Safety and durability are a top priority for products designed for use  among vulnerable populations. With this in mind the DREAM Infant Warmer has been developed to meet the highest standards and has successfully undergone significant testing both in the laboratory and field settings to ensure there is no risk of toxicity, irritation, or skin sensitization.

Laboratory Testing

The phase change material (PureTemp37) that comprises the candles is made from USDA-certified vegetable oil that is non-hazardous and safe for transport. It has been tested to perform for 10,000 cycles. Durability tests were also performed to confirm that the phase change material would not leak. In these tests, the candles in the Infant Warmer were repeatedly subject to heavy weights, horizontal and vertical drops, rapid cooling and exposure to bleach.

The external polymer film covering the warming pad is certified by USDA and FDA to hold air, fluids, and vacuum, and withstand very high temperatures. The plastic was tested by submerging each sample in boiling water for 30 minutes and cooling for 10 minutes. There was no apparent wear or melting in these tests.

The insulating mat is made from non-toxic foam and neoprene.

Each of these components is manufactured in the United States. Initial laboratory tests of thermal performance indicate that under the estimated conditions of use the DREAM Infant Warmer functions as expected.

Field Tests

Phase 1

The warmer was tested 102 times for babies who were hypothermic or at risk for hypothermia because of low birth weight (< 2.5 kg) in two district hospitals in Rwanda. In 98% of uses, hypothermia was prevented or corrected and there were no adverse effects except seven instances of mild hyperthermia. There were no instances of incorrect preparation, use or cleaning despite only a short training period. We received overwhelmingly positive feedback from mothers and nurses.

Phase 2

We have recently completed use of the warmer an additional 102 times in the health center setting in Rwanda, along with qualitative interviews of nurses and mothers regarding their opinions of the warmers and its relationship with STS. There was only one instance of mild hyperthermia (37.7 °C) that resolved despite the infant remaining on the warmer. Of 81 hypothermic infants placed on the warmer, 79 became euthermic. Again there were no instances of incorrect use.

Qualitative Results were overwhelmingly positive:

There is nothing else I can add except for thanking you who have brought this infant warmer, because it has helped me. I don’t even know how my baby would get warm had you not given me the warmer.”

Rwandan Caregiver

“When you measure infants’ temperature and you find that they are cold, you can give one to each infant and then where you are in your job you can feel confident that your infants are warm.”

Rwandan Nurse

“There is a case where a mother delivers a baby and she does not have a maid. After giving birth, she finds her alone to treat herself and baby. She thinks how to carry the baby, clean materials, wash herself and the baby, and finally, all these issues bring the challenges. But if it was this infant warmer, she would put the baby on it and deal with the rest.”

Rwandan Nurse

Based on suggestions from qualitative interviews, in our training materials we will stress the need to anticipate preparation of the warmer in advance of need, and provide instructions in multiple languages.

Future Testing

Based on the encouraging results from our pilot field studies, we are now designing a step-wedge cluster randomized controlled clinical trial. With this trial design, each hospital will serve as its own control: The control data is collected during an initial pre-intervention period for each hospital. During this time, hospitals will provide current standard of care based on the Rwandan National Neonatal Protocols, with STS offered for infants who meet certain criteria based on degree of prematurity, body weight and temperature while collecting data regarding vital signs, rate of growth, hospital length of stay and mortality statistics. Then a randomized cluster of hospitals will be added at regular intervals (steps) at which time the infant warmer will be introduced and that cluster will move from the control arm to the treatment arm (with the treatment being the provision of the DREAM Infant Warmer). Gradually all of the study hospitals will move from the control arm to treatment arm over an approximately 6 month period.

This design is intended to provide increased statistical power relative to a standard randomized controlled trial due to repeated observations and a lower intra-cluster coefficient. Furthermore, this design helps to sidestep the ethical issue of randomly choosing a sample of hypothermic infants as a control group who are offered a lesser level of care than another infant in the same hospital at the same time. Finally, this design allows for efficiency of staffing as only a subset of “cluster” hospitals needs training at any one time.