Developing and implementing a family-led infection prevention bundle for hospitalized neonates

doi:10.21203/rs.3.rs-7695391/v1

Download PDF

Research Article

Developing and implementing a family-led infection prevention bundle for hospitalized neonates

https://doi.org/10.21203/rs.3.rs-7695391/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

For many neonatal units in resource-limited settings, family involvement in patient care is both a logistic necessity and a cultural norm. However, incorporating families into hospital infection prevention and control (IPC) activities remains under-implemented and under-studied. We developed a family-led infection prevention (FLIP) bundle for a neonatal unit in Botswana and evaluated its impact on multidrug-resistant organism (MDRO) colonization, bloodstream infection (BSI) incidence, and all-cause mortality.

Methods

The FLIP bundle was developed with input from a multidisciplinary group of stakeholders including subject matter experts, staff, and families. The bundle components consisted of staff-led family orientation on 1) hand hygiene, 2) twice-weekly neonatal skin cleansing with 2% aqueous chlorhexidine gluconate, 3) lactation support, and 4) skin-to-skin contact. The bundle was implemented over 18 months (August 2023 – January 2025) at a 33-bed neonatal unit in Botswana during which time it was iteratively revised based on process metric performance. MDRO colonization was assessed twice-monthly using perirectal and skin swabs to detect extended-spectrum cephalosporin-resistant and carbapenem-resistant Enterobacterales and Acinetobacter spp. BSI incidence was determined from growth on neonatal blood cultures, excluding contaminants. In-hospital mortality data were extracted from hospital records. Colonization, BSI, and mortality rates were compared before and after pilot and continuation phases using two-proportion z-tests and Chi-square tests.

Results

Implementation of the FLIP bundle was temporally associated with a significant relative reduction in all-cause mortality by 28% across the unit, decreasing from 15.4% (165/1069) pre-FLIP to 11.1% (208/1866) during FLIP (p < 0.001). BSI incidence decreased modestly from 11.3% (121/1069) to 9.7% (181/1866) during FLIP (p = 0.14). MDRO colonization trends showed mixed results, with initial improvements to Acinetobacter spp. colonization but an overall 9.4% increase in skin MDRO colonization observed (p < 0.001) in the implementation period.

Conclusion

Implementation of the FLIP bundle demonstrated the potential of systematically engaging families in neonatal IPC programs, with a notable improvement in infant survival observed. The lack of clear improvement in BSI incidence and MDRO colonization prevalence may reflect incomplete adoption or poor impact of the bundle and highlights that FLIP should be used in conjunction with robust healthcare worker–led IPC measures.

In many hospitals, especially in places with limited resources, family members play an important role in caring for newborn babies. However, they are often not included in hospital programs designed to prevent infections. To address this, we created a family-led infection prevention (FLIP) program in a neonatal unit in Botswana. This program helped families understand how they could help to reduce infection risks through simple actions like proper handwashing, regular cleaning of the baby’s skin with an antiseptic solution, maximizing the provision of human milk and breastfeeding, and engaging in skin-to-skin contact with their baby.

We introduced this program over 18 months and measured rates of newborns with antibiotic-resistant bacteria, blood infections, and deaths during hospitalization. We found that after implementing the FLIP program, there was a substantial decrease in newborn deaths, though we did not observe consistent reductions in infections or colonization with antibiotic-resistant bacteria. This study shows that including families in infection prevention efforts in neonatal units can improve newborn survival and highlights the importance of family-centered care in hospital settings.

Neonatal deaths account for nearly half of all under-five mortality worldwide, about a third of which are attributable to neonatal sepsis.¹ The incidence and mortality from neonatal sepsis are higher in low- and middle-income countries (LMICs), partially due to the emergence of multidrug-resistant organisms (MDROs), which are now the leading causes of bloodstream infection among hospitalized neonates in sub-Saharan Africa and South Asia.^2,3 In Botswana, neonatal infections due to extended-spectrum beta-lactamase-producing Enterobacterales and carbapenem-resistant Acinetobacter baumannii have increased over the past decade.^4–6 Infections with MDROs are difficult to treat and are associated with high mortality; on average, one in three infants with an MDRO infection will die.^4,7

A combination of maternal, host, and environmental risk factors are thought to influence a newborn’s susceptibility to infection. In LMICs, common maternal factors include maternal infection and prolonged rupture of membranes, which are associated with high neonatal pathogen colonization and infection.⁸ Preterm labor and perinatal asphyxia are host factors that frequently occur in LMICs and increase the risk of sepsis due to the need for prolonged hospitalization.^9–11 Conversely, early human milk exposure and skin-to-skin contact can protect against neonatal infection and all-cause mortality.^12–15 Skin-to-skin contact is hypothesized to reduce infection through decreased neonatal stress response,¹⁶ increased skin barrier integrity,¹⁷ and transfer of protective maternal flora.¹⁸ Breastfeeding promotes maternal antibody production via the enteromammary pathway in response to shared microbial exposures; in comparison to formula-fed infants, infants fed exclusively human milk have a lower burden of gut colonization with Klebsiella pneumoniae and antibiotic resistance genes.¹⁹

Hospitalized neonates are often colonized by MDROs due to the presence of environmental reservoirs in hospital wards and frequent handling by healthcare workers.²⁰ Infection prevention and control (IPC) efforts to reduce environmental contamination and patient colonization with MDROs include multimodal cleaning strategies and routine skin cleansing with aqueous chlorhexidine gluconate (CHG).^21–23 However, IPC efforts in LMICs are often hampered by overcrowded wards, equipment re-use, and a critical shortage of healthcare workers to implement even basic IPC measures.²¹

