- HBPB —
- Healthy Baby Prenatal Benefit
Can society buy a better infant? This is, in essence, the question posed in this month’s Pediatrics article by Brownell et al.1 Studying a group of low-income Canadian women who participated in an unconditional cash transfer program (Healthy Baby Prenatal Benefit [HBPB]), the authors compared the birth outcomes of participants with outcomes among a group of low-income nonparticipants. They found that the group that received the equivalent of $81.00 per month over the second and third trimesters of their pregnancies experienced a 21% decrease in the rates of low birth weight and a 17.5% decrease in the rates of prematurity among other clinically important outcomes.
The findings are startling and raise some important policy questions. Few would disagree that society has a general interest in preventing low birth weight infants for many reasons, not the least of which is that the care they require is expensive to provide. Couple this recognized societal interest with the findings that poor women have disproportionately higher rates of delivering low birth weight infants and the argument for directing support to these women is highly persuasive.
But how should that support be provided to be most effective? In 1 approach, custodians of society’s resources presume to know what types of goods are most important for pregnant women to receive. This leads to in-kind transfers of so-called “merit goods,” such as food or health care services. The authors of the current article are making a case that unconditional cash transfers constitute a superior strategy not least because we don’t know exactly what mechanism must be disrupted to avert an undesired clinical outcome. Such a policy assumes that, like others operating under budget constraints, poor women know how to allocate available resources to achieve optimal results for themselves and should be given maximum leeway to do so. This argument is much in keeping with standard neoclassical economic theory that posits that utility is maximized when budgets are expanded unconditionally rather than when they are increased through the provision of specific goods that may or may not be valued by the recipient.2
And yet this approach is rare in practice. In 2002, it has been estimated that of the total safety net expenditures for children in the United States, almost 93% went to in-kind transfers notably for health care, nutrition, housing, and education services.3 Even the fastest growing cash-transfer payment program in the United States that has been shown to have positive impacts on infant health outcomes, the Earned Income Tax Credit, is conditioned on the recipient being active in the labor force, suggesting an abiding predisposition in US policy circles toward the goal of using incentives to manage the behavior of poor families rather than to maximize their choices.4 In an age of conservative politics, unconditional cash transfers occupy an ironic niche in public discourse: a policy supported by free-market theory that is eschewed in current practice.
How well do Brownell and her colleagues1 make their case? The validity of their conclusions, absent a randomized design, hinges on the potential impact of unobservable confounders as the authors themselves have taken pains to acknowledge. They suggest that the sensitivity analysis they conduct should reassure readers in this regard. Yet their results warrant significant caution.
The 2 groups being compared differ substantially in observed characteristics, and the authors go to considerable lengths to correct for these differences; yet we are assured that any unobservable differences between the 2 groups should not compromise the findings. Is it likely that unobservable confounders would be any less asymmetrically distributed than observable ones? We are, additionally, asked to accept on the 1 hand that unobserved confounders might readily affect estimates of the program’s impact on the likelihood of having a small for gestational age infant but on the other hand not affect the likelihood of delivering a low birth weight infant. What type of confounder could produce such an outcome?
Methodological considerations aside, the magnitude of the program’s impact should give pause. A payment of $81.00 per month translates into less than $3.00 per day. What is being purchased for this sum that could lower the risk of having a low birth weight infant by one-fifth? Another way of looking at this is to compare this investment with the financial gain resulting from it. Estimates of the hospitalization costs in the first year of life alone for an average low birth weight infant in the United States are on the order of $15 000.5 This would mean that the $480 invested over 6 months in the program produces a 21% chance of avoiding this $15 000 cost for a savings of $3150 or a return of >650%. Even more striking, if the infant mortality rate for low birth weight infants is ∼5%,6 then a decrease of 21% in the rate of low birth weight translates into an ∼1% decline in infant mortality (0.05 × 0.21). For every 100 enrollees in the HBPB program, on average 1 life would be saved at a cost of $48 000. Because the statistical value of a life saved has been calculated to be on the order of $6 million,7 this would suggest that HBPB is generating a return of $125 for every dollar invested: a phenomenal sum.
Brownell and colleagues1 have advanced an important debate on what type of transfer payments best help avoid adverse birth outcomes, and many economists would agree with their argument in principle. The assertion, however, that $81 per month represents an accurate market clearing price sufficient to purchase a 21% improvement in low birth weight rates should be received with the skepticism reserved for findings that sound too good to be true.
- Accepted March 1, 2016.
- Address correspondence to Andrew D. Racine, MD, PhD, Albert Einstein College of Medicine and Montefiore Health System, 111 East 210th St, Bronx, NY 10467. E-mail:
Opinions expressed in these commentaries are those of the author and not necessarily those of the American Academy of Pediatrics or its Committees.
FINANCIAL DISCLOSURE: The author has indicated he has no financial relationships relevant to this article to disclose.
FUNDING: No external funding.
POTENTIAL CONFLICT OF INTEREST: The author has indicated he has no potential conflicts of interest to disclose.
COMPANION PAPER: A companion to this article can be found online at www.pediatrics.org/cgi/doi/10.1542/peds.2015-2992.
- Copyright © 2016 by the American Academy of Pediatrics