Using diagnostic information, we obtained a sample of more than 400,000 admissions for delivery from 2020 inpatient delivery claims. We defined a delivery admission by the unique combination of individual, diagnostic related group (DRG), and admission and discharge dates. For each admission, we associated all facility claims with professional claims that occurred in the period between admission and discharge dates. We included only claims associated with a female member between the ages of 18 and 45.
We measured spending per admission as the sum of the plan payment and member out-of-pocket costs. We measured the cost per birth as total spending divided by a count of admissions for delivery. For state-level results, we aggregated claims at the state level and reported on states that had data on at least 400 deliveries attributed to at least five providers in 2020, including 100 vaginal birth deliveries and 100 c-section deliveries.
To the extent possible, we aimed to align methodologies across payers. We examined approximately 1.4 million inpatient stays. Claims had to have an enrollment record with at least one day of eligibility in the month the delivery discharge occurred. We included only enrollees who were female and between the ages of 18 and 45 years old. We excluded all women who were dually eligible for Medicare and Medicaid. Less than one percent of individuals had more than one inpatient childbirth claim during the calendar year; in these cases, we kept all claims and treated each as an independent stay.
The following states were excluded because the count of births in the T-MSIS claims data was more than 50% below the CDC reported Medicaid-covered births in the state: Alabama, Maryland, Mississippi, New Hampshire, New Jersey, Rhode Island, and Tennessee. Note that some states do not reconcile all or some Medicaid presumptive eligibility records, and there may be some variation in the number of records lost due to our data cleaning processes.
For payments, we included inpatient stays where the Medicaid paid amount was greater than $0. We included other non-facility payments that occurred between the start and end date of the index inpatient stay; for the stays without corresponding non-facility payments, we imputed the average non-facility payments at the state level. Additionally, we did not include any amounts paid by Medicare or other third-party payers. DRGs were calculated using ICD-10-CM/PCS MS-DRG v37.0 definitions.
Caution must be used when interpreting the average Medicaid payments made as payments are calculated from non-managed care claims only. Payment information may have varying representativeness across states given that rates of managed care enrollment vary from 0% in Alaska to nearly 100% in Hawaii. States not only have variation in the mix of managed care and fee-for-service beneficiaries, but some services may be carved in or carved out of managed care or care-coordination plans. These arrangements may affect not only the payment data we observe but also potentially the mix of services beneficiaries receive.