Infection prevention and control and related practices in African neonatal units: The Pan-African neonatal care assessment study (PANCAS)

Background The burden of neonatal mortality is primarily borne by low- and middle-income countries (LMICs), including deaths due to healthcare-associated infections (HAIs). Few studies have assessed infection prevention and control (IP&C) practices in African units caring for small and/or sick newborns aimed to reduce HAIs. Methods We performed a mixed-methods study composed of a survey and virtual tour to assess IP&C and related practices. We created a survey composed of multiple-choice and open-ended questions delivered to site respondents via Zoom or video equivalent. Respondents provided a virtual tour of their unit via video and the study team used a checklist to evaluate specific practices. Results We recruited 45 units caring for small and sick newborns in 20 African countries. Opportunities to optimize hand hygiene, Water, Sanitation and Hygiene (WASH) practices, Kangaroo Mother Care, and IP&C training were noted. The virtual tour offered further understanding of IP&C challenges unique to individual sites. All respondents expressed the need for additional space, equipment, supplies, education, and IP&C staff and emphasized that attention to maternal comfort was important to IP&C success. Discussion This study identified opportunities to improve IP&C practices using low-cost measures including further education and peer support through learning collaboratives. Virtual tours can be used to provide site-specific assessment and feedback from peers, IP&C specialists and environmental engineering experts.


Introduction
The burden of neonatal mortality is primarily borne by low-and middle-income countries (LMICs) where the risk of death in newborns is 10-15 times higher than in high-income countries (HIC) (Wang et al., 2016).While most deaths are attributed to delivery complications or early onset sepsis, deaths due to potentially preventable healthcare-associated infections (HAIs) in hospitalized small and/or sick neonates are also of increasing concern, (Olivier et al., 2018) particularly those caused by multidrug-resistant organisms (MDROs), requiring treatment with costly, broad-spectrum antibiotics (Morkel et al., 2014).
In May 2022, the World Health Organization (WHO) attributed adverse impacts on human health to lack of infection prevention and control (IP&C) strategies in LMICs, including suboptimal water supplies, sanitation, hygiene, and waste management (WASH) services (World Health Organization, 2022).The WHO recommended local, national, and global priorities (including political commitment); policies to develop and sustain IP&C programs; training and continuous education to strengthen the specialty of IP&C; and surveillance systems to monitor HAIs, emerging pathogens, and WASH practices.
In LMICs and HIC, improving hand hygiene, environmental and equipment cleaning, and policies to decrease crowding have reduced HAIs in newborns (Harris et al., 2019;Seale et al., 2009;Conde-Agudelo and Díaz-Rossello, 2014;Cross et al., 2019;Gon et al., 2021).Furthermore, in LMICs, Kangaroo Mother Care (KMC) has significantly reduced healthcare-associated sepsis, and thus serves as an IP&C strategy (Arya et al., 2023).Yet, few studies have systematically evaluated IP&C practices in African hospitals caring for small and/or sick neonates (Harris et al., 2019;Cross et al., 2019;Asare et al., 2009;Dramowski et al., 2016Dramowski et al., , 2021)).Thus, the objectives of this study were to [1] assess specific IP&C practices, including KMC, in African hospitals caring for small and/or sick newborns <30 days old, [2] identify gaps in IP&C practices to inform future strategies to improve these practices, and [3] describe the needs and priorities of clinicians caring for this population.

Study design, site recruitment, survey respondents
To assess IP&C practices in diverse hospitals in sub-Saharan Africa caring for small and/or sick newborns, we developed a mixed-methods study consisting of a survey complemented by a virtual assessment of the care units and by clinicians' self-reported needs and priorities for newborn care.
Potential sites were identified from past partnerships, referrals from global health physicians, or snowball referrals from participating sites.Sites received an email invitation (English or French) that provided an overview of the study's purpose and survey content.Three attempts to contact each site were made.Each individual site selected a respondent (s) familiar with IP&C practices, e.g., unit medical director, to answer the survey and/or conduct the video tour.The respondents were encouraged to consult other team members (e.g., the charge nurse, environmental workers) for assistance completing the survey.The Columbia University Irving Medical Center (CUIMC) team indicated particular sections of the survey for which multidisciplinary team input might be needed.
Approval from the CUIMC Institutional Review Board was obtained.Sites obtained local ethics approval as per local requirements.Sites received $1000 USD to defray the costs associated with this research, e. g., IRB fees.