Hospital IPC programs have historically centered their focus on the role of healthcare workers (nurses, doctors, and allied health professionals), neglecting the role of families in IPC. Family involvement in IPC has been identified as a promising strategy for implementing IPC measures, particularly in resource-limited settings.^21,24 However, incorporating families into hospital IPC policies and practices remains a challenge, and is under-studied and variably implemented, particularly in neonatal units. A 2022 systematic review of published reports of IPC bundles from LMICs identified only one study that included an element of family involvement.²⁴ That study, performed in a Vietnamese tertiary hospital, implemented a bundle promoting early breastfeeding and skin-to-skin contact for preterm infants which resulted in significant reductions in neonatal intensive care unit admissions, hypothermia, sepsis, and formula use, resulting in decreased monthly costs.²⁵ Engaging families of hospitalized patients around hand hygiene and other IPC activities has been shown to be acceptable in LMICs. One study from South Africa reported maternal involvement was key to achieving improved environmental and equipment cleaning.²¹

Although bundled IPC interventions limit assessment of individual components, they are consistently the most effective strategy for reducing healthcare-associated infections in neonates, whose complex infection risks require coordinated action.²⁶ In this article, we describe a quality improvement project to design and implement a family-led infection prevention (FLIP) bundle at a referral neonatal unit in Botswana. We hypothesized that such a bundle would be a low-cost, high-yield IPC strategy to reduce sepsis, mortality, and, potentially, antibiotic resistance in our setting.

Study design and setting

This prospective implementation study used a quasi-experimental before-and-after design to assess the temporal impact of the FLIP bundle on patient outcomes. The bundle was piloted at a 33-bed neonatal unit in a 530-bed public tertiary referral hospital in Botswana, where 6,000–8,000 deliveries occur annually. Common neonatal diagnoses in this unit include prematurity-related complications, hypoxia-related injuries, and sepsis, reflective of global patterns of neonatal mortality, and the average length of stay is about three weeks.²⁷ The nurse-patient ratio ranges from 1:6 to 1:12, with family members assisting in neonatal care tasks such as feeding, changing/cleansing babies, and medication administration due to staffing shortages. The hospital has a dedicated IPC program, including two full-time infection prevention nurse practitioners. Access to soap, water, and alcohol-based hand sanitizer is generally reliable in the neonatal unit; however, personal protective equipment such as gloves and gowns are frequently out of stock.

Intervention: The FLIP Bundle

The FLIP bundle was developed and implemented in six stages from October 2022 – January 2025: 1) Survey of subject matter experts (SMEs); 2) Identification of priority thematic areas; 3) Critique by local stakeholders; 4) Feedback from families; 5) 12-month pre-pilot surveillance; and 6) Six-month FLIP Pilot followed by a 12-month continuation phase (Fig. 1).

Involvement of subject matter experts

We invited 32 international SMEs to contribute to the development of the FLIP bundle. SMEs were identified through investigators’ networks and referrals, selected for expertise in neonatology, nursing, microbiology, IPC, behavioral science, lactation, and public health. Among those invited, 12 SMEs with experience in resource-limited settings across sub-Saharan Africa, Central America, and South Asia completed an electronic survey and participated in a collaborative teleconference. The survey covered key topics related to IPC and family involvement in neonatal care, including lactation, infant feeding techniques, skin-to-skin contact, skin integrity, hand hygiene, environmental hygiene, and maternal symptom screening (Supplemental Document 1: FLIP Bundle Development Survey). SMEs graded proposed interventions based on feasibility and effectiveness in resource-limited settings. The modified Delphi approach was used to build consensus through three teleconferences conducted over the course of six weeks.²⁸

SMEs identified four core components for the FLIP bundle: hand hygiene, CHG skin cleansing, lactation support, and skin-to-skin contact. While some SMEs supported family involvement in environmental cleaning, citing potential empowerment and increased cleaning frequency, this component was ultimately excluded due to concerns about cleaning product misuse and reduced time for family-infant bonding. SMEs pointed out that in most LMIC neonatal unit settings, mothers assume > 99% of family caregiving in the hospital (by tradition and hospital policy); however, using the term ‘family’ normalizes language which is inclusive of other family members like fathers and grandparents.

Additionally, SMEs identified key barriers to implementation, including ward hierarchy dynamics, limited leadership buy-in, and insufficient staff capacity to orient families. They suggested involving ward staff in bundle development and agreed on three primary outcome metrics for evaluating the FLIP bundle: prevalence of MDRO colonization, incidence of bloodstream infections (BSIs), and in-hospital mortality rates.²⁶

Involvement of local stakeholders

Various local stakeholders were engaged to further assess topics and interventions identified by SMEs based on perceived effectiveness and feasibility during an in-person meeting. Fifteen local stakeholders consisting of neonatal unit staff (doctors and nurses) and Ministry of Health representatives (Child Health Services) participated. Neonatal unit staff suggested FLIP bundle orientation materials be summarized in a succinct checklist to be completed with families, with priority tasks listed in order of descending priority and time-sensitivity (e.g., hand hygiene, feeding, and early human milk expression being high priority to discuss with families shortly after admission) (Supplemental Document 2: Family Handbook)

