trade-off between local access and safety considerations in childbirth: rural tasmanian women's...

Original Article

Trade-off between local access and safety considerations inchildbirth: Rural Tasmanian women’s perspectivesajr_1266 144..149

Ha Hoang, MD, and Quynh Le, PhD

Department of Rural Health,University of Tasmania, Launceston, Tasmania, Australia

AbstractObjectives: This study investigates: (i) Tasmanian ruralwomen’s preferences for different models of intrapar-tum care; (ii) their preferences for travel time to safedelivery; and (iii) factors which influence thesepreferences.Design: Mixed-methods study using a survey question-naire and semistructured interviews was adopted. Aquestionnaire explored women’s preferences for differ-ent models of care and preferred travel time. Interviewswere conducted to validate the survey results andprovide insightful information on their preferences onthe models of care. Women who have had rural child-birth experiences from six Tasmanian rural communi-ties were invited to participate in the study.Results: Two hundred and ten women completed thequestionnaire with a response rate of 35%. Twenty-twofollow-up interviews were conducted. The survey foundthat women preferred to give birth in a hospital settingto homebirth despite having to travel for two hours.Midwifery-led care with one hour travel time was thesecond preferred model of care. Women were willing totravel to access the regional hospital but within limit.Their preferences suggest that women have to trade-offbetween local access and safety considerations. Theinterviews validate the survey results. Three mainthemes emerged from the interview data namely (i)safety; (ii) distance from hospital; and (iii) type of deliv-ery. Their preferences were associated with their mater-nal care experiences in the past.Conclusion: In order to achieve the maternity servicesthat are woman centred and respond to the needs andpreferences of women, the service design and provisionshould take into account these women’s preferences.

KEY WORDS: Australia, models of care, maternitycare, rural area, rural woman.

IntroductionIn Australia, about 32% of the population live in rural(29%) and remote (3%) areas.1 People living in thoseareas encounter a number of health inequities, many ofwhich result from difficulties in accessing health careservices.2 Accessing appropriate maternity services isone of the particular concerning issues in rural andremote communities where over 50% of small ruralmaternity units had closed since 1995.3 There are threemain reasons which led to these closures includingworkforce shortage,4 safety and quality considerations3

and cost considerations.3,5

The recent report of maternity services review2 notedthat the provision of maternity services to rural commu-nities will require trade-offs between access to serviceslocally and considerations of safety and quality. In orderto improve access to maternal services for rural commu-nities, developing collaborative care models whichrespond to community needs and integrate with regionalservice systems is recommended. There is no such thingas a ‘one size fits all’ model of maternity care for allcommunities. Thus, a range of models which allowsflexibility to respond to community needs, priorities andworkforce availability should be considered. Further-more, the National Maternity Service Plan6 states thatmaternity care will be woman centred, reflecting theneeds of each woman within a safe and sustainablequality system, and close to where women live.

Currently, the needs and preferences of women inrural Tasmania on different models of maternity careremain unknown. This paper reports significant insightsof Tasmanian rural women’s preferences for a range ofmodels of intrapartum care and travel time to the hos-pital. The results of this study will inform policy-makersand State Government about the service designs andprovision of maternity services in rural Tasmania.

Methods

Research questions

• Which model of care do women prefer: hospital,midwifery or planned homebirth model of care?

Correspondence: Ha Hoang, University Department of RuralHealth, University of Tasmania, Locked Bag 1372 Launceston,Tasmania 7250, Australia. Email: [email protected]

Accepted for publication 25 February 2012.

bs_bs_banner

Aust. J. Rural Health (2012) 20, 144–149

© 2012 The AuthorsAustralian Journal of Rural Health © National Rural Health Alliance Inc. doi: 10.1111/j.1440-1584.2012.01266.x

• How long would they be willing to travel from theirhome to safe delivery?

• What factors affect their preferences for choosinga model of care and willingness to travel to safedelivery?