Survey development
The CUIMC study team, comprised of experts in global health and IP&C, developed a survey based on guidelines for evidence-based IP&C practices for units caring for small and/or sick newborns (Infection Prevention and control Assessment, 2018).The survey assessed multiple domains including space and unit layout, visitor policies, KMC, systems capacity, environmental and equipment cleaning, WASH practices, and available supplies (Supplement 1).In addition, the survey asked about hospital and patient characteristics and the impact of the COVID-19 pandemic.Two open-ended questions investigated respondents' greatest needs, i.e., what they would fund if money was available.
The draft survey was reviewed by physicians from different African hospitals, physicians with global health experience from the American Academy of Pediatrics Global Child Health and Life Support section, and the Bill and Melinda Gates Foundation, and was modified accordingly.CUIMC public health students, infection preventionists, and pediatricians with experience in neonatal care pilot-tested the survey resulting in additional modifications.The final survey was translated into French.
The survey used Yes/No/Unsure responses, multiple choice questions, and open-ended questions.In multiple-choice questions, the option of "other" was offered to allow additional options that were then incorporated into the survey.To reduce recall bias for specific items, e. g., average daily census or availability of certain supplies, respondents provided responses based on the day the survey was administered and were asked if "this was a typical day?"

Virtual assessments
To complement the survey, we developed a virtual assessment tool and conducted a video tour via Zoom or WhatsApp on the same day as the survey, whenever feasible.Areas to assess were selected based on team members' prior site visits to pediatric hospital wards across Sub Saharan Africa (Tarun et al., 2023) (e.g., patient and visitor crowding, KMC capacity, sink accessibility), as well as informed by IP&C observational studies conducted in the U.S. and other countries (e.g., hand hygiene supplies, rubbish bin accessibility, sharps safety).IP&C observations that would identify patients, staff, or visitors were avoided.Thus, during the tour, respondents provided images of the items on the virtual assessment checklist avoiding patient, visitor, or staff faces or other identifiers.Respondents were asked to clarify images if something was not easily visible, e.g., presence of window screens or running water.If a specific area was unavailable for observations, it was not scored.Alternatives e.g., a video or photographs via encrypted messaging, were arranged if internet connectivity was suboptimal.Given the potentially subjective nature of the items scores (e.g., crowded conditions), at least two team members independently assigned a score and then discussed the results.To ensure further consistency of interpretation, one team member (IF) reviewed recordings of each virtual assessment.Discrepancies were resolved by adjudication from the rest of the team.

Analysis
Data were entered into REDCap for analysis.Observations of each item on the virtual assessment were scored 0-2 by two members of the CUIMC study team; higher scores reflected more favorable IP&C observations.Discrepancies were resolved by consensus with the study team.Descriptive statistics (frequencies, means, medians, ranges) were performed for survey responses and virtual observations.Responses to open-ended questions were transcribed; French responses were translated to English.The transcripts were reviewed to develop themes used to analyze responses to both questions.Transcripts were coded independently by two team members.Coded themes were not mutually exclusive.Discrepancies were resolved through consensus.

