JMIR Publications

ABSTRACT

Background:

Medical resource allocation is important to ensure availability of care, also in challenging circumstances as a pandemic. In fields of unpredictable care demand such as obstetrics, forecasting may help manage scarce resources.

Objective:

The development, validation, and implemention a regional birth forecasting tool to support obstetrical staff planning in the Utrecht region during the COVID-19 pandemic.

Methods:

We combined predicted birth dates retrieved from Saltro, a large regional primary care laboratory, with data from the Dutch national perinatal registry (Perined) and Statistics Netherlands for model development. We created and implemented an html-tool visualising these forecasts, which were discussed during the regional acute obstetric health care network meetings. Six months after implementation we assessed the impact of the tool using an evaluative stakeholder meeting.

Results:

We achieved a performance accuracy (R2) of 0.45, 0.61 and 0.67 (all actual number of births within 95% CI) forecasting the number of births in the region, pooled in one- two and three-weekly bins respectively. After presenting these findings to stakeholders, we implemented a forecasting tool using the two-weekly bin model. The evaluative stakeholder meeting proved that the tool improved communication, awareness of health care need, and collaborations among health care providers in the Utrecht region. Additionally, stakeholders identified additional applications, such as communication with patients, and training of obstetric healthcare providers.

Conclusions:

Implementation of a forecasting tool for number of births based on available data across the healthcare system added value to obstetrical care by providing insight into care demand, and increasing communication, awareness and collaboration between health care providers. Further research should aim at improving regional obstetric acute care by fostering data sharing, in order to improve health care demand forecasts.