Involvement of families

Parents of infants previously admitted to the neonatal unit were invited to review an early version of the FLIP bundle. Four parents, including one whose infant survived neonatal sepsis, volunteered to share their experiences. Families expressed that such an intervention was long overdue and requested clear, written guidance outlining their roles in care tasks. The FLIP bundle was well received, aligning with botsetsi—a traditional postpartum confinement practice in Botswana, typically lasting three months—which promotes maternal and infant health and infection prevention through nutritional support and restricted visitation.²⁹

FLIP bundle components

The piloted version of the FLIP bundle included an admission care package for families which included a moisture-wicking bag to secure personal items during hand hygiene and care tasks. The bag contained a miniature dispenser of alcohol-based hand rub, nail clippers (to optimize hand hygiene), surgical masks to reduce droplet-spread of respiratory infection, a plastic cup and cover to support hand-expression of colostrum, a pen for recording feeding volumes and skin-to-skin contact hours, and a handbook written in English and Setswana containing information on skin-to-skin contact, CHG cleansing, and general information on the care of newborns (Fig. 2). Care package contents were chosen for their affordability and importance in infection prevention.

During disbursement of care packages, families received a 20-minute group orientation shortly after admission led by study staff, covering essential FLIP bundle practices. Each session typically included three to six families. Core practices covered during the orientation included:

Breastfeeding: Families were encouraged to initiate breastfeeding and/or milk expression (by hand or pump) within 60 minutes of birth and aim for at least eight feeding sessions/expressions per day.

Skin-to-Skin Contact: Target of eight hours daily, adjusted for space limitations.

Hand Hygiene: Eleven-step handwashing protocol using soap and water before unit entry, after using the bathroom or changing diapers, and when hands were visibly soiled.³⁰ Alcohol-based rub was used before and after infant care tasks.

CHG Cleansing: Conducted twice weekly by families with 2% aqueous CHG-impregnated wipes, supervised initially by healthcare staff. Cleansing was performed neck-to-toes for infants ≥ 1 kg and > 24 hours old. It was deferred for infants with skin breakdown, hypothermia, or when ambient temperatures were low. When families were not available, staff assisted with CHG cleansing.

Process Metrics

In the pre-FLIP and pilot phase, human milk feeding rates and skin-to-skin hours among admitted infants were monitored through weekly point prevalence surveys reviewing infant bedside logs where mothers recorded the number of feeds and/or the amount of human milk provided if fed by tube, cup, or syringe. Monitoring for key FLIP process metrics began in August 2023 with the initiation of the pilot phase and was performed by the hospital lactation support team—comprising two part-time lactation consultants and two assistants whose aims were to educate staff and families regarding breastfeeding/expression and troubleshoot common breastfeeding challenges. In January 2024, there was a monitoring gap during which time the hospital lactation team revised their monitoring strategy to address lack of observed improvement in breastfeeding and skin-to-skin contact rates. In February 2024, revised surveillance was launched, focusing on practices within the first 72 hours after birth, including time to first breastfeeding/expression across the entire postnatal ward. Twice weekly, mothers of newborns under 72 hours were interviewed at their bedside regarding feeding and skin-to-skin practices. This change aimed to prioritize the critical window for establishing lactation and skin-to-skin contact and provide timely education. Due to this methodological change, monitoring data from the revision and continuation phases are not directly comparable to earlier periods.

We monitored hand hygiene adherence of family members through direct observation by study staff during unannounced monthly audits conducted by study staff (range: 40–48 observations per month). CHG adherence was measured twice weekly through direct observation as the proportion of eligible infants (> 1kg, > 24 hours of age) who received CHG cleansing. Prior to FLIP implementation, hand hygiene and CHG cleansing were routine but not systematically monitored due to staff shortages. These metrics were formally tracked from the pilot phase onward (Supplemental Fig. 4). CHG cleansing was not conducted in the final three months of the continuation period due to a hospital-wide stock-out of CHG. An additional sensitivity analysis of the association between FLIP and outcome variables was conducted removing the last 3 months of surveillance when CHG cleansing was no longer performed; the results did not significantly differ. Informal feedback on bundle components was gathered once a week through family listening sessions and ward huddles in the pilot phase during which time misconceptions were discussed and barriers were identified (Supplemental Table 1).

Outcome metrics

The temporal impact of the FLIP bundle was assessed using three key outcome measures:

MDRO Colonization: Point prevalence surveys were used to ascertain MDRO prevalence before and during the FLIP intervention. All inpatient infants underwent culture-based screening twice monthly using perirectal and periumbilical skin swabs, regardless of previous screening results Samples were collected with flocked swabs and transported in Eswab® (COPAN Diagnostics, Murrieta, USA) systems. Samples were inoculated within 24 hours onto chromogenic media selective for: extended-spectrum cephalosporin-resistant Enterobacterales (ESCrE) carbapenem-resistant Enterobacterales (CRE) and Acinetobacter spp. (CHROMAgar ESBL, mSuperCarba, & Acinetobacter, Paris, France). MDRO colonization was defined as having visible colony growth on any selective media by 48 hours. Although Acinetobacter species were not further differentiated by resistance profile, > 90% of Acinetobacter infections in this unit are carbapenem-resistant. Therefore, any Acinetobacter growth was considered a surrogate marker for multidrug-resistant colonization. Colonization screening ran from November 2022 to October 2024 as part of a concurrent study. Thus, cumulative colonization prevalence comparisons were restricted to this period.