HAI
Healthcare associated infections LMIC Low-and middle-income countries IP&C Infection prevention and control WASH Water sanitation and hygiene KMC Kangaroo mother care MDRO Multi-drug resistant organisms

Hospitals characteristics
Ninety-six hospitals from 33 countries were contacted of which 45 (47%) participated from 20 countries (Fig. 1).Twenty-one (46%) hospitals were academically affiliated.Thirty were national/regional and 8 were district/provincial hospitals; 7 did not report these classifications.Forty (89%) had public funding, Two each reported being privately funded or funded by Missionary services, and one hospital was unsure about funding sources.
Infrastructure of the units that cared for small and/or sick newborns is shown (Table 1).Most (67%) had neonatal ICUs and a KMC room.Fewer had units to separate preterm and term newborns or an isolation area.Six hospitals had a single room for all newborns, and one had no designated area.All hospitals accepted newborns born at home or transferred from another hospital, and 33 (73%) readmitted newborns who had been discharged.
On the survey day, the mean census was 28 (median 19, range 0-95) small and/or sick newborns.On average, 9 newborns (median 5, range 0-41) had a birthweight <1500 g.Ten (22%) sites reported their census was usually higher than the survey day.A median of 10 (range 3-50) nurses, and 5 (range 1-20) physicians were working throughout the 24h period of the survey day.The median physician-to-patient ratio was 1:5 (maximum 1:34).All reported that this staffing reflected a typical day.Forty (89%) hospitals employed a physician(s) who completed pediatric training including 19 (42%) that employed a physician(s) who also completed neonatology training.

Kangaroo Mother Care
KMC was available in 42 (93%) sites.Nine (21%) were able to provide KMC at the newborn's bedside, 26 (62%) emphasized starting KMC as soon as possible, and 12 (29%) provided intermittent KMC to  critically ill newborns.On the survey day, 23% (mean per site, range 0-100%) of newborns were receiving KMC which 76% of sites reported was a typical day.All respondents reported their staff believed in KMC benefits, 37 (82%) reported that most mothers believed in KMC benefits, and 26 (62%) reported families supported mothers to perform KMC.Sixteen (38%) respondents reported their hospital was unable to optimally support KMC, primarily due to staffing shortages (n = 12/16, 75%).Respondents reported that 6 (mean per site, range 0-30) additional newborns could receive KMC if more resources were available.The most common challenges to providing KMC were family support/cultural incompatibility (40/45, 89%) and maternal and newborn medical conditions (69% and 61%, respectively).The most common support needed to increase KMC was educational material and training (35/43, 78%).Additional challenges and resources needed for KMC are shown (Table 3).

Resources for managing healthcare-associated infections
On the survey day, 25 (56%) sites were caring for newborns with HAIs.Clinical sepsis was most common followed by gastrointestinal illness, meningitis, pneumonia, and skin/soft tissue infections.Among the 11 sites with microbiologic data available on the survey day, 8 (73%) HAIs were caused by Gram-negative organisms.During the previous year, 23 (51%) sites experienced at least one HAI cluster/outbreak, 14 (61%) of which were caused by Gram-negative pathogens.Ten sites were able to cohort affected newborns.
Hand hygiene techniques were taught at 42 (93%) sites using posters, demonstrations, and lectures that described WHO recommendations for the "Five Moments for Hand Hygiene".Nine (20%) sites did not teach the WHO recommendation to perform hand hygiene after touching patient surroundings.IP&C staff were available for the unit at 21 (47%) sites, and 11 (24%) formally monitored hand hygiene adherence.On the survey day, 35 (78%) sites reported available sinks in newborn care areas, but 7 (15%) reported fewer working sinks than usual.Four sites (9%) had no functioning sinks on the day of the survey.Most sites reported enough soap (n = 38, 84%), alcohol hand sanitizer (n = 44, 98%), and disposable towels (7/11, 64%).The virtual assessment demonstrated that 19 (42%) sites had barriers accessing sinks and 38 (84%) lacked one or more hand hygiene supplies (Table 2).
Environmental cleaning and disinfection were performed by an agency at 39 (87%) sites or by nurses and/or families at 6 (13%) sites.The types of cleaning agents and frequency of cleaning are shown (Table 5).During the virtual assessments, 20 (44%) sites had bedside clutter, 8 (18%) had inaccessible or overflowing rubbish cans, and 15/31 (48%) had unclean spaces in procedure areas (Table 2).