This study was part of a wider research project iden-tifying women’s needs in rural areas of Tasmania, Aus-tralia. Ethics approval for the study was granted by theTasmanian Social Sciences Human Research EthicsNetwork. To seek answers to the earlier research ques-tions, the study used mixed-methods approach.7 Thedata collection methods included a survey and semis-tructured interviews. The results from the interviewswere used to validate and interpret the findings from thesurvey.8

Survey

Six rural communities in the north, south, west and eastof Tasmania were chosen to conduct the survey. Thehealth centres in these communities currently providevery limited maternity care services. It takes from45 min to more than two hours from these sites to themajor hospitals. Third parties were approached for par-ticipant recruitment. The selection criteria for partici-pants were:• Being a woman over 18 years of age;• Having had childbirth experiences and• Living in rural areas of Tasmania or Australia at the

time of giving birth.Through those centres, 600 surveys were distributed

to women who met the study criteria. When the surveywas closed, 210 questionnaires were returned (35%response rate).

Interviews

Survey participants were asked to indicate their willing-ness to participate in the interviews on the returned

survey form. The interviews were especially focused onwomen who have had child birth experience within fiveyears. Forty-eight women consented to participate in theinterviews. Saturation was determined by the team ofresearchers. This was achieved after conducting 22interviews as no more new themes and categoriesemerged from the data.

Data analysis

Survey analysis

Data from the questionnaires were coded and enteredinto SPSS version 15.0.9 Descriptive statistics and c2

tests were used to analyse the data. Selected character-istics of the survey respondents are presented inTable 1.

Interview analysis

The qualitative data were analysed using groundedtheory and thematic analysis which require theresearchers to constantly analyse and compare newlygathered information before going back to new par-ticipants.10 For the analysis, QSR – NVivo v9.0 soft-ware11 was used to organise transcripts and codes. Toensure the reliability of the study, another researcher(an independent judge) who was doing research in thesame general field was asked to review the raw data ofthe interviews. The independent judge reviewed verba-tim transcripts of interview files. This researcher alsoindependently coded four interviews of a randomsample of data. The researchers and the independentjudge discussed the coding until agreement wasreached.

Selected characteristics of the participants are pre-sented in Table 2.

What is already known on this subject:• Women in rural areas have poorer access and

outcomes.• There is a need to develop models of care that

respond to community needs as there is nosuch thing as ‘one size fits all’ models ofmaternity care.

• Currently, the needs and preferences ofwomen in rural Tasmania on differentmodels of maternity care remain unknown.

What this study adds:• Despite the risks and costs associated with

travel, this group of women preferred to givebirth in a hospital setting which was per-ceived as a relatively safer environment.

• Women are willing to travel to access tomaternity services but within limit.

• These results provide useful insights forservice design and provision of maternitycare services which should ensure the safetyas well as within accepted travel time.

RURAL CHILDBIRTH: ACCESS VERSUS SAFETY 145

© 2012 The AuthorsAustralian Journal of Rural Health © National Rural Health Alliance Inc.

Results

Survey results

Preferences on different models ofintrapartum care

The survey asked the participants to choose one of themodels of care from the following:• Hospital (conventional) care: Care is provided by the

maternity care team including doctors and mid-wives. The hospital is equipped to cover unexpectedsituations during childbirth. It takes two hours travelto this delivery hospital.

• Midwifery-led care: Midwives are primary carerswho provide care for low-risk women antenatally,during labour and postnatally. Epidural is availableonly in case of complication. Doctors are involvedwhen complications occur. It takes one hour travel toaccess this model of care.

• Planned homebirth: Midwives are primary carers forplanned homebirth women. If complications occur,

women will be transferred from their homes to thehospital. Epidural is available only in case of com-plication. It does not involve any travel to access thismodel of care. The survey of 210 women demon-strated that a small majority number of women(54.4%) chose hospital care as their preferredmodels of care. More than a third (36.8%) ofrespondents preferred midwife-led care. Approxi-mately 9% of women would choose to have theirbabies at home with the assistance of midwives.These results are illustrated in Figure 1.

