1 | 2 | All | Print this article

Can Healthcare IT Save Babies?

Catherine Tucker

Catherine TuckerCatherine Tucker is the Mark Hyman, Jr. Career Development Professor and an Associate Professor of Marketing

Each year, more than 18,000 babies in this country die within their first 28 days of life. This grim statistic ranks the United States 43rd in the world in terms of neonatal mortality—tied with Montenegro, Slovakia, and the United Arab Emirates, and behind 24 of the 27 members of the European Union.

In addressing these and other disparities in U.S. healthcare, much policy emphasis has been placed on moving the country toward digital medical records and abandoning the current paper-based system. This effort culminated in 2009 with the Health Information Technology for Economic and Clinical Health (HITECH) Act, which promised large subsidies for healthcare providers who moved to digital records. However, there was little broad-scale evidence about the effectiveness of these investments.

To determine whether the United States would see a marked improvement in neonatal mortality if it were to adopt electronic medical records (EMRs), my colleague Professor Amalia Miller from the University of Virginia and I examined 12 years of birth and death records for all American counties. We combined these with national data on the adoption of EMRs by U.S. hospitals. Then we ran a regression analysis—a statistical technique for estimating the relationships among variables—to include controls for hospital and county characteristics.

We found that a 10 percent increase in basic EMR adoption would reduce neonatal mortality rates by 16 deaths per 100,000 live births. Beyond this, a 10 percent increase in EMR adoption plus the same increase in obstetric-specific computing technology would cut neonatal mortality by 40 deaths per 100,000 live births. We documented that this increase was driven by the kind of conditions that could be helped by detailed documentation, such as those stemming from difficulties during a pregnancy, rather than by conditions that EMRs were less likely to help, such as congenital defects and accidents. Our study provides cautious optimism about the potential value of healthcare IT and EMRs in improving neonatal health outcomes.

When studying technology-based healthcare improvements, one concern is whether they reinforce or reduce inequalities in healthcare outcomes. Therefore, a striking additional empirical finding was that the adoption of EMRs produced larger gains for historically disadvantaged groups, such as less educated mothers and black mothers.

Healthcare IT can play a role in standardizing care and ensuring that best practices are pursued in all cases. Currently, babies born to black mothers are twice as likely to die within their first 28 days of life than are those born to white mothers. Mothers with less than a high school degree represent about 22 percent of our sample, and their children have a slightly elevated neonatal death rate compared to all births.

There is some evidence that these differences in outcomes reflect the differences in treatments offered to different groups. To the extent that healthcare IT systems reduce this variation in treatment decisions—by decreasing the burden on mothers to advocate for certain treatments and recalling past medical histories—they may improve outcomes for these historically disadvantaged groups.

Our rough calculations suggest that EMRs are associated with a cost of $531,000 per baby saved. The estimated impact of basic EMR adoption is a reduction of 1.6 deaths per 1,000 live births. In other words: A complete national transition from paper to computer records could save as many as 6,400 infants per year, out of about 4 million births.

The cost of doing so is substantially lower than the cost of expanding Medicaid, which would cost $840,000 per infant life saved. By comparison, the Office of Management and Budget endorses values between $1 million and $10 million for a statistical life in cost-benefit evaluations. This indicates that spending money on this kind of healthcare IT is relatively cost-effective.

Our findings suggest that investing in healthcare IT, and specifically in EMRs, is an effective way to improve neonatal health outcomes. EMRs have the potential to standardize treatment options, which can improve mortality rates for babies born to mothers of historically marginalized groups. They also represent a cost effective investment for hospitals. These findings offer support for current health policy that is directed toward increasing the usage of EMRs and other technologies.