Respondents' perceptions of newborn care needs
Themes elicited from respondents in open-ended questions included needs for space, equipment and infrastructure, nutrition, IP&C, staffing, and maternal and staff well-being (Table 6).Across many of these themes, respondents expressed concern for the experiences of mothers and stated that adequate support would encourage mothers to participate in KMC and reduce the risk of "contamination".
Space: Respondents identified the need for dedicated spaces for KMC, to meet maternal needs (e.g., a dining room, for preterm newborns, for isolation, and to accommodate equipment.One respondent noted, "I cannot ventilate because I do not have the equipment.And even if I do, then I cannot put it into practice because I have a very, very tight space." Equipment, Supplies, and Infrastructure: Respondents frequently cited the need for respiratory support including continuous positive airway pressure (CPAP) and pulse oximetry.One respondent reported having one CPAP machine for 40 beds, which lead to sharing and re-using supplies and equipment.Several described the need for PPE, e.g., gowns and gloves.Storage space for supplies and equipment and for storage of breastmilk and mothers' belongings was another focus.
Nutrition: Some expressed the need for nutrition for infants and mothers.One respondent said, "We don't have breastmilk, we don't have parenteral nutrition for babies who stay on the ventilator for a long time.We rely on IV fluids.We cannot stop giving care, but we have no nutrition".
Infection Prevention & Control: Many respondents highlighted the need for IP&C education and laboratory support including antimicrobial susceptibility testing.Others emphasized the need for clean water and avoiding reusing equipment to reduce risk of contamination.One respondent stated, "We would spend [money] on monthly running water.The times when we run out of water, it puts our patients at risk of infection".Staff Training: Respondents noted the need for neonatology education and training.One respondent said, "Just our major priority or the first thing we want to have is just getting a neonatologist or training manpower".
Well-being: Respondents emphasized the need to improve the wellbeing of staff.They voiced that in addition to the training mentioned above, staff needed resources such as laundry, storage, and breaks.Respondents also highlighted needs such as water, food, restrooms, showers, and recreational materials for mothers.One respondent expressed "I am sad each day when mothers struggle to go to use the restroom and they come back to stay with their babies and so I would change the setup to be more comfortable for moms".

Discussion
This multi-method study offers a unique look at IP&C practices and related resources available at diverse hospitals in Africa providing care to newborns.We found that most hospitals faced significant challenges addressing IP&C needs.These challenges included the need for additional space, equipment, supplies, education, training, and staff.Many of these challenges can only be addressed with substantial funding and extensive infrastructure improvements.However, data from studies

