Satisfaction and the three C's: continuity, choice and control. Women's views from a randomised controlled trial of midwife-led care

Authors


Correspondence: Ms V. A. Hundley, Aberdeen Maternity Hospital, Cornhill Road, Aberdeen AB9 2ZA, UK.

Abstract

Objective 1. To explore whether there are differences in women's satisfaction with care in a midwife-managed delivery unit compared with that in a consultant-led labour ward. 2. To compare factors relating to continuity, choice and control between the two randomised groups.

Design A pragmatic randomised controlled trial.

Setting Aberdeen Maternity Hospital, Grampian.

Sample 2844 women, identified at booking as low risk, were randomised in a 2:1 ratio between the midwives’ unit and the labour ward.

Main outcome measures Satisfaction, continuity of carer, choice, and control.

Results Satisfaction with the overall experience did not differ between the groups. Satisfaction with how labour and delivery was managed by staff was slightly higher in the midwives’ unit group, but this did not reach the 0.1% level of significance. Women allocated to the midwives’ unit group saw significantly fewer medical staff and were less likely to report numerous individuals entering the room. They were more likely to report having had a choice regarding mobility and alternative positions for delivery and were significantly more likely to have made their own decisions regarding pain relief.

Conclusions The issues surrounding the measurement of satisfaction with childbirth need further investigation. Until this area is clarified it would be unwise to use an overall measure of satisfaction as an indicator of the quality of maternity service provision. In particular, the current measures are not sensitive enough to examine the specific factors which affect women's satisfaction. Further research is required to assess which factors are important to women if they are to have a positive experience of childbirth and how these priorities change over time.

INTRODUCTION

The three C's, continuity, choice and control, have been identified as important aspects of maternity care1–3. Recommendations for improvements in continuity of care have come from both government1,2,4 and professional bodies5–7 with continuity of carer (care by one individual) being recognised as the gold standard. In addition, there is a commitment to providing women with greater choice and control in childbirth2,7–9. The fact that women themselves want these aspects of care has been well documented10–12.

In Scotland various systems of providing maternity care have been set up in response to government recommendations for improved continuity and choice13. One such system operates in the midwives’ unit in Aberdeen. The unit consists of five single rooms in a separate unit located 20 yards from the consultant-led labour ward. The philosophy of care behind the unit is to provide a safe,‘homely’ environment where women can retain choice and control in the management of their labours14. Between January 1992 and August 1993 the midwives’ unit was the site of a large randomised controlled trial of midwife-managed care. The study compared intrapartum care and delivery of low risk women in a midwife-managed delivery unit with that in a consultant-led labour ward. The morbidity results confirmed that low risk women allocated to midwife-managed care had a lower rate of intervention than women allocated to standard consultant-led care14. The findings suggest that care in a midwife-managed delivery unit is as safe as care in a consultant-led labour ward14. Midwife-managed care has also been shown to increase continuity of carer and, as a result, the satisfaction of the midwives15. This paper is concerned with data on women's satisfaction, continuity of carer, choice and control.

METHODS

Study aims

The main objective was to compare women's satisfaction with care and delivery in a midwife-managed delivery unit with that in a consultant-led labour ward. A secondary objective was to compare factors relating to continuity, choice and control, between the two randomised groups of women.

Study population

Low risk women were identified by the researcher from general practioner referral letters. The selection criteria for the study were those established to identify low risk women for booking for delivery in General Practitioner Units in Grampian. Low risk women who requested domino deliveries were excluded from the study because they were a self selecting group who had care given by one midwife throughout the antenatal and intrapartum period. A total of 2844 low risk women were recruited over a 14 month period between October 1991 and December 1992 (Fig. 1). Recruitment was voluntary and by informed consent. Furthermore, women were assured that they could withdraw from the study at any time and that this would not affect their care. All women had delivered by August 1993. The study received ethical approval from the joint ethical committee of Grampian Health Board and the University of Aberdeen.

Figure 1.

Trial recruitment. Values are given as n and n (%).

Trial design

This was a pragmatic randomised controlled trial. At booking, women identified as low risk were randomised to deliver in either the midwives’ unit or the labour ward by opening consecutive sealed opaque envelopes. An initial allocation of 2:1 in favour of the midwives’ unit was used because of the expected transfer of women with complications from the midwives’ unit to the labour ward. This ratio was necessary to ensure that the space in the midwives’ unit was fully utilised. The antenatal care of all women participating in the study was otherwise identical to that received by other women booking at Aberdeen Maternity Hospital. The full details of the study design have been reported previously14.