BSI Incidence: BSIs were identified from positive blood cultures. Likely contaminants—such as coagulase-negative Staphylococcus spp., Micrococcus spp., and diphtheroids—were excluded.³¹ Crude BSI incidence was calculated as the number of BSIs per total neonatal admissions. Analyses were performed both including and excluding outbreak-associated BSIs, with outbreaks defined as five or more BSIs of the same species within a four-week period, derived from previous ward BSI surveillance data. A period of four weeks with four or fewer BSIs of the same species marked the end of an outbreak period. BSI incidence was tracked for one year before FLIP implementation and throughout the entire FLIP implementation period.

In-Hospital Mortality: Mortality data were extracted from hospital registers and analyzed for temporal trends. Sepsis-related mortality was not included as hospital surveillance data does not currently link mortality data and microbiology results. Mortality surveillance spanned 12-months pre-implementation and the full 18-month FLIP implementation period.

A timeline of FLIP implementation and surveillance periods for process and outcome metrics is summarized in Fig. 3.

Statistical Analysis

We compared ward-aggregated outcome metrics on colonization, BSIs, and mortality across study periods – pre-FLIP, pilot phase, continuation phase, and both pilot plus continuation phases – using two-proportion z-tests for binary outcomes and Chi-square tests for categorical variables. We evaluated colonization based on detection of ESCrE, CRE, Acinetobacter, or any MDRO. Additionally, we assessed colonization by site of detection (skin vs perirectal) to account for differential effects of the interventions on site-specific colonization. The median number of monthly admissions during the study periods were compared using the Mann-Whitney-U test.

Ethics statement

This study was approved by the Institutional Review Boards at the University of Botswana, the Health Research and Development Committee at Botswana’s Ministry of Health (2022-HPDME18/13/1), Stellenbosch University (HREC#: S23-04-085), and the healthcare facility where this study was carried out (2022-2/2A (7)/201). FLIP orientation was voluntary for families and given that the study relied on a review of ward surveillance data, a waiver of written individual informed consent from families was granted. All process metric data were collected via paper form and then immediately entered on password-protected electronic spreadsheets.

Implementation of the FLIP bundle

During the 30-month surveillance period (August 2022 – January 2025) there were 2,935 admissions, 262 BSIs, and 373 deaths (Figure 4). A total of 1,866 neonates were admitted during the FLIP implementation period (Aug 2023-Jan 2025); 62.2% (n=1,161) of families received the FLIP orientation. The median number of monthly admissions was 88.5 in the pre-FLIP period, and 106.5 during the FLIP implementation period (p<0.001).

Process metric performance and bundle revision

The proportion of admitted neonates receiving any human milk remained stable between the pre-FLIP (66.3%) and pilot (66.0%) periods (Supplemental Figure 1). During the continuation phase—when monitoring feeding practices shifted to focus on the first 72 hours—80% of mothers breastfeed or expressed milk. Time to first breastfeeding was tracked in the continuation phase, averaging 2 hours for vaginal births (range 1–22 hours) and 9 hours for Caesarean births (range 1–23 hours), with a significant downward trend among vaginal deliveries (p<0.001) (Supplemental Figure 3).

Among mothers providing skin-to-skin contact to admitted infants, the median recorded duration was 3.75 hours/day in the pre-FLIP period and 3.4 hours/day during the pilot. In the continuation phase—which focused on practices in the first 72 hours—neonates received a median of only one hour/day of skin-to-skin contact, indicating that immediate and continuous contact was rarely practiced during this critical window (Supplemental Figure 2).

Hand hygiene adherence remained stable (74.9% pre-FLIP vs. 75.3% during FLIP; p=0.788). CHG cleansing improved from 55.3% in the pilot phase to 94.6% during the continuation phase (p<0.001), after staff began assisting families unable to perform the task (Supplemental Figure 5).

Impact of FLIP on colonization prevalence

FLIP implementation was not associated with a sustained change in colonization with MDRO. Although we observed a 6.8% decrease in Acinetobacter spp. colonization in the pilot phase (p=0.008), increases in both ESCrE colonization and Acinetobacter spp. were observed in the continuation phase, resulting in no sustained changes or, in the case of skin ESCrE colonization, a 6.7% increase overall (p=0.004). CRE colonization remained low throughout the study period, with no significant changes. The temporal impact of the FLIP intervention on outcome metrics are summarized in Table 1 and Figure 5.

Impact of FLIP on bloodstream infection incidence and outbreaks

During the surveillance period, three outbreaks were identified, all of which were caused by Klebsiella pneumoniae (Figure 4 and Supplementary Figure 6). One occurred during the pre-FLIP phase (41 cases in a 9-week period), and the second and third occurring during the FLIP continuation phase (39 cases in a 6-week period and 5 cases in a 4-week period). In the two largest outbreaks, epidemiologic investigations conducted by ward staff confirmed that contaminated intravenous fluid bags, which are accessed multiple times for intravenous line flushes and reconstitution of medication administered to multiple patients, were likely major drivers of transmission. Intensified efforts to curb these practices through routine audits of unlabeled and expired intravenous fluids and medications were introduced in the continuation phase. We observed a non-significant reduction in BSI incidence (11.0% to 9.3%; p=0.15) coinciding with the FLIP implementation period. Notably, BSI incidence declined significantly in the pilot period from 11.0% to 6.1% (p<0.001). However, this progress was reversed during the continuation period, possibly due the K. pneumoniae outbreak associated with contaminated intravenous fluids, although a sub-analysis of non-outbreak-associated BSIs before and after FLIP implementation showed no significant differences (Table 1 and Figure 5).