Well-being
Mothers & communities (n = 10) "I would like to have shower for the mothers, and recreation, video for the mothers to watch while kangaroo mother care is done, I want the mothers to have tea and coffee in the middle of the night.""So I think the main thing that we would stand to benefit is to make our kangaroo mother care bigger and to find some way to help mothers-to make their lives easier.""The greatest need is to help the communities.That's a difficult thing to do.Staff (n = 4) "I want each physician to have his own stuff laundered every day, shoe to enter into the ICU, locker for every physician to put his clothes and things away from the neonatal ICU for the safety.""Our nurses are ICU nurses they work 24/7 and almost all the time they are here somehow they miss refreshments, trainings, updates […] it's very heart wrenching they don't go out for lunch, they don't go out to pee, they don't go hangout with their friends because once they enter the ICU they have to stay very long hours in the evenings a lot." I. Frantzis et al. such as this one are important as they vividly describe the extent of common challenges and provide priority areas for investment.The countries represented by the hospitals in this study bear a disproportionate burden of HAIs, (Decraene et al., 2018) particularly caused by Gram-negative organisms.When microbiologic data were available, Gram-negative pathogens caused most HAIs and clusters/outbreaks.Notably, a minority of sites used the evidence-based practice of cohorting newborns during clusters/outbreaks to avoid additional transmission to other newborns (Hayward et al., 2022).Furthermore, many relatively narrow spectrum antibiotics used by the sites were consistent with WHO recommendations which may not reflect the susceptibility patterns seen in these countries (Richard et al., 2001).Thus, without consistent laboratory support for obtaining cultures and performing susceptibility testing to inform treatment and surveillance, HAIs caused by multidrug-resistant Gram-negative pathogens may be inadequately treated and under-estimated.
We also found that relatively low-cost IP&C practices could be further strengthened.While most sites taught the WHO's Five Moments for Hand Hygiene, 20% did not teach that hand hygiene should be performed after touching the patient care environment, an area that can be heavily contaminated by patient flora and serve as a reservoir for potential pathogens (FitzGerald et al., 2013;Bhalla et al., 2004).
Hand hygiene monitoring has been associated with an improved understanding of hand hygiene opportunities and increased adherence to hand hygiene practices, (Bhalla et al., 2004) but only a quarter of sites monitored hand hygiene.Many sites used alcohol-based hand sanitizers, but both the survey and video assessments revealed a lack of hand hygiene supplies and barriers to accessing sinks.Future work could develop strategies to support hand hygiene such as eliminating physical barriers to sink access, providing alcohol sanitizer at point of care, and monitoring hand hygiene adherence.IV dressing change policies, along with other preventive 'bundle strategies', have been associated with decreasing central line-associated bloodstream infections, (Huang et al., 2021) yet only half the sites that provided IV therapies had such policies for dressing changes.Providing window screens would both improve ventilation and reduce the risk of airborne and zoonotic infections (Ogoma et al., 2009).
Many sites had WASH practices that aligned with WHO recommendations to use chlorine-based/bleach products and perform a minimum of daily cleaning of environmental surfaces.However, the video assessments revealed crowding and clutter that could impede environmental cleaning suggesting that efforts to reducing crowding and clutter could improve the effectiveness of cleaning and disinfection.The U.S. Centers for Disease Control and Prevention have also increasingly emphasized safe water practices to reduce HAIs as municipal water can be contaminated by coliform bacteria and hospital pipes may harbor biofilms of potential bacteria and mycobacteria (Best Practices for Environmental Cleaning; Decraene et al., 2018;Hayward et al., 2022).Thus, sterile water is recommended to fill respiratory therapy equipment and prepare formula, but not all sites used sterilized water for these activities.
Additionally, IP&C strategies can be implemented to improve staff and visitor safety.For example, during the video assessments, sharps containers were frequently inaccessible, overflowing, and/or not secured on the wall.Focused attention to secure sharps containers and avoid overfilling could reduce the risk of sharps injuries to staff and potentially to visitors (Richard et al., 2001).Additionally, some PPE practices at the sites are not evidence-based such as wearing head or protective shoe coverings; disbanding these practices could be explored locally and ultimately be cost-savings.
KMC offers protection against HAIs likely due to reduced newborn contact with staff and improved breastfeeding (Arya et al., 2023).While passionate about performing KMC, respondents identified gaps in resources for KMC.Strengthening KMC through staff, mother, and community education and by improving maternal comfort through availability of water, food, and showers could expand the use of KMC.
Many respondents emphasized addressing maternal needs as they noted that mothers who are unwell or frustrated cannot stay to look after their newborn or provide protection through KMC.Local work to reduce barriers to KMC due to incompatibility with cultural norms should also be considered (Best Practices for Environmental Cleaning).
While this study identified many shared experiences and challenges, the unique characteristics of each hospital including differing settings, resources, and specific needs means that a simple unifying solution to fit all does not exist.In this context, a more customized approach to address each care units' IP&C challenges may prove more successful.The COVID era accelerated the use of virtual platforms, and we were able to review care settings from distances not possible just a few years ago.This medium facilitated our understanding of IP&C efforts and opportunities unique to each site.We anticipate that this method could be further developed to link remote hospitals with peers and with global experts who can address particular needs and provide tailored feedback.