Data collected

The majority of the data presented in this paper were collected from a questionnaire given to women participating in the study, with the exception of those women who had suffered a stillbirth or neonatal death, on discharge from the hospital. This contained two questions about satisfaction. One question asked women how they felt their labour and delivery was managed by staff. The other asked women to grade their overall satisfaction with the experience on an ordinal scale from 0 to 10, with 0 being “Thoroughly unsatisfactory. Nothing good to be said about it” and 10 being “An absolutely wonderful experience that could not have been better”. Other questions related to pregnancy and antenatal care; outcomes such as mode of delivery; staff involved and support given during labour and delivery; interventions in labour; levels of pain and pain relief; continuity of carer, choice and control; information about infant feeding and whether mother and baby were separated; and demographic details. Women who did not respond by three weeks after delivery were sent a second copy of the questionnaire. Those women not returning their questionnaire by six weeks after delivery were reminded by telephone.

Data were also collected from staff questionnaires completed by the midwife present at the delivery for each woman in the study. These questionnaires provided details about the midwifery staff, their role, experience and satisfaction.

Analysis and statistical methods

Data were analysed using the statistical package SPSS16 and analysis was by intention-to-treat. That is, all women were analysed in the group to which they were initially allocated, whether or not they completed or received, that care. This pragmatic method of analysis permits unbiased estimates of the performance of the midwives’ unit under normal clinical conditions which include transfer to the labour ward both before and during labour. This method of analysis avoids misinterpretation of the data17.

Categorical variables were analysed using χ2 tests and continuous variables with a normal distribution by Student's t test. Data which did not follow a normal distribution were analysed using the Mann-Whitney U test. Significance levels and confidence intervals, where appropriate, are quoted in the tables.

Due to the large number of variables to be tested and to keep the risk of type I error small, the P value which was considered to be statistically significant was reduced from the conventional 5% to 0.1% using the Bonferroni correction. The Bonferroni correction takes account of the number of variables to be examined and alters the P value accordingly18.

RESULTS

Although 2844 women were recruited, questionnaires were sent to only 2578 women (Fig. 2). The remaining women were lost to follow up (n= 106); had requested an alternative location for delivery (n= 105); were omitted due to staff error (n= 9); had suffered neonatal death or stillbirth (n= 21); or it was considered inappropriate that they complete the questionnaire (9 women had twins and 16 delivered before arrival at hospital). Of those questionnaires sent out, 2463 were returned giving a 95% response rate. Twenty-two percent of women (n= 556) required reminders. Table 1 shows the baseline characteristics of the women participating in the study.

Figure 2.

Response rates to postal questionnaire. Values are given as n and n (%).

Table 1.  Demographic data. Values are given as n, n (%) or mean [SD]. (Source: Hundley V, Cruickshank F, Lang G et al. Midwife managed delivery unit: a randomised controlled comparison with consultant led care. BMJ 1994; 309: 1400–1404. Reproduced with kind permission of the BMJ Publishing Group.)
 Midwives' unitLabour ward
Age at delivery (years)28 [4.4]28 [4.5]
 No.1675789
Maternal height (cm)163 [5.8]163 [5.9]
 No.1674793
Parity  
 Primiparous929 (56)451 (57)
 Multiparous745 (44)338 (43)
Social class  
 I190 (12.0)97 (12.2)
 II317 (20.0)141 (17.8)
 Id nonmanual165 (10.4)91 (11.5)
 HI manual453 (28.6)44 (30.8)
 rv377 (23.8)173 (21.8)
 V80 (5.1)47 (5.9)
Education: age on leaving  
full–time education (years)  
 Woman17.5 [2.5]17.4 [2.4]
 No.1640772
 Partner/husband17.7 [2.9]17.5 [3.0]
 No.1552737

Satisfaction

Table 2 shows women's satisfaction with how labour and delivery was managed and their satisfaction with the overall experience. More women allocated to the midwives’ unit group expressed approval about how their labour and delivery were managed by staff, although this did not reach the 0.1% level of significance. There was no difference in satisfaction with the overall experience.

