Short term non-programmed hospital visits after NICU discharge

The growing technical and scientific developments in neonatal care in recent years have led to an increase in the survival rate of premature newborns (NB) and a consequent increase in the number of NB discharged from Neonatal Intensive Care Units (NICUs) [1,2]. Hospital readmissions after discharge affect the quality of life of both the newborns and their families and imply increasing hospital costs and familiar costs related to parental absenteeism from work [1, 3-5]. 

Hospital readmissions have been the subject of several studies using different methodologies and evaluating different periods after discharge [1-2, 4-11]. Several studies have indicated that short-term readmissions are those that best reflect the appropriateness of medical decisions at discharge and those with the greatest potential to be prevented [1]. Emergency department (ED) observation rates have been described in the literature for different time periods, with an observation rate of 5.4% for NBs with a GA of 31-33 S in the period up to 30 days after discharge from the NICU [11]. Regarding readmission rates in the first 30 days, the reported figures vary widely, with rates of 5-30.7% [4, 6, 9-11]. 

The different clinical characteristics of NBs discharged from NICUs may be associated with different risk factors for readmission and may require more targeted approaches and planning at discharge [1, 7]. This study aims to identify the factors associated with unplanned hospital readmissions after discharge and how current clinical practice can be modified to reduce them. The objectives of this study are to examine the rate of short-term unplanned hospital visits in NB with gestational age (GA) < 34 weeks (wk) and to identify factors associated with readmission.

Unicentric, observational, retrospective, and longitudinal study over a period of 14 years conducted at the Bissaya Barreto Maternity.

The inclusion criteria were PT NB with GA < 34 wk, admitted to the NICU between the 1st of January 2009 and the 31st of December 2022, discharged home and with an unplanned hospital visit up to 30 days after discharge.

Data were collected by reviewing the NBs' clinical records, hospital summaries and electronic health records (RSE^®). The clinical and demographic characteristics of the two groups were compared according to whether they attended ER or not (Table 1) or were re-hospitalized or not (Table 3).

Fetal growth restriction was defined as an estimated fetal weight less than the 3rd percentile for GA or less than the 10th percentile for GA associated with altered dopplers [12]. Complete course of antenatal corticosteroids consisted of two maternal instramuscular injections of 12 mg betamethasone given every 24 hours or four instramuscular injections of 6mg dexamethasone given every 12 hours [13]. Peri-intraventricular hemorrhage was classified according to Volpe definition [14]. Bronchopulmonary dysplasia was defined by the need of supplementary oxygen therapy at 36 weeks of postmenstrual age (PMA) [15]. Sepsis was diagnosed in the setting of compatible symptoms associated with leucocyte count > 30.000/mm3 ou < 5.000/mm3 and/or C reactive protein >2mg/dL, with or without a positive hemoculture [16]. Necrotizing enterocolitis was classified according to the modified Bell’s staging criteria [17].

Social risk was considered if there was a referral for social service assessment described in the patient's medical record or a referral for hospital consultation due to the identified social risk.

Firstly, NBs who were observed in the emergency department for up to 30 days after discharge from the NICU were analysed, and secondly, NBs who required hospitalisation were assessed.

The study was approved by the Ethics Committee of the Coimbra Hospital and Universitary Centre.

The data was statistically analysed using IBM SPSS Statistics Viewer® for Machintosh version 28.0.1.0. Normality of quantitative variables was assessed using the Kolmogorov-Smirnoff test. None of the variables examined showed normality, so they were expressed as median and interquartile range (IQR), followed by the Mann-Whitney U test. Qualitative variables were expressed as frequencies and percentages and analysed using the Chi-squared test or Fisher’s test. A p value of < 0.05 was considered statistically significant.

The study population included 298 NBs. These patients were assessed for observation in the ED and then for re-hospitalization.

