The value of routine laboratory screening in the neonatal intensive care unit

Abbreviations

1 Background

The Department of Health and Human Services projects that healthcare spending will grow at a faster rate than the national economy over the coming decade. Projected national health expenditures for 2017 to 2026 will outpace the gross domestic product (GDP) growth rate by one percentage point. Healthcare spending is expected to grow at a rate of 5.5% per year, reaching a new peak of $5.7 trillion by 2026 [1].

Prematurity has a large impact on healthcare spending. The incidence of preterm birth ranges from 5% to 18% worldwide and the societal economic burden of preterm infants is estimated to be about $26.2 billion annually in the United States [2]. Though some literature has shown that a small percentage of neonatal intensive care unit (NICU) costs go towards futile care [3], surviving infants do have an increased risk of morbidity and mortality. The period of gestation remains one of the most important predictors of an infant’s future health and survival [4]. The infant mortality rate per 1,000 live births is 186.4 for infants less than 32 weeks gestation, which is 75 times the rate for full term infants [5].

Preterm infants often require a prolonged intensive care admission, along with other life sustaining interventions such as the use of ventilators, medications, blood product transfusions, invasive procedures, and laboratory work [6, 7]. The cost of initial hospitalization in preterm infants remains very high [8–10], and it increases with decreasing birth weight or gestational age [8, 10–14]. The challenge for clinicians is to reduce the incidence of short-term complications and long-term disability by providing comprehensive care in the most cost-effective manner possible.

Routine screening labs are often drawn in the preterm population looking for issues such as anemia, cholestasis from total parenteral nutrition (TPN), and metabolic bone disease. These lab tests are drawn weekly or every other week in many institutions, with heel sticks being the most common invasive procedure performed in the NICU [15]. The purpose of this study was to assess the frequency at which an intervention was performed after drawing a routine hemogram or hepatic panel at our institution. We wanted to explore if these routine tests are necessary or if they create a burden of nonessential testing without affecting patient care. We hypothesized that many of the routine labs drawn in patients less than 34 weeks gestation would lead to no intervention.

2 Method

We conducted a retrospective study of preterm infants admitted to the NICU at New York University Langone Health between June 2013 and December 2014. The study was approved by the institutional review board. The eligible study population included any inborn infant less than 34 weeks gestation who was admitted to the NICU for prematurity and who had a length of stay longer than two weeks. Infants were excluded if they had any congenital anomalies, bleeding disorders, or syndromes and if they were transferred in after three days of life. Subjects were stratified into three groups based on their gestational age: <28 weeks, 28–31 6/7 weeks, and 32–34 weeks.

Data was collected on each subject, including baseline infant demographics and hospital course. The infant hospital course data included all routine hemogram and hepatic panel results and any intervention performed due to those results.

Routine laboratory testing was defined as testing that occurred on a weekly basis after two weeks of life, on Sunday or Monday as per our unit policy, when subjects no longer required invasive ventilation. Lab tests that were drawn on subjects while they were intubated were not considered to be screening tests but rather testing done due to the infant’s clinical status. Each subject’s respiratory status and volume of feeds was noted at the time of testing, looking for any major deviations in the infant’s clinical status that would no longer make the testing “routine”.

Intervention after a routine hemogram was defined as a packed red blood cell (pRBC) transfusion within 24 hours, in subjects who were found to be anemic, as per the transfusion guidelines used in the NYU Langone Health NICU. The guidelines use a hematocrit cut off of 40 for severe cardiopulmonary disease, 30 to 35 for mild to moderate cardiopulmonary disease, and less than 23 for stable infants requiring no support [16]. Intervention after a routine hepatic panel was defined as the initiation or termination of ursodiol within 48 hours, in subjects found to have TPN cholestasis, when the direct bilirubin level was greater than or less than 20% of total bilirubin level or the direct bilirubin level was less than two. Lastly, if a subject demonstrated signs of metabolic bone disease with elevated alkaline phosphatase levels, an intervention was defined as a change in vitamin D dosing within 48 hours. The definition of elevated alkaline phosphatase levels was at the discretion of the medical team.

Statistical analysis was performed using SPSS 20.0 (IBM SPSS Statistics for Windows, Version 20.0. Armonk, NY: IBM Corp). Continuous data was expressed as means or medians with standard deviations and categorical data was expressed as percentages. The three groups were compared using chi-square or ANOVA testing, with a p-value of ≤0.05 indicating significance.

3 Results

A total of 356 infants who were ≤34 weeks gestation were admitted to the NICU at NYU Langone Health, of which 163 were admitted for longer than two weeks. After excluding patients with congenital anomalies and those who were transferred in, a total of 135 infants were included in the study(Fig. 1).

OPEN IN VIEWER

Fig.1

Study population attrition.

Infant demographics and clinical characteristics at birth were compared across the three gestational age groups. As expected, statistically significant differences in growth parameters, length of stay, and the need for positive pressure ventilation at birth were seen. Otherwise, there were no significant differences in sample size, sex, Apgar scores, or need for chest compressions noted amongst the three groups (Table 1).

