Multisensory stimulation in preterm infants: the trials have not answered the question

Design

Cochrane systematic review and meta-analysis of randomised trials, fixed-effect models, GRADE certainty assessment

Population

19 randomised trials including 1554 preterm infants; 18 against standard care, one against tactile stimulation

Primary outcome

major neurodevelopmental disability at 18 to 24 months corrected age, death during initial hospitalisation, and total weight gain at discharge

Effect

death RR 0.97 (95% CI 0.54 to 1.73), low certainty; weight gain MD 72.67 g (68.23 to 77.12), low certainty; no trial reported neurodevelopmental disability at 18 to 24 months

A Cochrane review searched to November 2025 and included 19 randomised trials in 1554 preterm infants - 18 comparing multisensory stimulation with standard care, one against tactile stimulation alone. The critical outcomes were set in advance: major neurodevelopmental disability at 18 to 24 months corrected age, death during the initial hospitalisation, and total weight gain by discharge.

Not one included trial reported developmental delay, intellectual impairment, blindness or sensorineural deafness at 18 to 24 months. A single trial of 18 infants reported cerebral palsy at 12 months, giving a risk ratio of 0.67 (95% CI 0.28 to 1.58) rated very low certainty. One trial of 395 infants found little to no difference in death (RR 0.97, 95% CI 0.54 to 1.73), low certainty. Three trials in 474 infants found greater total weight gain before discharge, mean difference 72.67 g (95% CI 68.23 to 77.12), also low certainty. Thirteen further studies are ongoing.

The honest reading is that this is an intervention with a plausible rationale, widespread informal use, and an evidence base that has spent two decades measuring the wrong things. Seventy-two grams of extra weight gain, on low-certainty evidence, is not a reason to change practice and not a reason to abandon it either. Nothing here argues against developmentally appropriate care in a neonatal unit; it argues that the trials have not tested whether structured multisensory programmes deliver what they claim.

• Do not present multisensory stimulation to parents as proven to improve development - no trial measured that at 18 to 24 months.
• Continue developmentally supportive care on its own merits; this review does not weigh against it.
• Treat the 72.67 g weight difference as a surrogate of uncertain meaning, not as an outcome parents should be promised.
• If your unit runs a structured programme, record the outcomes Cochrane says are missing - neurodevelopment at 18 to 24 months above all.
• Thirteen studies are ongoing; this is a question that will be answerable, and is not yet.

The statistics, in plain English

GRADE certainty of low or very low means the estimates could change substantially with better evidence - here driven by risk of bias, very small samples and wide intervals. The cerebral palsy risk ratio of 0.67 comes from 18 infants, which is why its interval runs from 0.28 to 1.58 and tells you essentially nothing. The weight gain figure looks oddly precise - 72.67 g with an interval of 68.23 to 77.12 - because it pools three small trials under a fixed-effect model with no heterogeneity detected; a narrow interval around a pooled estimate is not the same as confidence that the effect is real, when the underlying trials are at risk of bias. The absence of reporting on the review's critical outcome is the finding, not a footnote.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.