In the 30 days after discharge, 47 patients (15.8%) were observed in the ED. The median number of days between discharge and observation in the ED was 14 days (IQR 1-30). Table 1 shows the demographic and clinical characteristics of the patients seen in the ED. Postmenstrual age at discharge was similar in NBs seen in the ED (median 35 years) and those not seen in the ED (median 36 years). A statistically significant association was found with birth weight (BW) < 1000 g (p < 0.001; OR 3.588) and ED observation. A diagnosis of BPD during the initial stay in the NICU was also statistically associated with short-term observation in the ED (p=0.012; OR 3.473). The other factors assessed in the study did not show a statistically significant association with short-term readmission. 

Table 1: Demographic and clinical characteristics in relation to observation in the Emergency Department up to 30 days after discharge from the NICU 

Abbreviations: BPD - Bronchopulmonary dysplasia; FGR – Fetal Growth Restriction; NEC - Necrotizing enterocolitis; NICU - Neonatal Intensive Care Unit; PIVH - Peri-intraventricular hemorrhage; Postmenstrual age - PMA

Regarding the causes of observation in the ED, shown in Table 2, the overall sample showed a predominance of gastrointestinal/nutritional problems (n=18, 38.3%).

Table 2: Causes of observation in the Emergency Department up to 30 days after discharge from the NICU

Evaluating the subgroup of 44 PT NB with BW < 1000 g, it was found that 15 (34.1%) of them resorted to the ED in the short term. The main cause of observation in the ED in NBs with BW < 1000 g was gastrointestinal/nutritional problems (n=9, 60.0%), namely neonatal gastroesophageal reflux (n=5, 33.3%) and failure to thrive due to feeding difficulties, infantile colic, unspecified constipation and vomiting of the NB, both with n=1 (6.7%). The other pathologies recorded were other non-infectious (n=4, 26.7%), respiratory infectious (2 (13.3%) and surgical (n=1, 6.7%). 

Evaluating the subgroup of 22 NB diagnosed with BPD during their initial stay in the NICU, it was found that 8 (17.0%) of them resorted to the ED in the short term. The main cause of observation in the ED for NBs with BPD was gastrointestinal/nutritional pathology (n=4, 50.0%), namely neonatal gastroesophageal reflux (n=2, 25.0%) and infantile colic and unspecified constipation, both with n=1 (12.5%). Other causes of observation were respiratory infectious pathology (n=2, 25%), surgical (n=1, 12.5%) and other non-infectious (n=1, 12.5%).

Within 30 days of discharge, 15 patients (5.0%) were re-hospitalized, all of whom had previously been seen in the ED. Table 3 shows the demographic and clinical characteristics of the sample of NBs who were re-hospitalized. The median number of days between discharge and re-hospitalization was 12 days (IQR 4-26) and the median length of hospital stay was 9 days (IQR 1-20). 

Table 3: Demographic and clinical characteristics in relation to re-hospitalization up to 30 days after discharge from the NICU 

Abbreviations: BPD - Bronchopulmonary dysplasia; FGR – Fetal Growth Restriction; NEC - Necrotizing enterocolitis; NICU - Neonatal Intensive Care Unit; PIVH - Peri-intraventricular hemorrhage; Postmenstrual age - PMA 

There was a tendentially higher, but not statistically significant relationship between discharge in autumn/winter and short-term re-hospitalization (p=0.060). The other variables studied also showed a similar distribution between short-term re-hospitalized and non-re-hospitalized NBs. Regarding the causes of short-term re-hospitalization, described in Table 4, the more prevalent were infectious respiratory pathologies (n=5; 33.3%), namely pneumonia (n=2), acute bronchiolitis (n=2) and acute nasopharyngitis with apnea (n=1), and other non-infectious pathologies (n=5; 33.3%).

Table 4: Causes of re-hospitalization up to 30 days after discharge from the NICU

PT NBs have increased vulnerability and are at greater risk of readmission after discharge. Hospital readmissions have already been evaluated in several studies, albeit with very different methodologies. These methodologies differ in terms of the definition of readmission, the time period between discharge and readmission and the characteristics of the patients included in the sample [1, 2, 4-11]. 

