ed. 1999;80(3):F221–5.68. Norum J, Heyd A, Hjelseth B, Svee T, Mürer FA, Erlandsen R, Vonen B.Quality of obstetric care in the sparsely populated sub-arctic area of Norway2009-2011. BMC Pregnancy Childbirth. 2013;13:175.69. Viisainen K, Gissler M, Hartikainen AL, Hemminki E. Accidental out-of-hospital births in Finland: incidence and geographical distribution 1963-1995. Acta Obstet Gynecol Scand. 1999;78(5):372–8.70. Hemminki E, Heino A, Gissler M. Should births be centralised in higher levelhospitals? Experiences from regionalised health care in Finland. BJOG. 2011;118(10):1186–95.71. Viisainen K, Gissler M, Hemminki E. Birth outcomes by level of obstetric carein Finland: a catchment area based analysis. J Epidemiol Community Health.1994;48(4):400–5.72. Heller G, Richardson DK, Schnell R, Misselwitz B, Künzel W, Schmidt S. Arewe regionalized enough? Early-neonatal deaths in low-risk births by the sizeof delivery units in Hesse, Germany 1990-1999. Int J Epidemiol. 2002;31(5):1061–8.73. Merlo J, Gerdtham U, Eckerlund I, Håkansson S, Otterblad-Olausson P,Pakkanen M, Lindqvist PG. Hospital level of care and neonatal mortality inKornelsen et al. BMC Health Services Research  (2016) 16:381 Page 12 of 13low- and high-risk deliveries: reassessing the question in Sweden bymultilevel reassessing analysis. Med Care. 2005;43(11):1092–100.74. Tracy SK, Sullivan E, Dahlen H, Black D, Wang YA, Tracy MB. Does sizematter? A population-based study of birth in lower volume maternityhospitals for low risk women. BJOG. 2006;113(1):86–96.75. UNdata. United Nations Statistics Division, New York. 2016. http://data.un.org/ Accessed 20 July 2016.76. DesMeules M, Pong R. How healthy are rural Canadians? An assessment oftheir health status and health determinants. Ottawa: Canadian PopulationHealth Initiative, Canadian Institute for Health Information; 2006. p. 205.77. Finnström O, Berg G, Norman A, Otterblad OP. Size of delivery unit andneonatal outcome in Sweden. A catchment area analysis. Acta ObstetGynecol Scand. 2006;85(1):63–7.78. Serenius F, Winbo I, Dahiquist G, Källén B. Cause-specific stillbirth andneonatal death in Sweden: a catchment area-based analysis. Acta Paediatr.2001;90(9):1054–61.79. Urbach DR, Croxford R, MacCallum NL, Stukel T. How are volume-outcomeassociations related to models of health care funding and delivery? Acomparison of the United States and Canada. World J Surg. 2005;29(10):1230–3.80. Klein MC, Spence A, Kaczorowski J, Kelly A, Grzybowski S. Does deliveryvolume of family physicians predict maternal and newborn outcome?CMAJ. 2002;166(10):1257–63.81. Kornelsen J, Moola S, Grzybowski S. Does distance matter? Increasedinduction rates for rural women who have to travel for intrapartum care. JObstet Gynaecol Can. 2009;31(1):21–7.82. Ravelli AC, Jager KJ, de Groot MH, Erwich JJ, Rijninks-van Driel GC, Tromp M,Eskes M, Abu-Hanna A, Mol BW. Travel time from home to hospital andadverse perinatal outcomes in women at term in the Netherlands. BJOG.2011;118(4):457–65.•  We accept pre-submission inquiries •  Our selector tool helps you to find the most relevant journal•  We provide round the clock customer support •  Convenient online submission•  Thorough peer review•  Inclusion in PubMed and all major indexing services •  Maximum visibility for your researchSubmit your manuscript atwww.biomedcentral.com/submitSubmit your next manuscript to BioMed Central and we will help you at every step:Kornelsen et al. BMC Health Services Research  (2016) 16:381 Page 13 of 13


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items