A medical intervention with a high return on investment (ROI) means that the health benefits gained by investment in the intervention compare favorably to its cost. In this issue of the American Journal of Obstetrics & Gynecology , Dr Peterson and colleagues report the substantial health and cost burdens associated with pregestational diabetes (PGDM) that could be prevented with universal preconception care (PCC). The authors estimated the number of preterm births, birth defects, and perinatal deaths that could be reduced by the provision of PCC. Associated discounted lifetime costs averted for the affected cohort of children could be as high as $4.3 billion dollars. Including women with undiagnosed PGDM confers additional benefit in health outcomes and $1.2 billion dollars in averted cost. Although impressive, the authors acknowledge that some of these advantages are already being realized, as some women with PGDM are receiving PCC. Although there is also no consideration for what PCC costs and how it will be provided, we believe that the findings suggest PCC has a high ROI for women with PGDM.

Thirty plus years of data demonstrate that improving maternal preconception health improves reproductive health outcomes and may reduce societal costs. For example, a systematic review and metaanalysis found that PCC reduces congenital malformations (relative risk [RR], 0.25; 95% confidence interval [CI], 0.15–0.42), preterm delivery (RR, 0.70; 95% CI, 0.55–0.90) and perinatal mortality (RR, 0.35; 95% CI, 0.15–0.82) among women with PGDM. Targeting women with PGDM before pregnancy allows for optimal glucose control in anticipation of pregnancy and organogenesis. To turn data into action, in 2006 the Centers for Disease Control identified PGDM as a high priority area for PCC, given the known teratogenic effect of hyperglycemia seen in untreated or poorly controlled PGDM.

However, provision of universal PCC implies that pregnancy is intended or planned. Almost half of all pregnancies are not planned or are unintended, and unfortunately women with chronic illness are significantly more likely to have an unintended pregnancy than women without chronic illness (odds ratio, 1.8; 95% CI, 1.5–2.3). Approximately half of women reporting an unintended pregnancy also reported not using contraception at the time of conception. A qualitative study of pregnant women with PGDM who did not seek PCC reported that 70% of the women were not using contraception and that a major barrier to seeking PCC was conceiving sooner than expected. In the 2002 National Survey for Family Growth of 5955 women, up to 40% of women with diabetes did not use contraception, and additional studies of unplanned pregnancies suggest low rates of contraceptive use among women with diabetes. Addressing pregnancy planning is an essential component to delivering PCC to women with PGDM. However, women may receive inconsistent advice from varied providers, or may be overwhelmed by information and clinical recommendations. A poorly coordinated effort to achieve PCC will not work. Instead of adhering to provider recommendations, these data suggest women with PGDM may in fact avoid the health care system. Preparing for pregnancy has special meaning for women with PGDM because of the implications of periconceptional glucose control. We have a responsibility to understand and address all barriers to that preparation.

Pregnancy planning for women with PGDM must extend beyond simply recommending contraception until euglycemia is achieved. Reasons a woman chooses not to contracept and intendedness of pregnancy are complicated, thus complicating planning. Pregnancy may be unplanned but desired, and desire for pregnancy may better represent family planning practices than intendedness of pregnancies. To maximize the potential ROI of PCC, clinicians caring for women with PGDM should inquire about desire and intendedness for pregnancy, recognizing that unintended but desired pregnancies will occur. Such management may be more successful in reaching women who need it if focused on supportive measures and intensive glycemic control, rather than strictly contraception-focused family planning.

Although unplanned pregnancy may partially account for lack of PCC among with PGDM, other factors are important. Women with PGDM are more likely to receive PCC if they are married or living with a partner and employed than if single or unemployed; to have beyond a high school education and income >$20,000 than less than high school or <$20,000 income; and to have insulin-requiring PGDM. Interventions to provide PCC to women with PGDM must address social and economic barriers, and the cost of providing PCC to all women must be considered. PCC has the potential to be very costly because of the intensity and multidisciplinary nature of PCC for women with PGDM. In an older cost-benefit analysis of PCC for women with PGDM, the cost of PCC was estimated to be approximately $4000 per delivery. It is daunting to consider what those costs would be today for a taxed health care system. Nonetheless, to measure ROI these costs must be considered but were not in the current analysis.

The analysis by Peterson et al is a robust epidemiologic study, using nationally representative data to develop a model that illustrates the potential beneficial health impact of universal PCC on birth outcomes. The authors are appropriately conservative in their final estimates of potential health and cost burden and transparent in reporting model limitations. In particular, the metaanalysis data on which the authors based PCC benefits included older studies. Thus, they acknowledge that improved care in recent years may attenuate calculated benefit. On the other hand, this analysis did not account for the preventable burden of PGDM on maternal health and outcomes. Despite a substantial cost of PGDM treatment, the potential reduction in adverse maternal outcomes and associated cost savings are unknown, and likely significant. The authors cite previous studies of small-scale PCC interventions among women with PGDM are cost-effective and improve birth outcomes. Athough it is challenging to quantify the degree to which PCC improves outcomes, this is the first study to produce estimates of costs averted and to highlight essential areas of research.

An obvious criticism of the analysis would be to question the applicability of the reference group: a population of women with PGDM where none receive PCC. This is unrealistic, as despite all barriers, some women do receive what we currently consider comprehensive PCC, and others see providers who care for their PGDM, even if not specifically for preconception purposes. And if women do not receive PCC at a medical provider’s office, information is readily accessible from the Internet, social media, and lay press. It is important to recognize that the authors did not intend to imply PCC is nonexistent, or try to define what constitutes ‘comprehensive’ PCC for women with PGDM. The analysis quite convincingly demonstrates that PCC of PGDM has the potential to avert significant costs, which implies that we can achieve significant health returns with a modest investment.

The findings from this study yield very important information, but must be considered in the context of what is feasible in our current medical climate. Although PCC for women with PGDM can avert adverse outcomes and save money, what should the content of this care be and how do we provide it for all women? Who pays for it? Despite these limitations, the results suggest that the PCC-preventable health and cost burdens associated with PGDM are substantial. Thus innovative research is needed to assess the costs and practicality of delivering this care. The challenge is now to provide affordable, universal PCC for all women, but especially those with PGDM.