The People at the Cotside: How Nursing Care Shapes Infection Risk for the Smallest Babies

What the research really shows about staffing, handwashing, lines, skin care and breast milk in the care of babies born before 28 weeks

Infection is one of the biggest dangers facing a baby born before 28 weeks of pregnancy, and much of what protects them happens in the ordinary, hourly work of nursing. The strongest evidence points somewhere many families would not expect: how many nurses are on shift, and how heavy their workload is, is measurably linked to how often these babies get infections. This article looks honestly at what bedside care can and cannot change.

Why This Question Matters

A baby born at 24 or 25 weeks arrives with skin barely thicker than tissue paper, an immune system that has not finished assembling, and an urgent need for fluids and nutrition delivered through a fine tube threaded into a large vein. Every one of those tubes is also a possible doorway for bacteria. For decades, hospitals treated infection prevention mainly as a question of which products to use; the idea that the organisation of care mattered had to be proved.

The first convincing proof came from 54 hospitals in California working together between 2002 and 2006. Twenty-seven joined a shared quality-improvement programme; twenty-seven did not. Across the whole group, the infection rate in babies weighing under 1,500 grams at birth fell from about 17 in every 100 to about 14, and simply attending the collaborative's meetings was linked to roughly a 19% lower chance of infection [1]. That was encouraging, but it did not say what the participating hospitals had changed — and separating the poster on the wall from the person at the incubator has occupied researchers ever since. Formal guidance, when it arrived, was strikingly cautious: the United States Centers for Disease Control and Prevention published a guideline on bloodstream infections from central lines in newborn intensive care in February 2022, and for several of its questions concluded the evidence was not good enough to make a recommendation at all [2].

This article is Part 6 of the 6-part series "Infectious Diseases in the Extremely Preterm Infant". Part 1 covered how these infections arise and how common they are, Part 2 what they do to the body's tissues, Part 3 how they can be prevented, Part 4 how they are treated, and Part 5 what they mean for a child's future. This final part is about the bedside.

One honest limitation. Almost none of the research below studied only babies born before 28 weeks; most studied all babies under 1,500 grams, or everyone in the unit. The findings are the best available guide, but not a perfect match for the smallest babies.

The Number of Nurses Is Part of the Medicine

The clearest and most consistent finding in this field concerns staffing. A large American study using data from 67 newborn intensive care units compared the nursing each baby actually received against national guidelines for how much they should have received. In one study year, 47 out of every 100 babies received less nursing than the guidelines specify — and for the sickest babies, 80 in every 100. As understaffing worsened, infections rose steadily: from a predicted 9 in 100 babies in fully staffed conditions, to 14 in 100 at typical understaffing, to 21 in 100 in the most stretched units. Closing that gap would take roughly one extra nurse for every nine babies [3].

A study in Vienna, Austria, tracked how much work each shift actually demanded compared with how many nurses were rostered. When the workload passed the point where an extra nurse was needed, the odds of a baby developing a bloodstream infection roughly doubled. This study is unusually relevant to families of the very smallest babies: the infants who became infected had an average gestational age of about 25 and a half weeks [4]. Two further studies from Quebec, Canada, found the same pattern — more nursing per baby was protective, a fuller unit was riskier, and the overtime nurses worked over the preceding three days predicted infections the following day [5], [6]. Four studies, three countries, one message. So if you hear nurses on your baby's unit talking about ratios, occupancy or overtime, they are not only discussing working conditions. They are discussing infection risk.

Lines, Tubes and the Value of Asking "Does This Still Need to Be There?"

Central lines save these babies' lives and also carry the greatest infection risk. Hospitals manage them using a "bundle" — a fixed checklist performed every time. One unit in Beirut, Lebanon, published its checklist in full: hand hygiene, full sterile gowns and drapes, all equipment prepared before starting, skin cleaning, a written checklist, and — perhaps most important — explicit permission for any staff member to stop a non-urgent procedure on seeing sterile technique broken. Maintaining the line adds a daily check of whether it is still needed at all. Infections fell from about 5 per 1,000 line-days to about 1, and the unit went 23 consecutive months with none [7]. Pooling 24 such studies suggests bundles cut these infections by around 60% — but none of those 24 studies was a randomized trial, the strongest form of medical evidence; most simply compared a unit before and after [8].