Limitations
This study has several limitations.Respondents were selected because they were knowledgeable about the units providing care to small and/or sick newborns.This could have introduced potential biases including recall bias and socially desirable responses.Study sites were diverse; however, a larger cohort with more district hospitals may have offered further insights.Comparison between different types of hospitals was not done as many sites were unsure of their designation.While we routinely asked respondents if 'this was a typical day', we did not seek additional validation, e.g., alternative sources of census and staffing.Staffing for nursing was reported as total over 24 h and thus could not reliably calculate nurse to patient ratio per shift.Details regarding water treatment were not asked.
Due to limited internet connectivity in some areas, real-time virtual assessment of the sites could not always be performed.The virtual tour was planned in advance; thus sites could have made modifications prior to the tour.We did not ask the sites if they had made changes prior to the virtual assessment.Though there were separate questions in the survey to distinguish "shared" vs "reused" equipment, multiple respondents used the terms interchangeably.Thus we combined these responses in our analysis.Although we asked about the potential for cultural norms of families to be challenges to KMC, we did not explore staff cultural norms, which may also have been barriers to KMC.The qualitative results demonstrated the need to include questions about mother, community, and staff well-being in future studies.

Conclusions
This study highlights several opportunities for strengthening IP&C practices in hospitals caring for small and/or sick newborns.Results can be used by potential donors and health ministries for targeted areas of funding as well as by local neonatal units to inform their own practices and seek resources.
Respondents consistently identified the need for more space, equipment, personnel, supplies, education, and training for their units.These physical and workforce needs were not surprising and are costly.In addition, we identified economical strategies that largely involved strengthening implementation of evidence-based practices for hand hygiene, central line care, and WASH practices.To implement these strategies effectively further education and support are critical.Future work could explore creating a learning collaborative that promotes networking among hospitals and facilitates sharing practices, educational materials, ideas, and solutions (McMullan et al., 2013).Our findings suggest that virtual assessments are feasible and well accepted and could provide real-time assessment to hospitals by peers and/or global IP&C and environmental engineering experts.These approaches would offer opportunities to tailor IP&C strategies to the needs of individual hospitals.

Table 1
Infrastructure of sites providing care for small and/or sick newborns <30 days old.
a Self-designated.I.Frantzis et al.

Table 2
Virtual tour assessment checklist results.

Table 3
Available resources, challenges and support needed to optimize KMC.
a Maternal resources (n ¼ 42) a 42/45 sites provided Kangaroo Mother Care.b Financial (n = 2), distance to hospital (n = 2), equipment to facilitate monitoring/treatment of newborns while providing KMC (n = 1).cTwo sites reported no further support is needed.

Table 4
Shared and/or reused patient-care equipment and cleaning method.
a n = positive response/N = equipment or supply available.bNoavailable response at one site.c Respiratory equipment, e.g., oxygen circuit, endotracheal tube.d Feeding tools, e.g., spoons, bottles, cups.

Table 5
Cleaning agents and frequency of cleaning for different environmental surfaces.
a Respondents able to select more than one option.

Table 6
Themes and selected respondent quotes describing newborn care needs.
"You know even pulse oximetry is not available because the sensor is not a disposable one, so the pulse oximetry is available, but the sensor is not there." "[…] because there is only one suction machine, which is used for several patients and the contamination is very high."Specific needs (n = 35) "The greatest need now is actually one thing.We need heaters, we need to be warm.""We need incubators, we need KMC chairs.""I believe that is the most important thing in infection prevention in our unit … regular training of our staff in infection prevention and control.And working on stewardship with the staff.""There are many issues but the major one is the blood culture, urine basic culture, we don't have access to cultures so again we are guessing what bacteria we are