infants with NAS who require pharmacologic treatment generally have longer average hospital stays (23 days) compared with infants with NAS who do not require such medication (17 days).28 Health care providers from our selected hospitals also indicated a similar trend—the average length of hospital stay in calendar year 2016 for infants with NAS who received pharmacologic treatment ranged from 7 to 30 days, while the stays for infants who did not require such medication ranged from 3 to 7 days. 28The average length of stay for infants with NAS—depending on the treatment required— compares to an average of 2.1 days for all other hospital births. Patrick et al., “Increasing Incidence and Geographic Distribution of Neonatal Abstinence Syndrome: United States 2009-2012,” 653. Page 14 GAO-18-32 Newborn Health Medicaid generally pays for NAS treatment services in our four selected states using a diagnosis-related group (DRG) based payment system, in which hospitals receive a fixed amount for a bundle of services.29 In general, the DRG-based system used in Medicaid pays for the medical services necessary for treating infants with NAS, such as medication, bed space, and nursing staff, according to CMS officials. CMS officials said that the DRG-based system generally does not pay for professional services, such as physician visits; instead, these services are typically paid under a fee-for-service payment schedule, in which states or contracted managed care plans pay health care providers directly for their services. Officials in our selected states said information on total Medicaid payments for hospital-based NAS services was not readily available. Several DRGs are typically used to bill Medicaid for services provided to infants. However, these codes alone cannot provide an accurate estimate of Medicaid payments for NAS treatment services because the codes are not used exclusively for NAS. For example, according to some health care providers we interviewed, two DRG codes that may be used to bill Medicaid and other payers for NAS treatment services are 791 (prematurity with major problems) and 793 (full term neonate with major problems). However, these codes could be used to bill for over 2,000 diagnoses—for example, pneumonia or measles. One state official said that while they could provide us with information on Medicaid payments for these infants, they could not parse out the costs by diagnosis codes, such as those related to NAS. Thus, estimates of total Medicaid payments based only on DRG codes likely overstate the amount paid for NAS hospital-based services. Officials from two of the four selected states told us that their state has a public health surveillance system that tracks the incidence of infants diagnosed with NAS; however, the surveillance systems do not capture 29This payment system is based on DRG codes, which CMS classifies based on a number of factors, such as the patient’s age, diagnosis codes, and the presence of complications or multiple conditions. According to CMS officials, states set the DRG base payment rates, and managed care plans can further negotiate the DRG payment with health care providers. Health care providers from seven of our eight selected hospitals told us their state Medicaid program paid for NAS treatment services in calendar year 2016 using a DRG-based payment system. Health care providers from one hospital did not have knowledge on how their hospital bills Medicaid for these services. Medicaid Payments for NAS Treatment Services Provided in Hospital Settings Page 15 GAO-18-32 Newborn Health financial information, including Medicaid payments for NAS.30 At our request, one of the states cross-referenced their surveillance and Medicaid data and estimated that in 2016, their state Medicaid program spent over $22 million to treat 1,565 infants with NAS. While selected states generally could not provide information on total Medicaid payments for infants with NAS, some hospitals in our selected states were able to generate this information at our request using diagnosis codes that they identified as related to NAS from hospital claims data. Six of our selected eight hospitals reported that in calendar year 2016, the average Medicaid payment for treating infants with NAS ranged from about $1,500 to about $20,200 per infant per stay. The wide range in Medicaid payment averages may be because the averages included both infants who did and did not require pharmacologic treatment and because these hospitals treated infants in various settings, such as a nursery or a NICU. The literature we reviewed also had limited information on Medicaid payments for NAS treatment services provided in hospitals. A recent study reported that from 2009 through 2012—the most recent data available at the time of the study—Medicaid payments to hospitals for NAS treatment services increased from about $564 million to $1.2 billion nationwide. 31 30The two NAS surveillance systems rely on self-reported data submitted by hospital providers. One state requires these providers to indicate whether an infant is diagnosed with NAS, while the other state requires these providers to identify infants with NAS using International Classification of Diseases (ICD) diagnosis codes. Although one state requires these providers to identify the