Views on travel time to safe delivery

In order to inform policy-makers on how women wouldtrade-off between access to services locally and consid-erations of safety and quality, the survey asked womento indicate their willingness to travel for a safe delivery.The results are demonstrated in Figure 2.

Factors that affect their preferences

c2 test indicates that participants’ type of delivery intheir most recent childbirth experience significantly

TABLE 1: Characteristics of survey respondents

CharacteristicsNumber ofparticipants (n)

Percentage(%)

Age group18–21 years 2 1.022–30 years 42 20.131–40 years 77 36.8Over 40 years 88 42.1

EducationPrimary school 2 1.0Secondary school 111 53.1University 67 32.1Other 29 13.9

Language spoken at homeEnglish 203 97.1Other 6 2.9

An Aboriginal or TorresStrait Islander?

No 201 96.2Yes 8 3.8

Most recent babyLess than a year ago 33 15.9One to two years ago 33 15.9Over two to five years ago 51 24.5Over five years ago 91 43.8

Type of deliveryVaginal delivery 128 62.4Caesarean delivery 47 22.9Induced labour 21 10.2Mixed type of delivery 9 4.4

TABLE 2: Characteristics of participants

CharacteristicsNumber ofparticipants (n)

Percentage(%)

Age groups22–30 years of age 4 18.1831–40 years of age 15 68.18Over 40 years of age 3 13.64

Level of educationSecondary school 5 22.73University 11 50.00Other 6 27.27

Language spoken at homeEnglish 22 100.00Other 0 0.00

Torres Strait Island backgroundNo 20 90.91Yes 2 9.09

Most recent baby/babiesLess than a year ago 1 4.55One to two years ago 8 36.36Two to five years ago 13 59.09

The nearest maternity unitLess than an hour’s drive 11 50.00One to two hours’ drive 9 40.91More than two hours’ drive 2 9.09

Delivery typeVaginal delivery 11 50.00%Induced labour 5 22.73%Caesarean delivery 6 27.27%

146 H. HOANG AND Q. LE


affect their preferred model of care (c2 = 14.654,P-value = 0.023 < 0.05). For instance, women who hadcomplicated birth in the past are more likely to choosehospital care as their preferred model of care. Further-more, women’s past experience of the type of healthprofessionals who provided care during pregnancy doesaffect their preference on models of care (c2 = 14.720,P-value = 0.023 < 0.05). Particularly, for women whohad obstetricians as their care providers before tend toprefer to give birth in a hospital. Finally, in terms ofpreferred travel time, women tend to choose the traveltime to the delivery unit which matches their previousexperience (c2 = 98.106, P-value = 0.000 < 0.05).

Interview results

The results from the survey demonstrated that respon-dents most preferred hospital model of care, followed by

midwife-led model of care and homebirth model of care.This is consistent with the interviews. Among 22 par-ticipants, only three participants (13.6%) supportedhomebirth. The rest preferred to give birth in a hospitalsetting. The reasons given by the participants were cat-egorised into three groups namely: (i) safety; (ii) distancefrom hospital; and (iii) type of delivery.

Safety

Safety is the primary reason given by the participants fortheir preferences of hospital model of care. Womenbelieved that being in the hospital with medical staff andready access to equipment is the only way to ensure thesafety of themselves and their newborns.

. . . I would rather be at the hospital in case anythingwent wrong. (Participant 3)

Participants who have witnessed the risk of childbirthpreferred to give birth in a hospital setting where theycan access to medical help timely in case of any unex-pected emergencies.

I would prefer to be in a hospital in case of a scenariorequiring emergency treatment. Having been a supportperson for my sister in labour one time and seeingwhat can happen unexpectedly, her child wouldn’thave survived a homebirth. (Participant 14)

Distance from hospital

Another reason why the majority of women in this studypreferred to give birth in the hospital is their distancefrom the hospital. Because they were living in ruralareas, in case of an emergency, there would be a longdrive to the hospital, and it would not be safe for themother and the baby.

I however would not even consider homebirth becauseif anything was to go wrong the nearest maternity unitis a 2 hour drive away (4 hours return) (Participant 13)

Most participants were aware of unpredictable situa-tions in childbirth and were willing to travel a longdistance to a fully equipped hospital for safe delivery.This strengthens the results of the survey on travel timeto safe delivery.