Impact of FLIP on all-cause mortality

Following FLIP bundle implementation, a sustained and significant reduction of in-hospital mortality was observed, (15.4% pre-FLIP to 11.1% during FLIP; 28% relative reduction; p<0.001). Phase-specific analyses revealed an initial decline in mortality during the pilot phase from 15.4% to 11.3% (p=0.02), followed by a sustained reduction to 11.1% during the continuation phase (p=0.002).

To our knowledge, this is the first IPC bundle piloted in a resource-limited setting that specifically engaged families in preventing infections among hospitalized neonates. Although the FLIP bundle did not significantly reduce BSIs or MDRO colonization, we demonstrated the potential of systematically engaging families in neonatal IPC programs, with a notable improvement in in-hospital survival. The observed reduction in mortality suggests that even when colonization and infection are not fully prevented, components of the bundle—such as improved access to human milk —may enhance survival, possibly via increased immune protection. For example, there is evidence showing that even among newborns with sepsis, human milk feeding is associated with a reduced need for vasoactive medications.³² These findings underscore the indirect benefits of family-centered care, where practices supporting immunity and mitigating pathogen transmission can improve health outcomes.

While the observed decrease in BSI incidence was modest and not statistically significant, most BSIs were noted to occur during ward outbreaks linked to contaminated intravenous fluids. Outbreaks in neonatal units linked to contaminated intravenous fluids and medication are common in resource-limited settings.^33–35 This highlights that family engagement, while beneficial, must complement robust healthcare worker–led IPC measures, especially those targeting procedural safety and medication and intravenous fluid handling.

Family engagement in IPC did not significantly alter MDRO colonization which may reflect global trends toward increasing colonization in both hospitals and communities which may not be overcome by family-based interventions alone.³⁶ Maternal MDRO colonization, which poses a risk for subsequent neonatal colonization, has been shown to be high in both community and hospital settings in LMICs.³⁷ Additionally, seasonal surges of human MDRO colonization have been observed during wet seasons.^38–40 Thus, it is possible that high neonatal colonization with MDROs in this setting is inevitable despite FLIP bundle implementation. Additionally, the optimal frequency and dosing of CHG cleansing as a decolonization measure remain uncertain, particularly in reducing gram-negative bacterial colonization; more frequent applications may have yielded better results in reducing MDRO colonization.²²

Study Limitations

This study has several limitations. First, some of the process and outcome metrics were based on point prevalence surveys and maternal recall, potentially missing temporal variability and introducing recall bias. Additionally, MDRO colonization was assessed using chromogenic media, which may lack specificity compared to automated antimicrobial susceptibility testing.⁴¹ Because colonization among families, staff, or patients in other wards was not assessed during this period, it is difficult to determine how local colonization pressure may have influenced the results. Mortality analyses could not distinguish infection-related deaths from those due to other causes, limiting interpretation of survival benefits. Moreover, using temporal associations to assess the bundle’s impact risks conflating improvements with unrelated concurrent changes, such as changes in clinical practice or IPC protocol adherence, and thus the association between FLIP implementation and improved survival cannot be considered causal. For example, we noted an anecdotal increase in patient transfers to private facilities for ventilator support during the FLIP implementation period, which may have reduced the number of deaths occurring in this hospital independent of the FLIP bundle intervention. An increase in median monthly admissions during the FLIP period (88.5 to 106.5) may have affected results by inflating the denominator of patients at risk or conversely by obscuring the bundle’s impact through additional overcrowding in the implementation period, which is known to increase transmission and mortality risk. A future analysis of the FLIP intervention should take advantage of a multicenter and/or a cluster randomized study design to better understand the precise effect that FLIP has on outcome metrics.

The FLIP bundle also faced challenges in uptake, reflected by only 62% of families receiving the FLIP orientation. This suboptimal uptake was driven by staff shortages; we relied on staff-led orientations, which were unavailable on weekends and evenings. Consequently, some families received orientation several days after admission, missing critical opportunities to establish early lactation and skin-to-skin care. In light of nurse shortages in LMICs, future implementations of the FLIP bundle should consider task-shifting orientation activities to trained healthcare auxiliaries.⁴² Sustained improvements in skin-to-skin contact were not observed, resulting from several barriers in our site such as limited space, lack of privacy, and competing clinical priorities which may similarly apply in other settings.⁴³ In settings which cannot safely support rooming-in, broad advocacy is needed with hospital administrators and policy makers to ensure space for a mother to sit beside each cot/incubator, facilitating extended family visiting hours to enable prolonged intermittent skin-to-skin contact and frequent breastfeeding. Although breastfeeding rates were higher in our study than in many settings, rates of early initiation of breastfeeding, which has been showed to decrease rates of sepsis and mortality,⁴⁴ were low, particularly among infants born via Caesarean delivery, highlighting the importance of antepartum and early postpartum lactation support.

Implementation of all bundle elements was further hindered by resource limitations—with a stock-out of CHG experienced in the last three months of the FLIP implementation period.

Resource constraints also undermined implementation. For example, lack of human milk storage facilities and supplies may have undermined mothers’ ability to establish adequate milk supply. Hospitals looking to implement a FLIP bundle should consider the cost of essential consumables needed for successful implementation; at the very least this would include hand hygiene products and containers for expressing human milk that are not shared between patients. Breast pumps, while posing a challenge for cleaning and disinfection for some facilities, can facilitate adequate milk expression in mothers of preterm infants when hand expression is not sufficient.

This pilot implementation of the FLIP bundle demonstrates the promise of systematically engaging families in neonatal IPC programs, with encouraging improvements in survival among hospitalized infants. The lack of significant reductions in BSI incidence and MDRO colonization highlights the need for iterative refinements and consideration of additional strategies to explore how units can more consistently engage families and better implement the individual components of the family engagement in infection prevention to ensure its optimal effect. Future studies should employ more rigorous study designs, such as multi-centered and a cluster-randomized approaches, to establish a more direct causal link between the FLIP bundle and improved sepsis-related mortality.