Table 2.  Satisfaction. Values are given as n, n (%) and median {interquartile range}.
 Midwives' unitLabour wardP
Thinking back now about what happened to you and what the staff did, do you feel your labour and delivery was managed by the staff16547680.02
 As you liked it in every way1291 (78.1)564 (73.4) 
 As you liked it in some ways but not in others342 (20.7)193 (25.1) 
 Not as you liked it at all21 (1.3)11 (1.4) 
Satisfaction with overall experience8.0 {7–9}8.0 {7–9}0.1

Continuity

Women allocated to the midwives’ unit group had greater continuity of carer, as reported by the midwife at delivery, both during labour, at delivery and after delivery15.

The number of midwives caring for the woman while she was in the delivery unit was the same in both groups (Table 3). However, women allocated to the labour ward group reported seeing significantly more members of medical staff during labour and delivery. A small number of women in both groups reported that they were unsure who they saw.

Table 3.  Continuity of carer as recorded in the women's questionnaire. Values are given as n, n (%) and median {interquartile range}.
 Midwives' unitLabour wardP% difference (95% CI)
Approximately how many midwives (trained) looked    
after you while you were in the delivery unit?1618762  
 2.0 {1–3}2.0 {1–2}0.03 
Staff that women reported seeing during either labour or delivery1665784  
 Midwife1616 (97.1)760 (96.9)1.00.2 (-1.3 to 1.6)
 Hospital doctor624 (37.5)354 (45.2)< 0.001-7.7 (-ll.9 to -3.5)
 Student midwife1033 (62.0)378 (48.3)< 0.00113.8 (9.6 to 18.0)
 Student nurse204 (12.3)100 (12.8)0.8-0.5 (-3.3 to 2.3)
 Medical student169 (10.2)149 (19.0)< 0.001-8.8 (-12.0 to -5.7)
 Paediatrician382 (22.9)204 (26.0)0.1-3.1 (-6.8 to 0.6)
 Anaesthetist284 (17.1)172 (21.9)0.004-4.8 (-8.3 to -1.5)
 Other17 (1.0)25 (3.2)< 0.001-2.2 (-3.5 to -0.8)
 Don't know65 (3.9)40 (5.1)0.2-1.2 (-3.0 to 0.6)

The majority of women in both groups had their chosen companion with them during labour and delivery (Table 4). Where the chosen companion was present, they were most likely to be with them for all or most of the time. More women allocated to the labour ward group reported feeling that there were a lot of people coming in and out of the room while they were in labour. However, this did not reach our set level of statistical significance and the majority of women in both groups said that they were happy with the number of staff around during labour/delivery. A small proportion of women in both groups reported that they were left alone either during labour or after delivery when it worried them to be left alone.

Table 4.  Women's views of support during labour and delivery. Values are given as n or n (%).
 Midwives' unitLabour wardP% difference (95% CI)
Chosen companion present during labour/delivery?16687860.40.9 (−0.8 to 2.6)
 Yes1611 (96.6)752 (95.7)  
Present16087490.80.2 (-1.4 to 1.8)
 For all or most of the time1554 (96.6)722 (96.4)  
 For some of the time only54 (3.4)27 (3.6)  
Did you feel there were lots of different people coming    
in and out of the room while you were in labour?16597740.003 
 Yes: a lot50 (3.0)33 (4.3)  
 Yes: quite a few271 (16.3)163 (21.1)  
 No: hardly any1338 (80.7)578 (74.7)  
During your labour/delivery did you feel you had16637800.2 
 Too many staff around43 (2.6)16 (2.1)  
 Too few staff around29 (1.7)22 (2.8)  
 Were happy with the number1591 (95.7)742 (95.1)  
Were you and your companions left alone by the staff at any    
stage when it worried you to be left alone?16747890.0034.1 (1.3 to 6.9)
 No: neither1514 (90.4)681 (86.3)  

Choice

Few women in either group said that they had been given a choice as to the way their baby's heartbeat was monitored (Table 5). Women allocated to the labour ward group were more likely than women allocated to the midwives’ unit group to report that they had been given a choice. However, the majority of women who had electronic monitoring, either intermittently or continuously, were happy with this (midwives’ unit 89%, labour ward 91%). In contrast, the findings of the midwives’ questionnaire suggested that the majority of women in both groups were involved in making management decisions about labour (Table 6).

