The feasibility and acceptability of using the Mother-Generated Index (MGI) as a Patient Reported Outcome Measure in a randomised controlled trial of maternity care

Abstract

Abstract
Background: Using patient-reported outcome measures (PROMs) to assess Quality of Life (QoL) is well established, but commonly-used PROM item-sets do not necessarily capture what all respondents consider important. Measuring complex constructs is particularly difficult in randomised controlled trials (RCTs). The Mother-Generated Index (MGI) is a validated antenatal and postnatal QoL instrument in which the variables and scores are completely respondent-driven. This paper reports on the feasibility and acceptability of the MGI in an RCT, and compares the resulting variables and QoL scores with more commonly used instruments.
Methods:
The single-page MGI was included at the end of a ten page questionnaire pack and posted to the RCT
participants at baseline (28–32 weeks’ gestation) and follow-up (six weeks postnatal). Feasibility and acceptability
were assessed by ease of administration, data entry and completion rates. Variables cited by women were analysed
thematically. MGI QoL scores were compared with outcomes from the EQ-5D-3 L; Edinburgh Postnatal Depression Scale; Satisfaction With Life Scale; and State Trait Anxiety Inventory.
Results:
Six hundred and seventy eight pregnant women returned the pack at baseline; 668 completed the MGI (98.5 %); 383/400 returns at follow up included a completed MGI (95.7 %). Quantitative data were scanned into SPSS using a standard data scanning system, and were largely error-free; qualitative data were entered manually. The variables recorded by participants on the MGI forms incorporated many of those in the comparison instruments, and other outcomes commonly used in intrapartum trials, but they also revealed a wider range of issues affecting their
quality of life. These included financial and work-related worries; moving house; and concerns over family illness and pets. The MGI scores demonstrated low-to-moderate correlation with other tools (all r values p < .01).
Conclusions:
Without face-to-face explanation and at the end of a long questionnaire, the MGI was feasible to use, and acceptable to RCT participants. It allowed individual participants to include issues that were important to them,
but which are not well captured by existing tools. The MGI unites the explanatory power of qualitative research with
the comparative power of quantitative designs, is inexpensive to administer, and requires minimal linguistic and conceptual translation.