How long a line should stay in is genuinely unsettled. A Swiss hospital following 615 babies found every infection occurred after the line had been in place for at least five days, with risk climbing after nine [9]. But the one randomized trial here — a single-hospital study comparing umbilical lines left for 8 to 14 days with removal within a week — found no significant difference in complications — and the early-removal group needed more replacement lines and more X-rays [10]. Reasonable units make different choices, and it is fair to ask your baby's team what theirs is and why.

Skin: Where the Smallest Babies Are Most Vulnerable

This is where extremely premature babies differ most from everyone else, and where research has served them worst. A review gathering every published report of skin injury from antiseptics used before procedures found that 98 out of every 100 injuries happened in extremely premature babies, most often from chlorhexidine, the standard skin-cleaning agent, ranging from redness through skin breakdown to chemical burns [11]. Yet the trials testing whether these antiseptics are safe have generally excluded the very babies at risk: a recent trial in South Africa and Bangladesh, designed to test chlorhexidine safety, enrolled only babies of 28 weeks or more [12]. A British feasibility trial comparing two chlorhexidine preparations in babies born before 34 weeks found no major skin injuries, which is reassuring as far as it goes [13]. Nurses caring for the smallest babies therefore balance two real harms — infection on one side, skin damage on the other — with less evidence than they deserve. Bathing is gentler ground: pooling 11 trials in 828 premature babies showed that a "swaddle bath", where the baby stays wrapped, keeps them warmer, calmer and breathing more steadily than an ordinary tub bath, though these studies did not measure infection [14].

Two Results That Deserve Honesty

Not everything intuitive turns out to work. A hand-hygiene campaign in a 132-bed newborn unit in Cape Town, South Africa — training, posters, personal bottles of hand gel, parent leaflets and performance feedback — watched 2,430 opportunities to clean hands. Compliance was about 62% before and about 62% afterwards, hand gel use did not change, and infection rates did not fall [15]. This does not mean handwashing is unimportant; it means campaigns alone do not change deeply worn habits, and units need to change systems, not just remind people.

Similarly, Family Integrated Care — a model in which parents are trained to give much of their baby's daily care — was tested in a randomized trial across 25 hospitals in Canada, Australia and New Zealand. Babies gained weight better and parents were less stressed, both real gains, but there was no difference in deaths or serious complications, and the trial demonstrated no reduction in infection [16]. A smaller American study of a similar model did find fewer infections where parents took part in ward rounds, but the result was statistically borderline [17]. Being present and involved is good for your baby and good for you; it is not yet proven to prevent infection.

The Nurse Who Notices, and the Milk You Provide

One of the most valuable things in the unit cannot be bought. In a survey of 181 nurses at an American academic hospital, 73% reported having suspected a baby was becoming septic before any test confirmed it — often from behavioural changes such as unusual sleepiness, irritability or floppiness rather than from monitor readings. That ability tracked experience in newborn intensive care, not formal qualifications [18]. Units increasingly formalise this instinct into "trigger tools" so a nurse's concern reliably starts a fast response; one such tool cut the average time to the first dose of antibiotics from 126 minutes to 102, with no cases missed [19]. If a nurse tells you they are "not happy" with your baby today, that is a clinical observation with real evidence behind it.

Your milk matters enormously, and handling it follows strict rules — fresh milk refrigerated for no more than 48 hours, thawed milk 24 hours, feeds hanging no longer than four hours, and all preparation in a dedicated clean room, never at the bedside [20]. These rules exist for good reason: a study of 118 mothers found nearly 90% of expressed milk samples carried bacteria, with delayed refrigeration and hand recontamination when turning taps off among the culprits [21]. Nothing here is anyone's fault; it is why hospitals ask for particular washing and storage steps. Technology helps: one American children's hospital cut milk-handling errors from about 97 to about 11 per 1,000 bottles using bedside barcode scanning and a dedicated milk technician [22]. And the amount matters: in 175 very small babies, every extra 10 millilitres of mother's milk per kilogram per day reduced the odds of sepsis by 19% [23]. Early drops of colostrum placed inside the baby's cheeks — typically 0.4 mL every three hours for seven days — are given for the same reason; a recent Iranian trial found fewer infections in the colostrum group, though the difference was not statistically certain [24]. In[Neo]Sight covers this in its dedicated cluster on oropharyngeal colostrum.