Table 5.  Choice. Values are given as n or n (%). SVD = spontaneous vaginal delivery.
 Midwives' unitLabour wardP% difference (95% CI)
Were you given any choice as to the way    
 your baby's heartbeat was monitored?14297410.002-3.7 (-6.2 to -l.2)
 Yes88 (6.2)73 (9.9)  
During your labour, did you feel you wanted to move around or change position16327570.004 
Yes937 (57.4)380 (50.2)  
 No, not really675 (41.4)365 (48.2)  
 Don't know20 (1.2)12 (1.6)  
Where women wanted to move, were they able to?9373800.0077.9 (2.2 to 13.5)
 Able to move most of time663 (70.7)239 (62.8)  
 Unable to move273 (29.1)140 (36.8)  
Did the hospital staff encourage you to move around and change position?1544722< 0.0018.4 (4.4 to 12.5)
 Yes1163 (75.3)483 (66.9)  
Where the woman had a SVD, would they have liked to have tried another position for delivery13385790.3 
 Yes, definitely35 (2.6)21 (3.6)  
 Yes possibly115 (8.6)54 (9.3)  
 No, not really1057 (79.0)460 (79.4)  
 Don't know131 (9.8)44 (7.6)  
Did you have any particular preferences about what happened in the third stage?14706730.6 
 Yes108 (7.3)42 (6.2)  
 No1206 (82.0)555 (82.5)  
 Don't know enough about it156 (10.6)76 (11.3)  
If you had a preference for third stage, did you get what you wanted?108420.11-13 (-28 to l)
 Yes73 (68)34 (81)  
Table 6.  Control. Values are given as n or n (%).
 Midwives' unitLabour wardP% difference (95% CI)
Labour management decisions16638090.21.7 (−0.7 to 4.1)
 Woman involved in the decision making1535 (92.3)733 (90.6)  
 Decisions made by staff only128 (7.7)76 (9.4)  
Did you have any say in whether your waters were broken?8323910.8 
 It was discussed fully and my approval was sought543 (65.3)263 (67.3)  
 It was discussed but the staff made the final decision165 (19.8)69 (17.6)  
 It was never discussed, the staffjust got on with it120 (14.4)57 (14.6)  
Waters broke during vaginal examination - accident4 (0.5)2 (0.5)  
 How was the decision made about the type of pain relief to use?1616765< 0.001 
 I made my own decision with the staff's approval880 (54.5)375 (49.0)  
 I made my own decision against the staff's advice14 (0.9)9 (1.2)  
 I was happy to follow the staff's advice on the matter594 (36.8)275 (35.9)  
 The staff were insistent I take their advice and    
 I didn't feel I could refuse9 (0.6)11 (1.4)  
It all just happened and there was no decision made as such107 (6.6)89 (11.6)  
 Other12 (0.8)6 (0.8)  

Significantly more women allocated to the midwives’ unit reported that they were able to move around and change position during labour (midwives’ unit 63.5%, labour ward 51.6%; P < 0.001). However, there was no significant difference in the number of women who both wanted to and were able to move around (Table 5). Women allocated to the midwives’ unit group were also more likely to report that staff encouraged them to move around and change position (midwives’ unit 75.3%, labour ward 66.9%; P < 0.001). The most common reason in both groups for restricted mobility was that the woman was attached to a monitor, drip or epidural infusion (midwives’ unit 22.2%, labour ward 30.3%; P < 0.001). A small number of women (n= 34) allocated to the midwives’ unit group reported that they were unable to move because they were told to keep still. There were no reports of this among women allocated to the labour ward group.

Of the women who had a normal delivery, the majority in both groups delivered on their back in bed propped up with pillows. Women allocated to the midwives’ unit group were more likely to have tried an alternative position (midwives’ unit 14.3%, labour ward 8.7%; P < 0.01), but the majority of women in both groups were happy with the position they were in for delivery.

Few women in either group had any particular preferences about what happened in the third stage of labour. Women who did have a preference about what happened were more likely to get their preference if they were allocated to the labour ward group.

Control

Women in both groups were equally likely to be involved in making management decisions about labour and artificial rupture of membranes (Table 6). However, significantly more women allocated to the midwives’ unit group made their own decision about the type of pain relief to use. Women allocated to the labour ward group were more likely to say that there was no decision made as such.

DISCUSSION

This study has tried to measure women's satisfaction in a midwife-managed delivery unit compared with that in a consultant-led labour ward. Issues relating to continuity, choice and control have also been examined.