A total of 536 routine hemograms were drawn on all subjects during the course of their admission, with an average of four hemograms per subject. The <28 week group had the highest number of hemograms drawn. A total of 31 interventions were performed post-hemogram results on a total of 23 subjects in all three groups combined. The frequency of intervention after a hemogram was 8.4% in the <28 week group, 4.6% in the 28–31 6/7 week group, and 0% in the 32–34 week group (Table 2). The frequency of intervention was found to be significantly different between infants born at <32 weeks and infants born between 32–34 weeks gestation, with a p-value of 0.02.

A total of 151 routine hepatic panels were drawn on all the subjects, with an average of one hepatic panel per subject. The <28 week group had the highest number of hepatic panels drawn. A total of seven interventions were performed post-hepatic panel results on seven different subjects in all three groups combined. The frequency of intervention after a hepatic panel was 4.2% in the <28 week group, 5.7% in the 28–31 6/7 week group, and 0% in the 32–34 week group (Table 2). No significant difference was found when comparing the frequency of intervention between the three groups.

Total health expenditure in the United States, which includes money spent on healthcare and health-related activities, increases annually. Health spending totaled $74.6 billion in 1970, which has increased to approximately $3.3 trillion in 2016 [17]. The cost of prematurity remains multifaceted. According to the March of Dimes in 2015, in the acute period after birth, about $17 billion are spent annually in medical and healthcare costs for the infant and about $2 billion are spent in labor and delivery costs for the mother. Once the infant leaves the NICU, approximately $611 million are spent on early intervention services to help provide therapies for children with disabilities and developmental delays until the age of three. An additional $1 billion is spent on special education services once this population starts school [2].

Many studies have been performed looking at the costs of major morbidities in the NICU, such as bronchopulmonary dysplasia, intraventricular hemorrhage, and retinopathy of prematurity [6, 19]. Very few studies have looked at the effects of changing lower cost interventions and their contribution to the overall cost of the initial hospital stay. Therefore, we focused on the utility and cost of routine laboratory testing in this unique patient population that frequently requires a prolonged intensive care admission.

In this study, we found that infants 32–34 weeks gestation underwent no interventions following a routine hemogram or hepatic panel. This suggests that routine lab testing in this population may be unnecessary, adding cost while not affecting patient care. This is an extremely important area for intervention, given that the majority of preterm infants admitted to the NICU are born in this gestational age range [20].

Infants who were <28 weeks gestation had the highest number of routine lab tests drawn, likely due to their prolonged hospital stay, as well as the highest frequency of interventions performed after a routine lab test. In general, infants born at less than 28 weeks gestation have longer NICU courses with a higher rate of complications. Even though the subjects less than 28 weeks gestation only underwent interventions 8.4% of the time after a hemogram and 4.6% of the time after a hepatic panel, many of them required multiple interventions during their hospital stay. Based on these results, it appears that routine lab testing may be helpful in this patient population, though the ideal frequency remains unclear.

True healthcare costs are very difficult to determine, with the most costly factor being human resources. Charges are not an accurate measure of true hospital costs due to different pricing policies between institutions. A study of very low-birth-weight infants in California demonstrated that the average charge for the initial hospitalization overestimated treatment costs by 53% [10]. Despite this, charges can still be useful to help provide some estimates regarding actual costs. In this study, in the 59 subjects in the 32–34 weeks group there were a total of 70 hemograms and 4 hepatic panels that were drawn that had no intervention performed post results. According to standard charges at NYU Langone Health, there could have been approximate savings of $9,660 had these lab tests never been performed.

A major limitation of this study includes its retrospective nature and its small sample size, especially in the 32–34 weeks group. In order to strengthen the results, more data needs to be collected on more infants over a longer period of time. Another limitation to this study is that despite the fact that many institutions draw routine screening labs on preterm infants, there are variations amongst NICUs regarding how frequently these lab tests are drawn and in which patients.

Different NICUs also have varying guidelines regarding pRBC transfusions, as well as other possible interventions for anemia such as the use of erythropoietin and iron. We do not use erythropoietin routinely at our institution, and usually iron is initiated in all preterm infants after they reach full feeds, not in response to a hemogram. Other institutions may also have some variability in the initiation and modification of vitamin D, the use of multivitamins, and ursodiol dosing. Therefore, our findings may not be completely generalizable to all institutions. Lastly, in this study, though we had data regarding level of respiratory support and feeding volume at the time of each test, we did not assess other variables that may affect the frequency of intervention after a laboratory test, such as frequency of apnea, bradycardia, or desaturation events, or the presence of tachycardia.

Further studies investigating all the processes involved in obtaining a laboratory test could provide more insight into true costs. This includes the cost of the supplies, the nursing effort to obtain the sample, the laboratory personnel to run the test, physician interpretation; in addition to the immeasurable cost of discomfort to the infant. There is potential to perform a randomized control trial comparing infants who had routine lab testing against those that did not and studying differences in short and long-term outcomes, as well as a cost analysis.

In conclusion, it is crucial for physicians to be mindful when ordering tests and procedures on patients in order to help battle the rising costs of the healthcare system, and more importantly to most efficiently optimize patient care. We found that routine weekly laboratory testing may be unnecessary in patients greater than 32 weeks gestation at birth when screening for issues of prematurity, however larger studies are needed to deem if this is a safe practice and to further examine patient outcomes.

Disclosure statements

Funding Source: No external funding for this manuscript.

Financial Disclosure: The authors have no financial relationships to disclose.

Conflict of Interests: The authors have no conflicts of interest to disclose.