FUNDING:

This study received funding from the Children’s Hospital of Philadelphia Global Health Pilot Grant and CDC.

Author Contribution

CVT, J S , CM & S C: ConceptualizationCVT, JS, TG, TN, TZ , KR and KL : Data Curation:KL , JS and CVT: Formal AnalysisJS,SP,SM and EL: Funding Acquisition C V T, BG, TG, TN,TZ,KR and KL: Investigation JS Supervision CVT and JS: Writing – Original Draft Preparation: All authors: Writing -Review & Editing

Acknowledgement

The authors gratefully acknowledge the contributions of hospital staff, local and international stakeholders, and families who contributed to the development, iterative revision of the FLIP bundle. We also thank Ashley Styczynski and Gemma Parra from the Division of Healthcare Quality & Promotion at the U.S. Centers for Disease Control and Prevention (CDC) for their technical input and support during the design and implementation of this project.

Liu L, Oza S, Hogan D, et al. Global, regional, and national causes of child mortality in 2000-13, with projections to inform post-2015 priorities: an updated systematic analysis. Lancet Jan. 2015;31(9966):430–40. 10.1016/S0140-6736(14)61698-6.
Wen SCH, Ezure Y, Rolley L, et al. Gram-negative neonatal sepsis in low- and lower-middle-income countries and WHO empirical antibiotic recommendations: A systematic review and meta-analysis. PLoS Med. Sep 2021;18(9):e1003787. 10.1371/journal.pmed.1003787.
Licona G, Ballot D, Moon TD, et al. Multidrug-Resistant Bacterial Infections Among Very Low Birthweight Infants With Late-Onset Sepsis in Johannesburg, South Africa. Open Forum Infect Dis Aug. 2023;10(8):ofad362. 10.1093/ofid/ofad362.
Strysko J, Machiya T, Lechiile K, et al. Carbapenem-resistant Acinetobacter baumannii at a tertiary-care hospital in Botswana: Focus on perinatal environmental exposures. Antimicrob Stewardship Healthc Epidemiol. 2022;2(S1):S79–79. 10.1017/ash.2022.206.
Gezmu AM, Bulabula ANH, Dramowski A, et al. Laboratory-confirmed bloodstream infections in two large neonatal units in sub-Saharan Africa. Int J Infect Dis Nov. 2020;20:103:201–7. 10.1016/j.ijid.2020.11.169.
Mudzikati L. DA. Neonatal septicaemia: prevalence and antimicrobial susceptibility patterns of common pathogens at Princess Marina Hospital, Botswana. S Afr J Infect Dis. 2015;30:108–133.
Flokas ME, Karanika S, Alevizakos M, Mylonakis E. Prevalence of ESBL-Producing Enterobacteriaceae in Pediatric Bloodstream Infections: A Systematic Review and Meta-Analysis. PLoS ONE. 2017;12(1):e0171216. 10.1371/journal.pone.0171216.
Chan GJ, Lee AC, Baqui AH, Tan J, Black RE. Risk of early-onset neonatal infection with maternal infection or colonization: a global systematic review and meta-analysis. PLoS Med Aug. 2013;10(8):e1001502. 10.1371/journal.pmed.1001502.
Murthy S, Godinho MA, Guddattu V, Lewis LES, Nair NS. Risk factors of neonatal sepsis in India: A systematic review and meta-analysis. PLoS ONE. 2019;14(4):e0215683. 10.1371/journal.pone.0215683.
Samanta S, Farrer K, Breathnach A, Heath PT. Risk factors for late onset gram-negative infections: a case-control study. Arch Dis Child Fetal Neonatal Ed Jan. 2011;96(1):F15–8. 10.1136/adc.2009.169540.
Vogel JP, Lee AC, Souza JP. Maternal morbidity and preterm birth in 22 low- and middle-income countries: a secondary analysis of the WHO Global Survey dataset. BMC Pregnancy Childbirth Jan. 2014;31:14:56. 10.1186/1471-2393-14-56.
OuYang X, Yang CY, Xiu WL, Hu YH, Mei SS, Lin Q. Oropharyngeal administration of colostrum for preventing necrotizing enterocolitis and late-onset sepsis in preterm infants with gestational age = 32 weeks: a pilot single-center randomized controlled trial</at. Int Breastfeed J Aug. 2021;21(1):59. 10.1186/s13006-021-00408-x.
Arya S, Chhabra S, Singhal R, et al. Effect on neonatal sepsis following immediate kangaroo mother care in a newborn intensive care unit: a post-hoc analysis of a multicentre, open-label, randomised controlled trial. EClinicalMedicine Jun. 2023;60:102006. 10.1016/j.eclinm.2023.102006.
Group WHOIKS, Arya S, Naburi H, et al. Immediate Kangaroo Mother Care and Survival of Infants with Low Birth Weight. N Engl J Med May. 2021;27(21):2028–38. 10.1056/NEJMoa2026486.
Casper C, Sarapuk I, Pavlyshyn H. Regular and prolonged skin-to-skin contact improves short-term outcomes for very preterm infants: A dose-dependent intervention. Arch Pediatr Nov. 2018;25(8):469–75. 10.1016/j.arcped.2018.09.008.
Pavlyshyn H, Sarapuk I, Saturska U. The impact of skin-to-skin contact upon stress in preterm infants in a neonatal intensive care unit. Front Pediatr. 2024;12:1467500. 10.3389/fped.2024.1467500.
Abouelfettoh A, Ludington-Hoe SM, Burant CJ, Visscher MO. Effect of skin-to-skin contact on preterm infant skin barrier function and hospital-acquired infection. J Clin Med Res Feb. 2011;12(1):36–46. 10.4021/jocmr479w.
Eckermann HA, Meijer J, Cooijmans K, Lahti L, de Weerth C. Daily skin-to-skin contact alters microbiota development in healthy full-term infants. Gut Microbes Jan-Dec. 2024;16(1):2295403. 10.1080/19490976.2023.2295403.
Parnanen KMM, Hultman J, Markkanen M, et al. Early-life formula feeding is associated with infant gut microbiota alterations and an increased antibiotic resistance load. Am J Clin Nutr Feb. 2022;9(2):407–21. 10.1093/ajcn/nqab353.
Vurayai M, Strysko J, Kgomanyane K, et al. Characterizing the bioburden of ESBL-producing organisms in a neonatal unit using chromogenic culture media: a feasible and efficient environmental sampling method. Antimicrob Resist Infect Control Jan. 2022;24(1):14. 10.1186/s13756-021-01042-2.
Dramowski A, Aucamp M, Bekker A, et al. NeoCLEAN: a multimodal strategy to enhance environmental cleaning in a resource-limited neonatal unit. Antimicrob Resist Infect Control Feb. 2021;12(1):35. 10.1186/s13756-021-00905-y.
Dramowski A, Pillay S, Bekker A, et al. Impact of 1% chlorhexidine gluconate bathing and emollient application on bacterial pathogen colonization dynamics in hospitalized preterm neonates - A pilot clinical trial. EClinicalMedicine Jul. 2021;37:100946. 10.1016/j.eclinm.2021.100946.
Sharma A, Kulkarni S, Thukral A, et al. Aqueous chlorhexidine 1% versus 2% for neonatal skin antisepsis: a randomised non-inferiority trial. Arch Dis Child Fetal Neonatal Ed Nov. 2021;106(6):643–8. 10.1136/archdischild-2020-321174.