What Researchers Are Working On Next

Nurses themselves have driven much of the recent progress. A collaboration of 12 Brazilian units cut serious late infections from 27% to 22% of very small babies over three years — real improvement, but only half the units met their target and four got worse, and the units differed enormously, from one nurse supervising 4 beds to one supervising 30 [25]. A review of 29 such programmes, pooling 21 of them across more than 29,000 babies, suggests they roughly halve the odds of late infection — including in babies born under 1,000 grams, the closest group to the very smallest [26]. The field still needs a proper trial of staffing levels, a trial of skin antiseptics designed for babies born before 28 weeks, and real evidence for the sterile-handling techniques nurses are trained in but which have never been tested in this group. Until those exist, the most useful thing a family can know is that the quiet, repetitive, unglamorous work at the cotside — the hand rub, the daily question of whether a line is still needed, the careful milk label, the nurse who notices — is not background to your baby's care. It is a substantial part of the treatment.

References

  1. Wirtschafter DD, Powers RJ, Pettit JS, et al. Nosocomial infection reduction in VLBW infants with a statewide quality-improvement model. Pediatrics. 2011;127(3):419–426. doi:10.1542/peds.2010-1449 ↩
  2. Centers for Disease Control and Prevention; Healthcare Infection Control Practices Advisory Committee. Guideline for Prevention of Infections in Neonatal Intensive Care Unit Patients: Central Line-associated Blood Stream Infections. Atlanta, GA: CDC; February 2022. 37 pp. https://www.cdc.gov/infection-control/media/pdfs/Guideline-NICU-CLABSI-508.pdf ↩
  3. Rogowski JA, Staiger D, Patrick T, et al. Nurse staffing and NICU infection rates. JAMA Pediatr. 2013;167(5):444–450. doi:10.1001/jamapediatrics.2013.18 ↩
  4. Küng E, Waldhör T, Rittenschober-Böhm J, Berger A, Wisgrill L. Increased nurse workload is associated with bloodstream infections in very low birth weight infants. Sci Rep. 2019;9:6331. doi:10.1038/s41598-019-42685-x ↩
  5. Beltempo M, Patel S, Platt RW, et al. Association of nurse staffing and unit occupancy with mortality and morbidity among very preterm infants: a multicentre study. Arch Dis Child Fetal Neonatal Ed. 2023;108(4):387–393. doi:10.1136/archdischild-2022-324414 ↩
  6. Fazio M, Jabbour E, Patel S, et al. Association of shift-level organizational factors with nosocomial infection in the neonatal intensive care unit. J Pediatr Clin Pract. 2024;13:200112. doi:10.1016/j.jpedcp.2024.200112 ↩
  7. Al Bizri A, Hanna Wakim R, Obeid A, et al. A quality improvement initiative to reduce central line-associated bloodstream infections in a neonatal intensive care unit in a low- and middle-income country. BMJ Open Qual. 2023;12(2):e002129. doi:10.1136/bmjoq-2022-002129 ↩
  8. Payne V, Hall M, Prieto J, Johnson M. Care bundles to reduce central line-associated bloodstream infections in the neonatal unit: a systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed. 2018;103(5):F422–F429. doi:10.1136/archdischild-2017-313362 ↩
  9. Buttera M, Luhmann-Lunt C, Buettcher M, et al. The impact of central line-specific dwell-times for neonatal central line associated bloodstream infections. Antimicrob Resist Infect Control. 2025;14:106. doi:10.1186/s13756-025-01618-2 ↩
  10. Hess S, Poryo M, Ruckes C, et al. Assessment of an umbilical venous catheter dwell-time of 8–14 days versus 1–7 days in very low birth weight infants (UVC — You Will See): a pilot single-center, randomized controlled trial. Early Hum Dev. 2023;179:105752. doi:10.1016/j.earlhumdev.2023.105752 ↩
