December 11, 2017

expert reaction to results of trial of voucher scheme to encourage breastfeeding

Offering new mothers financial incentives may significantly increase low breastfeeding rates, new research published in JAMA Pediatrics reports.

Prof. Kevin McConway, Emeritus Professor of Applied Statistics at The Open University, said:

“In many contexts, inside and outside healthcare, there is evidence that financial incentives, even quite small ones, can affect behaviour. So to that extent, it’s hardly surprising that incentives may have an effect on mothers’ decisions on breastfeeding, as this study appears to show.

“But the researchers make it clear that their results apply only to areas where mothers previously had low rates of breastfeeding, because that’s the sort of area where they tested the incentives. And even in those areas, they only conclude that the incentives ‘may’ improve breastfeeding rates.

“Even though this is (in my opinion) a pretty well-designed and well-executed study, there are several good reasons for this rather cautious conclusion. The researchers’ primary outcome measure is based to a considerable extent on what mothers tell their healthcare professionals about how they are feeding their babies. That applies to all the mothers in the areas where vouchers were available, not just those who chose to sign up for the vouchers, but the existence of the voucher scheme may make it more likely that mothers mention to healthcare professionals that they have been breastfeeding. So we can’t be sure that all of the difference between areas with and without vouchers is because of real differences in breastfeeding. Also, the research found no statistically significant differences between areas with and without vouchers on their secondary measures (starting babies on breast milk from birth, and feeding babies only on the breast at 6-8 weeks). On both of these, the researchers did find higher breastfeeding rates in the areas with vouchers, but the differences were small enough that they could have been entirely due to chance. The researchers make these and other limitations on their conclusions perfectly clear in the research report.

“There are several other points to consider before any decisions could be taken on rolling out a scheme like this widely. Obviously the scheme would have a cost (in administration as well as the cost of the vouchers). Is it worth it? Answering that would depend on how much it changes breastfeeding rates – in this study, the rates seems to increase more the longer the scheme went on, which might be due to changes in the way mothers discuss breastfeeding with one another, and we can’t yet tell how big an increase in breastfeeding rates might be if a scheme like this went on for years. Maybe the difference in rates would continue to increase to a much higher level than the relatively small differences found in this study, or maybe after a time the incentive would stop being effective. Also, we don’t know what the effect might be in areas where breastfeeding rates are higher to begin with, and I suspect it wouldn’t be politically acceptable to roll out a scheme like this only in areas where breastfeeding rates are currently low. So there’s plenty more to consider and research on these ideas.”

Prof. Andrew Whitelaw, Emeritus Professor of Neonatal Medicine at the University of Bristol, said:

“This is a pioneering trial tackling the important problem of low breastfeeding rates in low-income areas in the UK. However, the trial design could not avoid the possibility that an economically deprived mother would be tempted to report she was breastfeeding (when she was not) in order to receive a 200 pound reward. Furthermore, the clinicians who were responsible for verifying breastfeeding would have known whether the mother was in an intervention area or a control area and might themselves have been biased by sympathy to the idea of the trial or to economically deprived mothers. There was no objective way of confirming that an infant was receiving breastmilk and the outcome reporting was not blinded.

“This trial is worth publishing because it highlights the difficulties in researching this problem but is not a justification for a general policy of economically rewarding mothers who reporting breastfeeding in areas with low breastfeeding rates.”

* ‘Effect of Financial Incentives on Breastfeeding’ by Clare Relton et al. will be published in JAMA Pediatrics on Monday 11 December.

3^rd party reaction when the trial was launched in 2013 can be found here: http://www.sciencemediacentre.org/expert-reaction-to-launch-of-feasibility-study-paying-mothers-to-breastfeed/

Declared interests

Prof. Kevin McConway: “Is a member of the SMC advisory board.”

None others received.