Effecting change in midwives' waterbirth practice behaviours on labour ward: an action research study.

Introduction: the use of water immersion for labour and birth has been shown to be beneficial for women in normal labour (Cluett et al, 2009). It was decided to use problem solving coordinator workshops to change in the way waterbirth practice was promoted and organised on labour ward. Findings from the first Action Research phase (Russell, 2011) led to the development of a waterbirth questionnaire to measure midwives' personal knowledge of waterbirth practice, waterbirth self-efficacy, social support and frequency of hydrotherapy and waterbirth practice. The aim of this paper is to share the questionnaire findings from an on-going action research study.

Method: prior to the first workshop 62 questionnaires were distributed to midwives (Bands 5, 6 and 7) working on labour ward. Subsequent questionnaires (n=53) were sent to Bands 5/6 midwives not involved in the workshops, at four (Group 2) and eight months (Group 3). N.B only Bands 5/6 midwives completed post workshop questionnaires. In total 169 questionnaires were distributed. One-way ANOVA with Tukey post-hoc test and the χ(2) test were used to determine statistical significance.

Findings: 96 questionnaires were returned (57%). Midwives' personal knowledge of waterbirth practice differed significantly between groups, (F2,85=3.67, p<0.05) with midwives in Group 1 giving significantly higher scores (X¯=45.6, 95% CI [43.0, 48.2]), than those in Group 3, (X¯=41.7, 95% CI [40.0, 43.3]), p<0.05. Midwives' waterbirth self-efficacy did not differ significantly between groups (F2,88=3.15, p>0.05). However scores for social support did differ (F2,75=4.011, p=0.022), with midwives in Group 1 giving significantly lower scores (X¯=8.0, 95% CI [6.4, 9.5]) than those in Group 3 (X¯=10.5, 95% CI [9.4, 11.6]), p=0.016. Fifty-five per cent of Group 1 midwives facilitated a waterbirth in the previous three months compared with 87% in Group 3. Changes in the frequency of waterbirth for these groups were statistically significant (x(2)=4.369, p<0.05, df=1).

Conclusions: it appears that the co-ordinators were able to influence waterbirth practice because of changes in social support and frequency of waterbirth practice. Given the widespread and continued impact of the intervention, on midwives who attended workshops and those that did not, we feel it likely that a significant proportion of this change could be attributed to the introduction of problem solving waterbirth workshop. The findings from this study suggest that problem solving waterbirth workshops based on an action research format have the potential to normalise midwifery care within medically dominated hospital birthing environments.