Maintaining human milk bank services throughout the COVID-19 pandemic: A global response

If maternal milk is unavailable, the World Health Organization recommends that the first alternative should be pasteurised donor human milk (DHM). Human milk banks (HMBs) screen and recruit milk donors, and DHM principally feeds very low birth weight babies, reducing the risk of complications and supporting maternal breastfeeding where used alongside optimal lactation support. The COVID-19 pandemic has presented a range of challenges to HMBs worldwide. This study aimed to understand the impacts of the pandemic on HMB services and develop initial guidance regarding risk limitation. A Virtual Collaborative Network (VCN) comprising over 80 HMB leaders from 36 countries was formed in March 2020 and included academics and nongovernmental organisations. Individual milk banks, national networks and regional associations submitted data regarding the number of HMBs, volume of DHM produced and number of recipients in each global region. Estimates were calculated in the context of missing or incomplete data. Through open-ended questioning, the experiences of milk banks from each country in the first 2 months of the pandemic were collected and major themes identified. According to data collected from 446 individual HMBs, more than 800,000 infants receive DHM worldwide each year. Seven pandemic-related specific vulnerabilities to service provision were identified, including sufficient donors, prescreening disruption, DHM availability, logistics, communication, safe handling and contingency planning, which were highly context-dependent. The VCN now plans a formal consensus approach to the optimal response of HMBs to new pathogens using crowdsourced data, enabling the benchmarking of future strategies to support DHM access and neonatal health in future emergencies. The core Virtual Collaborative Network (VCN) was formed over a 2‐month period from 17 March, just as the WHO declared a global pandemic. It was formed by using a WhatsApp group, which the founders G. W. and N. S. recognised was a technology available in every country, without censorship and available to anyone with a mobile phone. As such, the founders approached the heads of every milk bank association that represented milk banks in more than one country, as well as milk bank leads from countries where individual milk banks operated (e.g. Kenya). We also approached nongovernmental organisations (PATH and Alive and Thrive) who had expertise in this sector with links to the WHO, in order to facilitate the recruitment of milk bank leads into the VCN and information flow between the VCN and WHO. Academics with specific expertise in human milk banks, including neonatologists who were clinical directors for milk banks, and social scientists, including anthropologists, were approached to join by email, followed up with a link to the WhatsApp group. In the first 2 months, a weekly update was made available on a central Google Doc resource so that new members to the VCN could review the conversations and information that had already been exchanged by the group. In a similar manner, this manuscript was effectively ‘crowdsourced’ as all members had access to edit and submit country‐ or regional‐specific information. Predictions were made of the total number of premature recipients across the countries with operational HMBs. This was done by making use of publicly available per‐county birth rates (Central Intelligence Agency, 2020), UN estimates of population sizes (Worldometer, 2020) and preterm birth rates per region (Blencowe et al., 2012). Mortality rates were not factored in. Results were generated on a regional basis, with designations of countries to regions as specified by Blencowe et al. For the purposes of this estimation, the population of preterms considered include births below 32 weeks gestational age. An initial number of HMBs per country was obtained from PATH (2020), updated where necessary from information provided by members of the VCN local to those countries. VCN members were requested to share up to date information regarding HMB operations (numbers of recipients and DHM volumes). The granularity of the provided data varied. Regional data were obtained for instance in the case of North America (via Human Milk Banking Association of North America), and national data were received for several countries, for example, Brazil and India; otherwise, data were received for individual HMBs. As the information received was not complete, that data were only made available for a subset of countries/or individual HMBs within a country, and the number of recipients was not provided in many cases; approaches for estimating the missing data were required. Three such approaches were employed: (A) for countries where only the volume of DHM is reported, the number of recipients is estimated using the volume per recipient, averaged over all (global) responses that included the volume and number of recipients; (B) for countries reporting data for only a subset of known HMBs, data were extrapolated to the full set of known HMBs within that country; (C) for countries for which only the number of HMBs was known, the number of recipients was estimated based on the (global) average of the calculated number of recipients per HMB where data allows. Data from each milk bank leader and national associations (>80 members of the VCN as of 1 May) were collated by three authors (M. S., N. S. and G. W.). Data were collated and analysed using Excel (Microsoft 365, Microsoft, WA). A set of open‐ended questions were circulated to the group to ask for their experience of operating an HMB, or the experiences of a national or regional network, in approximately the first 2 months of the global pandemic. Evidence collection started on 23 March and concluded on 1 May. Experiential descriptions of challenges faced in milk bank service provision during the COVID‐19 pandemic were submitted to and analysed by G. W. and N. S. for themes regarding the challenges raised by the COVID‐19 pandemic. Examples of responses to the pandemic were communicated by each of the co‐authors in their country‐specific context, and each co‐author read and approved the mitigation steps as outlined.