Assessing satisfaction, particularly satisfaction with childbirth, is an extremely difficult task fraught with pitfalls19. It has become common practice in surveys, particularly of maternity care, to use an ordinal scale of measurement to allow a woman to grade her overall satisfaction with the experience. However, reducing women's perceptions of their care and childbirth experience to a single dimension of ‘satisfaction’ is often meaningless because of their complex nature. Clearly, satisfaction is determined by a wide variety of factors, such as previous experience and social status, as well as by the time of administration of the questionnaire. Thus, problems can arise from the lack of specificity of the ordinal scale which make interpreting actual satisfaction difficult. In particular, individuals may report general satisfaction with their care or experience while at the same time having specific dissatisfactions, such as with waiting times or starring levels. Michie and Kidd20 cite an example of this where an audit of day surgery found that one week after the operation 80% of patients reported overall satisfaction. However, further questioning revealed that 47% of patients were dissatisfied with the facilities and 23% were dissatisfied with the information they had received. Measuring satisfaction with childbirth is further complicated by the fact that the delivery of a healthy child may overshadow other considerations21.

In our study we found that 80% of respondents in both groups gave a satisfaction score of geqslant R: gt-or-equal, slanted 7 out of 10. However, it would be unwise to use this finding to plan service provision because differences between the groups did exist. It is probable that dissatisfaction with specific areas of service provision also existed. Therefore, we cannot be complacent with the finding that 80% of women were highly satisfied with the service in general.

The answer may be to look more closely at factors which can affect satisfaction, for example; continuity, choice and control. However, these findings may not always provide a clear answer and in some cases may appear to conflict with each other. For example; nearly one fifth of women in both groups answered yes to the question “Did you feel there were lots of different people coming in and out of the room while you were in labour?” Yet when asked how they felt about the number of staff in the room during labour and at delivery, the majority of women in both groups said they were “happy with the number”. One explanation for this could be that at least 80% of respondents will express satisfaction for any given question22. This phenomenon, it is argued, occurs because of a reluctance of patients in the National Health Service to express critical comments about their health care23. Furthermore, we must be hesitant about the findings of this study in relation to factors such as continuity, choice and control since it was not the primary purpose of the study to examine these.

A further problem in assessing satisfaction is the question of the optimum time to seek opinions, particularly if these opinions are to influence service provision. Studies have shown that women are more likely to express negative feelings about care when asked some time after the event24,25. Erb et al. suggest that expression of negative feelings about the birth experience are more common seven to twelve months after birth than in the first six months24. If the findings of surveys of maternity care, carried out within the first six months of birth, are used to plan service provision then it is likely that some problems with the service will not be addressed.

The study is limited in a number of ways. The aim was to measure women's satisfaction with their experience in the two arms of the trial. Although a difference was found in satisfaction between the two groups, this did not reach statistical significance. The finding that differences do exist for some factors which could affect satisfaction may suggest that the scale used to measure satisfaction was too crude. On the other hand, it could be argued that the small number of women who were excluded or who did not complete the questionnaire might significantly have affected the findings. This is particularly important with regard to satisfaction where it is possible that the least satisfied women may not have returned the questionnaire (Table 2). If we assume that women who were excluded or who did not complete the questionnaire were dissatisfied (i.e. responded that their care was “Not as you liked it at all”) then the increased satisfaction among women in the midwives’ unit group is significant (P= 0.0000). However, if we assume that women who were excluded or who did not complete the questionnaire were satisfied in every way (i.e. responded that their care was “As you liked it in every way”) then there is no significant difference between the groups (P= 0.14). The study is also limited by the timing of the questionnaire. Previous studies indicate that the questionnaire was probably sent to women too soon after delivery and as a result more negative feelings about care were not detected. However, this is likely to have affected both groups equally. Finally, caution is required in interpreting the findings of the secondary analysis carried out to explore some factors which have been identified as important to good maternity care since this was not one of the original aims of the study.

CONCLUSION

There are issues surrounding the measurement of satisfaction with childbirth which need further investigation. Until these issues are clarified it would be unwise to use an overall measure of satisfaction as an indicator of the quality of maternity service provision. Further research is required to assess what factors are important to women if they are to have a positive childbirth experience and how these priorities change over time.

Acknowledgements

The study was funded by the Scottish Office Department of Health (SODH). However, the opinions expressed in this paper are those of the authors and not SODH. The authors would like to thank midwifery, medical and clerical staff for their support and assistance; Professor I. Russell and Professor A. Templeton for consultation; Mr J. Lemon for computer advice; and in particular all the women who participated in the study.

Ancillary