Learning to Help a Newborn Breathe: How Practice Dummies With Feedback Are Changing Training

What two new 2026 studies, from a hospital in Tanzania and two hospitals in London, tell families about how doctors, nurses and midwives learn to give breaths to babies

When a newborn does not start breathing well, the most important thing the team can do is to give gentle breaths with a mask and a bag or a special device. Two new studies tested whether practising on dummy babies that give feedback, such as showing how much air is going in or how much is leaking out of the mask, helps health workers get better at this skill. The answer is encouraging but nuanced: feedback was linked to steadier breaths, particularly on the very smallest dummy baby, but which kind of feedback worked best seemed to depend on who was learning.

Why this skill matters so much

Most babies take their first breaths without any help. A small number need a little support to get started, and an even smaller number need breaths to be given for them. Giving those breaths is a skill of the hands and eyes. The person holds a small mask over the baby's nose and mouth, squeezes a bag or presses a control, and watches whether the baby's chest rises. It sounds simple, but it is surprisingly hard to do well. If the mask does not sit snugly, air leaks out around the edges and the baby does not get the breath. If the breaths are too forceful or too large, delicate lungs, especially those of premature babies, can be hurt.

Two new studies, one from Tanzania and one from London, take on this problem [1][2]. Doctors have known for years that even experienced people find it hard to tell how much air is really leaking. Research in delivery rooms found that clinicians often misjudged leaks when giving breaths to premature babies [3]. International guidelines give targets for the pressure and speed of the breaths, such as starting with a gentle pressure and giving 40 to 60 breaths each minute [4]. But it is hard to learn those targets by watching a chest move on a practice dummy.

What came before: attempts to make the invisible visible

Over the last decade, researchers have tried using a "respiratory function monitor," a small screen connected to the mask that shows how large each breath is and how much of it leaks. In one large study across 13 hospitals in Australia, 400 health workers practised on a dummy baby, half with the screen visible and half with it covered. Those who could see the screen had less leak (about 23% of the air versus 35%) [5]. Another study with 412 health professionals from several countries found leak fell from 31% to 10% after training with the screen [6]. Studies with experienced clinicians found the same direction of effect on both premature-sized and full-term-sized dummies [7].

Not everything has been clear-cut. When researchers used these monitors with real premature babies in a large trial, the results were mixed [8], and a recent review cautioned that more high-quality evidence is needed before every hospital rushes to buy them [9]. What was not well understood was whether seeing the numbers during practice teaches something that lasts after the screen is turned off.

In parts of the world with fewer resources, the problem is different. Hospitals may not have monitors, and a simple hand-squeezed bag may be the only way to help a baby breathe. Families in these places face higher risks, and research suggests that having skilled people present at birth could prevent a large share of newborn deaths [10]. Training programmes for newborn resuscitation have been linked to fewer early newborn deaths [11], and teams in Tanzania have already used adapted practice tools successfully [12]. The open question was which kind of practice tool works best, and for whom.

The two studies

The first study took place at Tosamaganga Regional Referral Hospital in the Iringa region of Tanzania in June 2025 [1]. It involved 79 people: 42 students in training to become clinical officers (a type of mid-level doctor), who had never ventilated a newborn, and 37 working health professionals, most of them midwives. A research team from Italy and Tanzania had built a low-cost practice dummy that can be set up in three ways. One version was simple, with no feedback. A second version had a chest and tummy that rise realistically, like a real baby. A third added a phone app that gave lights and sounds to guide the learner. People were randomly assigned to one of the three versions, and everyone attended four days of lessons and practice sessions, aiming for 40 breaths a minute at a set gentle pressure. At the end each person took a one-minute test on the realistic version, where the chest was the only clue.

The second study took place at two London hospitals, the Royal London Hospital and Whipps Cross University Hospital [2]. Forty-nine doctors and nurses of different seniority gave breaths to two dummy babies: a very premature one, about the size of a 23-week baby weighing roughly 500 grams, and a full-term one. Each person did this twice, once with a monitor screen showing the size of each breath and the leak, and once with the screen turned away. The order was decided by chance.

What they found

In Tanzania, average knowledge scores rose in both groups, and the trainers rated almost everyone's final performance highly [1]. The interesting differences were in how people practised. The new students did well on the simplest version: they were giving the right number of breaths by their third try. Students who practised on the realistic-chest version without any prompts tended to press too hard. The experienced professionals, on the other hand, did best in practice with the app-guided version, which was associated with keeping closer to the right pressure. When the test came, though, and the prompts were gone, the group that had trained with the app was giving breaths faster than the target, on average about 49 a minute compared with 40 in the group who had trained on the realistic chest. In other words, help during practice does not always carry over when the help is taken away.

In London, seeing the screen made a clear difference for the tiny premature dummy [2]. Among people who first practised without the screen, the share of breaths with only a small leak (30% or less) rose from about 78% to about 97% once they could see it, and their breaths became much more even. People who started with the screen visible kept up their good performance when it was hidden afterwards, which suggests that they had learned something that stayed with them for the rest of the session. With the full-term dummy, results were less clear: breath size improved somewhat for the group that started without the screen, but leaks stayed high for everyone. The researchers judged the full-term dummy to be less lifelike, which may have played a part. More than nine in ten said the screen was helpful and made them feel more confident.

What this means for babies and families

These studies used practice dummies, not real babies, and lasted for a few days or a few minutes, so they cannot tell us that babies born in these hospitals have better outcomes. What they do suggest is helpful for families to know.

First, hospitals are working to make this skill more consistent. Using a screen to show the size of a breath and the leak helps teams practise sealing the mask and giving the right-sized breath, and looks especially useful for premature babies.