I guess the deciding factor for my husband and I wasbeing quite isolated and the thought of not being ableto make it to a hospital in case of an emergency wasquite scary for both of us . . . (Participant 18)

Delivery type

The survey results indicated that there was a relation-ship between participants’ type of delivery and the pre-

60.054.4

36.8

8.8

50.0

40.0

30.0

20.0

10.0

0.0

Hospital care Midwife-ledcare

Models of intrapartum care

Perc

en

tag

e (

%)

Plannedhomebirth

FIGURE 1: Women’s preferences on models of care.

70.0

60.059.2

25.2

1.5

12.1

1.9

50.0

40.0

30.0

20.0

10.0

0.0

Less than an hour drive

One to two hours drive

More than two hours drive

Not comcerned about driving time

Other

Travel time to safe delivery

Perc

en

tag

e (

%)

FIGURE 2: Responses on willingness to travel to safedelivery.



ferred model of care. These are confirmed by theinterview results. Interviewed women who have experi-enced complications in childbirth and had caesareandelivery tended to choose hospital care as their preferredmodel of care.

As my birth was an emergency caesarean I am glad Iwas at the hospital. (Participant 15)

Some participants who have experienced unexpectedemergencies in their childbirth strongly believed that it issafer to deliver in the hospital.

I definitely would not have a homebirth in my experi-ence as we thought our twins were healthy and readyto be born, as I had a check-up the day before and younever know what can go wrong as one of my twin’slungs weren’t developed. (Participant 21)

DiscussionThe combined results from the study indicated thatwomen in rural areas considered the safety of them-selves and their newborns of high priority. Despitehaving to travel for two hours, they preferred to givebirth in a hospital setting where childbirth risk iscovered. These findings support the previous studiesshowing that safety is an important factor in women’sdecision on place of birth12,13 and women wanted todeliver in a fully serviced hospital.14,15

The results on the participants’ willingness to travel tosafe delivery seemed to contrast with the findings on thepreferred models of care as over 50% of the respondentssaid that they would be willing to travel for less than anhour drive to a safe delivery unit. These findings suggestthat women want to give birth in an adequatelyequipped hospital, but they have to weigh up this optionwith the time taken to travel to that place and difficultiesassociated with travel. Literature has largely docu-mented that women incur financial costs12,16,17 andencounter stress, fear and anxiety12,16–18 to meet theiraccess needs. These results indicate that women have totrade-off between access to services and considerationsof safety. This is validated by the interview resultsshowing that women did not choose to give birth athome with the assistance of a midwife despite nothaving to travel due to safety reason, their risk statusand distance from their home to the major hospital.Women were willing to travel but within limit to assuretheir safety rather than forgoing their safety giving birthat home. These results provide useful implications forservice design and provision of maternity services. Obvi-ously, further loss of maternity services in rural areasmight not be acceptable to women as it will increase thetravel time. Thus, the health system should design and

provide services which are within acceptable and pre-ferred distance as well as ensure the safety in terms ofpreferred models of care.

Findings about factors that influenced the women’spreferences on the models of care and travel time pre-sented interesting results. Our study supports the claimthat rural women formed their views through experi-ence and framed them realistically within current serviceprovision and their rurality context.14 Women tend tochoose the system of care which matches their experi-ence in the past.14,19,20 It suggests that either women arevery discerning or all the systems work for the patientswho use them. Moreover, due to the design of the study,the sample did not include nulliparous and primiparouswomen. Without an insight from these two groups, thehypothesis remains that women who have deliveredbecome more concerned about safety than women whohave not. An alternative hypothesis is that rural womentend to be older, and older women who have alreadygiven birth are older again than their nulliparous urbancounterparts and that we are witnessing an age-associated concern for security. Therefore, future inves-tigations should address to these hypotheses.

