Short term cost of care for the surviving periviable neonate

1 Introduction

Preterm births impose a heavy financial burden on the U.S. health care system, costing more than $26 billion in 2005 [1, 2]. Only 6% of preterm infants are born <28 weeks’ gestational age; however, this population accounts for 1/3 of the total cost [3]. Mortality is high in the extreme premature infant with rates of 74%, 45%, and 28% at 23, 24, and 25 weeks’ gestation respectively [4, 5]. In addition, moderate-to-severe neurodevelopmental impairment is common at 4–8 years of age ranging from 40% for 23 weeks’ gestation to 24% for 25 weeks’ gestation [6]. Very little is known regarding the current costs of these periviable infants during their initial hospitalization. The objective of this analysis is to describe the hospital costs and distribution of financial charges for the initial hospitalization of the surviving periviable neonate.

2 Methods

We studied the medical records and financial data for neonates of 23–25 weeks’ gestational age cared for in a single tertiary care, university-based NICU from 1/1/2012-6/30/2015. Gestational age was determined as the best obstetric estimate or postnatal assessment. The first trimester ultrasound was considered most accurate for gestational age, followed by any subsequent ultrasounds, last menstrual period and postnatal assessment. Hospital cost was determined by a detailed breakdown of cost components that included costs directly related to care including nursing staff salaries and benefits, respiratory therapy staff salaries and benefits, depreciation, administration and general (marketing, provider tax, malpractice insurance, business office, registration, information technology, and hospital administration), maintenance, housekeeping/laundry, medical records, social services, and house staff to give a rate per patient day for intensive and intermediate care. Allocations were based on statistics including square footage for maintenance, housekeeping and building depreciation. Fixed asset log was used for calculating depreciation on equipment. Rotation hours were used for house staff, time spent for social services, total admissions for medical records, and accumulated costs for administration. In addition, neonatologist cost was determined by salary, benefits, billing overhead and liability insurance to determine a cost per week for care divided by average daily census/week to give a cost/day/patient. Unfortunately, laboratory, radiology, and pharmacy services were billed by independent agencies not directly connected to the hospital finance department. Employee and physician salaries and benefits and medication and other costs were unavailable to us, so direct cost for services could not be determined. We used the Healthcare Cost Utilization Project (HCUP) Cost-to-Charge Ratio files to convert these charges to costs [7]. Each file contains hospital-specific cost-to-charge ratios based on all-payer inpatient cost for hospitals. Ancillary charge data were converted to hospital cost estimates by multiplying total charges with the appropriate cost-to-charge ratio. The cost-to-charge ratio used for this study was 31.8%. Surgery costs and physician consultant costs were not included as they made up only <1% of the charges and were considered non-essential to the cost analysis. Total charges included all hospital charges including ancillary services attributed to respiratory, pharmacy, laboratory, radiology (including radiologist charges), surgery (including surgeon and anesthesiologist fees), any pediatric specialty consultation fees, and neonatologist charges. Because our goal was to compare costs of survivors, non-survivors were excluded from analysis as were patients transferred to another facility prior to discharge.

All analyses were performed by using GraphPad Prism version 7.01 for Windows, GraphPad Software, La Jolla California USA. T test comparisons of length of stay, hospital cost, charges, and survival were calculated using the Bonferroni-Sidak method. P value <0.05 was considered statisticallysignificant.

This study was approved by the institutional review board of Marshall University. The authors have no conflict of interest to disclose.

3 Results

A total of 59 periviable neonates (23–25 weeks’ gestation) were admitted and cared for in our NICU over a 42 month period from 1/1/2012-6/30/2015. There were an additional sixteen deliveries of the same gestational age who were assigned one and five minute Apgar scores of zero who were labeled stillborn and not included in the study. Eight infants were transferred to another facility prior to discharge. Of these 4 were transferred for surgical care when our pediatric surgeon was unavailable and 4 were transferred to receive convalescent care. Seven of these infants survived; however, no financial data were available and they were excluded from the study. Eleven infants died prior to discharge and were also not included in this study. 40 infants survived until discharge and financial data and medical records were analyzed for all these infants (Table 1). All 40 received surfactant in the first two hours of life and 32 of the mothers received at least one dose of antenatal steroids and 22 received a complete course of 48 hours. Overall survival for this study population was 78%. Despite small numbers, survival increased between 23 weeks and 25 weeks and 24 weeks and 25 weeks (both p < 0.01) but not between 23 and 24 weeks. Average length of stay, total charges, and hospital costs decreased with increased gestational age between 23 and 25 weeks(p < 0.05).

Total charges were analyzed and allocated to the following distribution: NICU-56%, Respiratory-11%, Pharmacy-6%, laboratory-6%, radiology-6%, surgery-1%, neonatologist-13%, and misc.-1%.

Patient#1 was delivered at 23 4/7 weeks of gestation and weighted 540 grams. She remained in the NICU for 112 days. The calculation of her hospital cost is described in Table 2.

Literature on the short-term cost for the periviable neonate is scarce. In patients cared for from 1990–1994 at a single institution the cost to produce a survivor was $294,749 and $181,062 with lengths of stay of 120 and 85 days at 24 and 25 weeks respectively [8]. From cohort data for all births in California in 1998–2000, it was reported that the cost and length of stay for 24 and 25 weeks’ gestation were $297,629 (109.6 days) and $272,730 (101.7 days) respectively [9]. Recently, from a secondary analysis of the 2009 Agency for Healthcare Research and Quality National Inpatient Sample looking at mechanically ventilated patients across the age span, surviving, ventilated infants <24 weeks’ gestation had a mean hospital cost of $209,000 and those 24 weeks’ gestation a mean cost of $200,000 [10].

In all these studies, hospital charges were converted to measures of treatment costs using cost-to-charge ratios from hospital cost reports submitted to CMS. Despite different methodologies, the numbers from these previous studies compare closely to the hospital costs calculated in the current study.To our knowledge, this is the first study to attempt to compare hospital and physician costs, in large part, by a detailed breakdown of cost components.

Our study has several limitations. We have limited our study to short term direct hospital costs for the initial hospitalization. Also indirect costs including family expenses and lost income are excluded that would contribute to total cost. Information regarding long term costs of post discharge care was not reported as well as long term survival and morbidity are not included in this study. Finally, this is a small study population at a single institution and data from other institutions in different regions could be substantially different. Transferred patients were excluded from this study as costs would not be comparable to other patients in the study.

Our study describes a detailed breakdown of hospital cost and distribution of charges for the periviable neonate during the initial hospitalization providing essential information for administrators in this new era of value-based population management and may help to guide clinicians in using economic information for quality improvement and building a foundation for future cost management.