JMIR Publications

ABSTRACT

Background:

Typical measures of maternity performance are overly focused on the technical elements of birth, especially pathological elements, with insufficient measurement of non-technical measures and those prior to and post-partum. New technologies allow Patient Reported Outcome Measures (PROMs) and Patient Reported Experience Measures (PREMs) to be collected from large samples at multiple time points, and considered alongside existing administrative sources, but such models are not widely implemented or evaluated. In Tuscany, Italy, longitudinal, personalized and integrated user-reported data collection for the maternal care pathway has been in operation since 2018. This model has been through two methodological iterations.

Objective:

To compare the two sampling models of longitudinal user-reported data for the maternal care pathway, considering: participation and attrition rates, population factors influencing attrition, costs, and the strengths and weaknesses of the two data collection models for different stakeholders.

Methods:

Data were collected by: ‘cohort’ recruitment at birth hospital of predetermined sample size; and by continuous, ongoing ‘census’ recruitment of women at first midwife appointment. Surveys collected experiential and outcome data related to existing services. Included women could have reached surveys issued 12 months after enrolment. Data were collected from women in Tuscany, Italy between September 2018 and July 2020. The total population was 7,784 individuals with 38,656 observations. We compare the two models of longitudinal collection of user-reported data using descriptive statistics, survival analysis, cost comparison, and qualitative review.

Results:

Cohort sampling provides lower initial participation than census sampling, but very high subsequent response rates - 87% one year after enrolment. Census sampling has higher initial participation, but greater dropout (to 45% at one year). Both models have high response rates for online surveys. There are non-proportional dropout hazards over time. There are higher rates of dropout for women with foreign nationality (Hazard Ratio 1.88), and lower rates for those more highly educated (HRs 0.77 and 0.61 for women completing high school and college respectively), employed (HR 0.87), in a relationship (HR 0.84) and with previous pregnancies (HR 0.86). The census model is initially more expensive, with lower repeat costs, becoming cheaper if repeated more than six times.

Conclusions:

The digital collection of user-reported data enables high response rates to targeted surveys in the maternal pathway. The point at which pregnant women or mothers are recruited is relevant for response rates and sample bias. The census model of continuous enrolment and real time data availability offers a wider set of potential benefits, but at initially higher cost and with the requirement for more substantial data translation and managerial capacity to make use of such data