Second, good training does not have to be expensive. The Tanzanian project suggests that a low-cost practice dummy, used within a four-day course, was acceptable to both beginners and experienced staff, and that beginners did at least as well in practice with simpler tools. Researchers have long argued about whether the most realistic and high-tech dummies are always better; some work has found that fancier is not necessarily better for students [13].

Third, feedback is a helper, not a replacement for people. The studies reinforce the importance of lessons, demonstrations and teachers who can guide learners through the tricky parts, such as holding the mask so that it seals.

If your baby needs help with breathing after birth, you can ask the team about their training and practice routines. Asking about training is a reasonable question and is not a sign that anything is wrong.

What researchers want to find out next

The authors of both studies say that the next steps are to find out how long the benefits last, whether better practice in a simulation leads to better care of real babies, and what kind of feedback works best at which point in someone's training. Other researchers have found that people who trained with a monitor kept improvements in the size of the breaths they gave for up to three months, though leaks remained hard to fix [14]. Participants in London also asked for clearer screens, with colour cues and alarms for high leak.

For related reading on how teams support a newborn's airway and breathing, see In[Neo]Sight's articles on supraglottic airways during newborn resuscitation and video laryngoscopy in newborns.

A hopeful summary

Learning to give a newborn breaths is a hands-on skill, and like any skill it improves with good practice and good feedback. These two studies show that smart practice tools, whether a high-tech monitor in London or a low-cost dummy in Tanzania, can help health workers become steadier and more confident. They also remind us that the best training is tailored to the learner, and that the real goal is not a good performance on a dummy but safer first breaths for real babies.

References

  1. Mocellin D, Maglio S, Nanyaro MW, et al. Neonatal ventilation training using a low-cost, high-fidelity simulator in a low-resource setting: a randomized clinical trial. JAMA Netw Open. 2026;9(9):e2633261. doi:10.1001/jamanetworkopen.2026.33261 ↩
  2. Alrashed L, Bodanese E. Impact of real-time respiratory function monitoring on neonatal mask ventilation training: a multicentre simulation-based crossover study. BMJ Paediatr Open. 2026;10(1):e004894. doi:10.1136/bmjpo-2026-004894 ↩
  3. Schmölzer GM, Kamlin COF, O'Donnell CPF, et al. Assessment of tidal volume and gas leak during mask ventilation of preterm infants in the delivery room. Arch Dis Child Fetal Neonatal Ed. 2010;95(6):F393-F397. doi:10.1136/adc.2009.174003 ↩
  4. Hogeveen M, Monnelly V, Binkhorst M, et al. European Resuscitation Council guidelines 2025: newborn resuscitation and support of transition of infants at birth. Resuscitation. 2025;215:110766. doi:10.1016/j.resuscitation.2025.110766 ↩
  5. O'Currain E, Thio M, Dawson JA, et al. Respiratory monitors to teach newborn facemask ventilation: a randomised trial. Arch Dis Child Fetal Neonatal Ed. 2019;104(6):F582-F586. doi:10.1136/archdischild-2018-316118 ↩
  6. Ni Chathasaigh CM, Smiles L, Curley AE, et al. Newborn face mask ventilation training using a standardised intervention and respiratory function monitor: a before and after manikin study. Arch Dis Child Fetal Neonatal Ed. 2024;109(5):505-510. doi:10.1136/archdischild-2023-326416 ↩
  7. Tracy MB, Hinder M, Morakeas S, et al. Randomised study of a new inline respiratory function monitor (Juno) to improve mask seal and delivered ventilation with neonatal manikins. Arch Dis Child Fetal Neonatal Ed. 2024;109(5):535-541. doi:10.1136/archdischild-2023-326256 ↩
  8. van Zanten HA, Kuypers K, van Zwet EW, et al. A multi-centre randomised controlled trial of respiratory function monitoring during stabilisation of very preterm infants at birth. Resuscitation. 2021;167:317-325. doi:10.1016/j.resuscitation.2021.07.012 ↩
  9. Patel N, Collier LP, Aghababaie A, et al. Respiratory function monitoring for neonatal resuscitation: a narrative review. Arch Dis Child Fetal Neonatal Ed. 2025;111(1):F6-F12. doi:10.1136/archdischild-2025-328586 ↩
  10. Bhutta ZA, Das JK, Bahl R, et al. Can available interventions end preventable deaths in mothers, newborn babies, and stillbirths, and at what cost? Lancet. 2014;384(9940):347-370. doi:10.1016/S0140-6736(14)60792-360792-3) ↩
  11. Pammi M, Dempsey EM, Ryan CA, Barrington KJ. Newborn resuscitation training programmes reduce early neonatal mortality. Neonatology. 2016;110(3):210-224. doi:10.1159/000443875 ↩
  12. Kalabamu FS, Daudi V, Moshiro R, et al. Neonatal resuscitation skills acquisition among healthcare providers after Helping Babies Breathe simulation training using improved tools across two regions in Tanzania. Adv Simul. 2025;10(1):6. doi:10.1186/s41077-025-00338-2 ↩
  13. Massoth C, Röder H, Ohlenburg H, et al. High-fidelity is not superior to low-fidelity simulation but leads to overconfidence in medical students. BMC Med Educ. 2019;19(1):29. doi:10.1186/s12909-019-1464-7 ↩
  14. Carlomagno CEB, Mascaretti RS, Lyra JC, et al. Monitoring respiratory mechanics as a training tool for manual ventilation. Front Pediatr. 2025;13:1638566. doi:10.3389/fped.2025.1638566 ↩