Molina Garcia A, Cross JH, Fitchett EJA, et al. Infection prevention and care bundles addressing health care-associated infections in neonatal care in low-middle income countries: a scoping review. EClinicalMedicine Feb. 2022;44:101259. 10.1016/j.eclinm.2021.101259.
Tran HT, Mannava P, Murray JCS, et al. Early Essential Newborn Care Is Associated With Reduced Adverse Neonatal Outcomes in a Tertiary Hospital in Da Nang, Viet Nam: A Pre- Post- Intervention Study. EClinicalMedicine Dec. 2018;6:51–8. 10.1016/j.eclinm.2018.12.002.
Johnson J, Akinboyo IC, Schaffzin JK. Infection Prevention in the Neonatal Intensive Care Unit. Clin Perinatol Jun. 2021;48(2):413–29. 10.1016/j.clp.2021.03.011.
Mahtab S, Madhi SA, Baillie VL, et al. Causes of death identified in neonates enrolled through Child Health and Mortality Prevention Surveillance (CHAMPS), December 2016 -December 2021. PLOS Glob Public Health. 2023;3(3):e0001612. 10.1371/journal.pgph.0001612.
Nasa P, Jain R, Juneja D. Delphi methodology in healthcare research: How to decide its appropriateness. World J Methodol Jul. 2021;20(4):116–29. 10.5662/wjm.v11.i4.116.
Kgosiemang T, Motzafi-Haller P. Understanding Early Pregnancies: Sociocultural Factors of Teenage Childbearing in Gaborone, Botswana. Qual Health Res Dec. 2021;31(14):2641–52. 10.1177/10497323211041978.
WHO Guidelines on Hand Hygiene in Health Care: First Global Patient Safety Challenge Clean Care Is Safer Care. Geneva: World Health Organization. 2009. II, CONSENSUS RECOMMENDATIONS. Available from: https://www.ncbi.nlm.nih.gov/books/NBK144035/
Centers for Disease Control and Prevention. Common Commensals (CC). Retrieved from https://www.cdc.gov/nhsn/xls/master-organism-com-commensals-lists.xlsx
Moliner-Calderon E, Verd S, Leiva A, Ginovart G, Moll-McCarthy P, Figueras-Aloy J. The role of human milk feeds on inotrope use in newborn infants with sepsis. Front Pediatr. 2023;11:1172799. 10.3389/fped.2023.1172799.
Moore KL, Kainer MA, Badrawi N, et al. Neonatal sepsis in Egypt associated with bacterial contamination of glucose-containing intravenous fluids. Pediatr Infect Dis J Jul. 2005;24(7):590–4. 10.1097/01.inf.0000168804.09875.95.
Moodley P, Coovadia YM, Sturm AW. Intravenous glucose preparation as the source of an outbreak of extended-spectrum beta-lactamase-producing Klebsiella pneumoniae infections in the neonatal unit of a regional hospital in KwaZulu-Natal. S Afr Med J Nov. 2005;95(11):861–4.
Okomo U, Senghore M, Darboe S, et al. Investigation of sequential outbreaks of Burkholderia cepacia and multidrug-resistant extended spectrum beta-lactamase producing Klebsiella species in a West African tertiary hospital neonatal unit: a retrospective genomic analysis. Lancet Microbe Jul. 2020;1(3):e119–29. 10.1016/S2666-5247(20)30061-6.
Karanika S, Karantanos T, Arvanitis M, Grigoras C, Mylonakis E. Fecal Colonization With Extended-spectrum Beta-lactamase-Producing Enterobacteriaceae and Risk Factors Among Healthy Individuals: A Systematic Review and Metaanalysis. Clin Infect Dis Aug. 2016;1(3):310–8. 10.1093/cid/ciw283.
Robinson ML, Johnson J, Naik S, et al. Maternal Colonization Versus Nosocomial Transmission as the Source of Drug-Resistant Bloodstream Infection in an Indian Neonatal Intensive Care Unit: A Prospective Cohort Study. Clin Infect Dis Jul. 2023;5(Suppl 1):S38–45. 10.1093/cid/ciad282.
Mannathoko N, Mosepele M, Gross R, et al. Colonization with extended-spectrum cephalosporin-resistant Enterobacterales (ESCrE) and carbapenem-resistant Enterobacterales (CRE) in healthcare and community settings in Botswana: an antibiotic resistance in communities and hospitals (ARCH) study. Int J Infect Dis Sep. 2022;122:313–20. 10.1016/j.ijid.2022.06.004.
Cocker D, Chidziwisano K, Mphasa M, et al. Investigating One Health risks for human colonisation with extended spectrum beta-lactamase-producing Escherichia coli and Klebsiella pneumoniae in Malawian households: a longitudinal cohort study. Lancet Microbe Jul. 2023;4(7):e534–43. 10.1016/S2666-5247(23)00062-9.
Bulabula ANH, Dramowski A, Mehtar S. Maternal colonization or infection with extended-spectrum beta-lactamase-producing Enterobacteriaceae in Africa: A systematic review and meta-analysis. Int J Infect Dis Nov. 2017;64:58–66. 10.1016/j.ijid.2017.08.015.
Mannathoko N, Lautenbach E, Mosepele M, et al. Performance of CHROMagar ESBL media for the surveillance of extended-spectrum cephalosporin-resistant Enterobacterales (ESCrE) from rectal swabs in Botswana. J Med Microbiol Nov. 2023;72(11). 10.1099/jmm.0.001770.
Coales K, Jennings H, Afaq S, et al. Perspectives of health workers engaging in task shifting to deliver health care in low-and-middle-income countries: a qualitative evidence synthesis. Glob Health Action Dec. 2023;31(1):2228112. 10.1080/16549716.2023.2228112.
Cai Q, Chen DQ, Wang H, et al. What influences the implementation of kangaroo mother care? An umbrella review. BMC Pregnancy Childbirth Nov. 2022;18(1):851. 10.1186/s12884-022-05163-3.
Debes AK, Kohli A, Walker N, Edmond K, Mullany LC. Time to initiation of breastfeeding and neonatal mortality and morbidity: a systematic review. BMC Public Health. 2013;13(Suppl 3):S19. 10.1186/1471-2458-13-S3-S19.
Wambach K, Spencer B. Breastfeeding and human lactation. Sixth edition. ed. Jones & Bartlett Learning; 2021:p.
Botswana Integrated. HIV Clinical Care Guidelines (2023).
WHO. HIV/AIDS: Infant feeding and nutrition. 12. Feb, 2024. https://www.who.int/news-room/questions-and-answers/item/hiv-aids-infant-feeding-and-nutrition
Guideline. for Vertical Transmission Prevention of Communicable Infections (2023).
Mizuno K, Ueda A. The maturation and coordination of sucking, swallowing, and respiration in preterm infants. J Pediatr Jan. 2003;142(1):36–40. 10.1067/mpd.2003.mpd0312.
Russell N, Clements MN, Azmery KS, et al. Safety and efficacy of whole-body chlorhexidine gluconate cleansing with or without emollient in hospitalised neonates (NeoCHG): a multicentre, randomised, open-label, factorial pilot trial. EClinicalMedicine Mar. 2024;69:102463. 10.1016/j.eclinm.2024.102463.
Chapman AK, Aucott SW, Milstone AM. Safety of chlorhexidine gluconate used for skin antisepsis in the preterm infant. J Perinatol Jan. 2012;32(1):4–9. 10.1038/jp.2011.148.

