'The STORK Study' : the Scottish trial of refer or keep : Midwives' Intrapartum Decision Making

Abstract

The STORK Study Abstract Aim: This thesis describes a study which aimed to explore whether midwives decision making during labour care was affected by the midwives’ own attitude to risk. Specifically, whether midwives who scored highly on risk tendency would delay making a referral for medical assistance compared to those who scored lower. A secondary aim was to explore whether years of clinical experience or location had an effect on midwives’ decision to refer. Project Outline/Methodology: Research Questions 1. Do midwives vary in their general risk propensity, as assessed by scores on a standardised measure of risk propensity? 2. Are midwives risk propensity scores related to their decisions when to seek medical assistance or transfer women to medical care during labour (transfer decisions)? 3. Are ‘transfer’ decisions related to the experience of the midwife or the type of maternity unit in which she practices?

Design A correlation study examined the relationship between midwives personal risk tendency and the timing of their decision to make referrals in a series of fictitious case scenarios. Setting and Sample Four Scottish Health Board areas with one or more Consultant Led Units (CLU) and at least one associated Community Midwifery Unit (CMU). Sample - 100 midwives providing labour care. Permission for access and ethics Ethical approval was granted by the Research Ethics Committee, Department of Nursing and Midwifery, University of Stirling and the NHS Research Ethics Committee - Ref No. 05/S1401/44. Research & Development management approval was granted from each area. Permission for access was granted by the head of midwifery of each participating site. Data collection This was an internet based study using vignettes and questionnaires. Five vignettes were developed which represented a range of labour care scenarios. Each contained snap-shot information about five time points describing a worsening case history e.g. high blood pressure or fetal distress. Participants were asked to review each of the vignettes and decide at which time point they would refer the woman for medical care. For each case midwives could decide not to refer at all. Midwives also completed a questionnaire comprising: social and demographic information, two validated measures of risk attitude and a personality assessment. Key Results Despite being presented with the same information midwives made a range of referral decisions. There was no correlation between referral scores and measures of risk, personality or years of experience. No statistically significant difference between the referral scores of midwives working in CLUs or CMUs was found. However, a significant difference did emerge between the health board areas, with midwives from one area making referrals at a significantly earlier stage. It is interesting that maternity services in this area had experienced several high profile adverse events prior to this study; possibly impacting on the midwives’ timing of referrals. Conclusions The range of referral decisions was not due to risk propensity, personality factors, experience or location. Local factors may influence individual decision making choices. What does this study add to the field? This study contributes to the understanding of midwives’ decision making during intrapartum care. The study also involved the development of an innovative internet based study design which will be useful for other research studies.

Implications for Practice or Policy The study questions assumptions about midwives’ decision making being influenced by personality, place of work or length of service and highlights the range of decisions made by midwives when presented with the same case factors. Where to next? Further study is required to explore factors which may explain the variability of midwives’ decisions to refer. These factors may include individual differences for example, tolerance of ambiguity, the nature of past experience or individual thresholds for acceptable risk.