ConclusionThe results of the study suggest that in order to achievethe maternity care system which is woman centred andmeets their needs and preferences,6 service design andprovision of maternity care should take into accountwomen’s perspective shown in these findings about theirpreferences. This study provides an evidence base whichcan be built on in order to make more definitive recom-mendations about health service design. A further studywith a broader and larger target group including nul-liparous and primiparous women would be the next steptogether with a greater review of the literature on safetyof travel (driving) during pregnancy and labour in ruralareas.

AcknowledgementWe would like to thank the University of Tasmania forproviding funding under the Australian PostgraduateScholarship scheme.

References1 Australian Institute of Health and Welfare. Australia’s

Health 2008. Canberra: AIHW, 2008.2 Department of Health and Ageing. Improving maternity

services in Australia: report of the Maternity ServicesReview: Australian Government Department of Healthand Ageing. 2009.

148 H. HOANG AND Q. LE


3 Rural Doctors Association of Australia. Maternity servicesfor rural Australia. 2006.

4 NASOG. Submission to Maternity Services Review.National Association of Specialist Obstetricians and Gynae-cologists. 2008. [Cited 25 Mar 2011]. Available fromURL: http://www.health.gov.au/internet/main/publishing.nsf/Content/maternityservicesreview-355.

5 Klein M, Christilaw J, Johnston S. Loss of maternity care:the cascade of unforeseen dangers. CJRM 2002; 7: 120–121.

6 The Australian Health Ministers’ Conference. NationalMaternity Services Plan. Canberra. 2011.

7 Tashakkori A, Teddlie CE. Handbook of Mixed Methodsin the Social and Behavioural Science. Thousand Oaks,CA: Sage, 2003.

8 Greene JC, Caracelli VJ, Graham WF. Toward a concep-tual framework for mixed-method evaluation design. Edu-cational Evaluation and Policy Analysis 1989; 11: 255–274.

9 SPSS 15.0. Command Syntax Reference Chicago: SPSS Inc.2006.

10 Strauss A, Corbin J. Basics of Qualitative Research: Tech-niques and Procedures for Developing Grounded Theory.Newbury Park, CA: Sage, 1990.

11 QRS International Pty Ltd. Nvivo 9. Melbourne QRSInternational Pty Ltd. 2011.

12 Kornelsen J, Grzybowski S. Safety and community: thematernity care needs of rural parturient women. Journalof Obstetrics and Gynaecology Canada 2005; 27: 554–561.

13 Smith M, Askew DA. Choosing childbirth provider loca-tion – rural women’s perspective. Rural and RemoteHealth 2006; 6 (3): 1–9.

14 Pitchforth E, Watson V, Tucker J et al. Models of intrapar-tum care and women’s trade-offs in remote and rural Scot-land: a mixed-methods study. BJOG 2008; 115: 560–569.

15 Simth K, Humphreys J, Lenard Y, Jones J, Prince V, SooHan G. Still the doctor – by a country mile! Preferences forhealth services in two country towns in north-west NewSouth Wales. The Medical Journal of Australia 2004; 181:91–95.

16 Roach S, Downes S. Caring for Australia’s most remotecommunities: obstetric services in the Indian OceanTerritories. Rural and Remote Health (Online), 2007;7(699). [Cited 19 May 2010]. Available from URL: http://wwwrrhorgau..

17 Dietsch E, Shackleton P, Davies C, Alston M, McLeod M.‘Mind you, there’s no anaesthetist on the road’: women’sexperiences of labouring en route. Rural and RemoteHealth 2010; 10 (1371): 1–9.

18 Kornelsen J, Stoll K, Grzybowski S. Stress and anxietyassociated with lack of access to maternity services forrural parturient women. Australian Journal of RuralHealth 2011; 19: 9–14.

19 Hundley V, Ryan M. Are women’s expectations and pref-erences for intrapartum care affected by the model-of-careon offer? BJOG 2004; 111: 550–560.

20 Salkeld G, Ryan M, Short L. The veil of experience: doconsumers prefer what they know best? Health Economics2000; 9: 267–270.



trade-off between local access and safety considerations in childbirth: rural tasmanian women's...

Documents