Methods
All women attending Liverpool Women’s Hospital for antenatal care between January and December 2018 received information about the study in the form of a participant information sheet. These were distributed in antenatal clinics, community and hospital. Posters and leaflets informing women about the study were also displayed in antenatal clinics. Contact numbers for researchers and opportunities to discuss the study with them were given to all women. Women at known high risk of PPH, such as women with twin pregnancies, were approached for written informed consent by a researcher in the antenatal period. If a woman indicated that she did not want to be approached further, a sticker was placed in her notes to alert clinicians to this. See figure 1 for the recruitment process.
Obstetricians trained in use of the PPH Butterfly device were also trained in gaining consent for research in an emergency situation. This was provided as part of the overall intervention training process by a researcher (WT). In the event that a woman experienced a PPH, the attending clinical team would administer first aid to her, gaining intravenous access, taking blood for grouping, and administering oxytocin. During these initial procedures, the recruiting obstetrician would also discuss the use of the device and gain verbal consent to use the PPH Butterfly device should the bleeding continue. A script was provided to the recruiting physicians and was signed by them after recruitment (see Figure 2). The obstetricians were instructed to only use the device if the woman was certain that she wanted to continue; if she declined, appeared hesitant, uncertain, or confused then normal care would continue without use of the device. The time taken to discuss consent was typically 1-2 minutes. In those whose bleeding continued despite the initial measures and who provided verbal consent, the device was inserted, and uterine compression commenced alongside normal care. The clinician was instructed to remove the device if at any point the participant requested.
Following management of the situation and after sufficient recovery time, the woman was approached by a trained research midwife to discuss the study and obtain written consent. Women consenting to the study were then asked to complete a questionnaire, which included a Likert scale question on the acceptability of the consent process. Data were cleaned and entered into SPSS; analysis provided descriptive statistics of the responses to this question.
The qualitative aspect of the study explored the consent process through the views of the women participants, their birth partners, midwives and obstetricians. The qualitative approach was underpinned by Straussian grounded theory methodology.6 Inclusion criteria were that the participant was present at the birth and during the use of the PPH Butterfly device. Women included in the main study were provided with a participant information sheet by the midwife obtaining retrospective consent. If women were agreeable, a different research midwife (WT) contacted them by their preferred method, after a time of at least 24 hours had elapsed. Interviews were completed at a time of the woman’s choosing, usually at her home address, following informed written consent. If a woman consented to interview, then a further participant information sheet was provided to their birth partner and a similar process followed. Obstetricians and midwives present when the device was used were provided with participant information sheets. If they consented, face-to-face interviews were conducted within a private environment in the hospital setting. The sample size for all groups was guided by data saturation;7 that is when no new themes emerged from the data.
Interviews with all groups were semi-structured, with a topic guide enabling focus to be maintained during the interview. Data collection was completed in the three weeks following birth, in order to minimise recall bias. Data were analysed using a framework analysis approach.8 This approach is useful for facilitating the constant comparative techniques required in grounded theory.9 Data collection ceased when data saturation was achieved as agreed by the two researchers completing analysis (WT, CB). At this point data was becoming repetitive, with no new themes emerging. Data were collected between January and November 2018.