  11. Mulinda C, Suhail S, Sutherland B, Lauren CT, Hunt RD. Pre-procedural topical antisepsis in the neonate: a systematic review evaluating risk factors for skin injury. Pediatr Dermatol. 2025;42(1):31–40. doi:10.1111/pde.15773 ↩
  12. 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. 2024;69:102463. doi:10.1016/j.eclinm.2024.102463 ↩
  13. Clarke P, Soe A, Nichols A, et al. 2% chlorhexidine gluconate aqueous versus 2% chlorhexidine gluconate in 70% isopropyl alcohol for skin disinfection prior to percutaneous central venous catheterisation: the ARCTIC randomised controlled feasibility trial. Arch Dis Child Fetal Neonatal Ed. 2024;109(2):202–210. doi:10.1136/archdischild-2023-325871 ↩
  14. Sun X, Xu J, Zhou R, et al. Effectiveness of different bathing methods on physiological indexes and behavioral status of preterm infants: a systematic review and meta-analysis. BMC Pediatr. 2023;23:507. doi:10.1186/s12887-023-04280-y ↩
  15. Dramowski A, Erasmus LM, Aucamp M, et al. SafeHANDS: a multimodal hand hygiene intervention in a resource-limited neonatal unit. Trop Med Infect Dis. 2023;8(1):27. doi:10.3390/tropicalmed8010027 ↩
  16. O'Brien K, Robson K, Bracht M, et al.; FICare Study Group and FICare Parent Advisory Board. Effectiveness of Family Integrated Care in neonatal intensive care units on infant and parent outcomes: a multicentre, multinational, cluster-randomised controlled trial. Lancet Child Adolesc Health. 2018;2(4):245–254. doi:10.1016/S2352-4642(18)30039-730039-7) ↩
  17. Franck LS, Sun Y, et al. Neonatal outcomes from a quasi-experimental clinical trial of Family Integrated Care versus Family-Centered Care for preterm infants in U.S. NICUs. BMC Pediatr. 2022;22(1):674. doi:10.1186/s12887-022-03732-1 ↩
  18. Boettiger M, Tyer-Viola L, Hagan J. Nurses' early recognition of neonatal sepsis. J Obstet Gynecol Neonatal Nurs. 2017;46(6):834–845. doi:10.1016/j.jogn.2017.08.007 ↩
  19. Forget A, Adegboye C, Alfieri M, et al. A sepsis trigger tool reduces time to antibiotic administration in the NICU. J Perinatol. 2023;43(6):806–812. doi:10.1038/s41372-023-01636-1 ↩
  20. Steele C. Best practices for handling and administration of expressed human milk and donor human milk for hospitalized preterm infants. Front Nutr. 2018;5:76. doi:10.3389/fnut.2018.00076 ↩
  21. Gad S, Sheta MM, Al-khalafawi AI, et al. Expressed breast milk contamination in neonatal intensive care unit. Pediatr Health Med Ther. 2021;12:307–313. doi:10.2147/PHMT.S311632 ↩
  22. Oza-Frank R, Kachoria R, Dail J, et al. A quality improvement project to decrease human milk errors in the NICU. Pediatrics. 2017;139(2):e20154451. doi:10.1542/peds.2015-4451 ↩
  23. Patel AL, Johnson TJ, Engstrom JL, et al. Impact of early human milk on sepsis and health-care costs in very low birth weight infants. J Perinatol. 2013;33(7):514–519. doi:10.1038/jp.2013.2 ↩
  24. Zabihi A, et al. Investigating the effect of oropharyngeal colostrum in the prevention of late-onset sepsis in preterm infants: a randomized controlled trial. Sci Rep. 2025;15(1):17390. doi:10.1038/s41598-025-02309-z ↩
  25. Rugolo LMSdS, Bentlin MR, Melo FPdG, et al. Quality improvement initiative to reduce late-onset sepsis in very low birth weight preterm infants: a multicenter study from the Brazilian network on neonatal research. J Pediatr (Rio J). 2025;101(6):101459. doi:10.1016/j.jped.2025.101459 ↩
  26. Zhang X, Wan Z, Yuan K, Wu G, Yu Z. Quality improvement interventions to prevent late-onset sepsis in premature infants: a systematic review and meta-analysis. PeerJ. 2026;14:e20530. doi:10.7717/peerj.20530 ↩