No competing interests reported.

Download PDF

Editorial decision: Revision requested
09 Nov, 2025
Reviews received at journal
08 Nov, 2025
Reviews received at journal
07 Nov, 2025
Reviews received at journal
05 Nov, 2025
Reviewers agreed at journal
14 Oct, 2025
Reviews received at journal
11 Oct, 2025
Reviewers agreed at journal
09 Oct, 2025
Reviewers agreed at journal
09 Oct, 2025
Reviewers agreed at journal
09 Oct, 2025
Reviewers invited by journal
09 Oct, 2025
Editor assigned by journal
05 Oct, 2025
Submission checks completed at journal
26 Sep, 2025
First submitted to journal
23 Sep, 2025

You are reading this latest preprint version

Developing and implementing a family-led infection prevention bundle for hospitalized neonates

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

PLAIN LANGUAGE SUMMARY

BACKGROUND

METHODS

Study design and setting

Intervention: The FLIP Bundle

Involvement of subject matter experts

Involvement of local stakeholders

Involvement of families

FLIP bundle components

Process Metrics

Outcome metrics

Statistical Analysis

Ethics statement

RESULTS

Discussion

Study Limitations

Conclusion

Declarations

FUNDING:

Author Contribution

Acknowledgement

References

Additional Declarations

Supplementary Files

Status:

Version 1