In the present study, the rate of observation in the ED (15.8%) was higher than reported in a 2012 Californian study with NBs without major congenital anomalies who were discharged from the NICU to their homes [11]. This difference may be due to the fact that the sample size was much larger and that the NBs in this study were all assessed by a pediatrician or home health service 2 days after discharge. 

Regarding the re-hospitalization rate, a rate of 5.0% was obtained. A 2014 study with PT of 23-34 wk of GA data from California showed a similar median 30-day re-hospitalization rate (~5%) [10]. Another 2013 study with data also from California, had a 30-day re-hospitalization rate of 6.0% for PT with 23-33 wk of GA [9]. On the other hand, a 2016 study in Australia with NB of GA 24-33 wk showed rates of 30.7% for GA 24-27 wk, 19.9% for GA 28-31 wk and 10.1% for GA 32-33 wk [6]. 

Regarding the variables with a statistically significant association with re-hospitalization, none of the variables examined showed this association. Only autumn/winter discharge showed a tendentially higher, but not statistically association with short-term re-hospitalization (p=0.060). A 2016 study in Ohio also found no association between BW and re-hospitalization up to 30 days after discharge [4]. The same study found no association between BPD and re-hospitalization when patients with BPD were treated by a team specialized in this pathology [4]. Factors that have been associated with 30 days re-hospitalization in other studies were: longer duration of initial NICU stay [4-5] and lower GA [5, 9, 11]. 

The most common causes of re-hospitalization were infectious respiratory diseases (33.3%) and other non-infectious diseases (33.3%). The fact that infectious respiratory pathology is one of the most prevalent pathologies leading to re-hospitalization may be related to discharge in autumn/winter and this being the time of year when there is a higher prevalence of infectious respiratory pathology. A 2020 review by Hannan et al. [5] states that PTs with a GA < 33 wk are most commonly re-hospitalized for acute bronchiolitis in the first month after discharge. The same review also described acute bronchiolitis as a leading cause of re-hospitalization within 14 days of discharge in PTs with a GA < 32 wk [5]. A 2013 study with data from California in NBs with GA 23-44 wk described respiratory infections as the most common cause of re-hospitalization in all GA [9]. 

We consider the strength of this study to be the fact that it evaluated all PT NBs with a GA < 34 wk who were discharged home from the NICU of a specialized perinatal support hospital over a period of 14 years, and the fact that no NB lost follow-up during the period evaluated.

However, when analysing the results obtained, some limitations of the study should be taken into account. These limitations relate to the fact that as a retrospective study the data collection was limited to the information recorded in the clinical file. In addition, it was not possible to access patient examinations carried out in primary health care centres or private health care facilities. The social risk assessment did not specifically assess the socioeconomic power of the family [6-7, 9].

We concluded that gastrointestinal/nutritional problems were the most common cause of ED observation on the overall sample and in the subgroups of NB with BW < 1000 g and BPD. This alerts us to the importance of investing in caregiver training on these issues prior to discharge from the NICU, as well as improving the adequacy of the Primary Care services response.  The readmission rate in this study was lower than that generally described in the literature. The majority of NBs seen in the ED did not present with severe pathologies that would have led to their re-hospitalization and should have been assisted in a Primary Care setting. We therefore emphasize the importance of reorganizing Primary Care to improve its accessibility and response to NBs and their families, with the aim of reducing the inappropriate use of ED associated with high costs and infectious risk in this vulnerable population.

The authors do not state any acknowledgment.

All authors stated contributed to this article according to the International Committee of Medical Journal Editors guidelines.

No source of funding was obtained to publish this article.

The data set used during the study is available to the authors.

Ethics approval for the development of this study with informed consent waiving was obtained from Ethics Committee of the Coimbra Hospital and Universitary Centre (reference number OBS.SF.116-2023).

The authors agree to submit this article to the journal and, if accepted, to its publication in this journal. The authors warrant that this article is original, dose not contravene on any copyright and any other proprietary right of any third party, is not under consideration by another journal, and has not been previously published.

The authors express that they have no competing interests.

1. Bernardo J, Keiser A, Aucott S, Yanek LR, Johnson CT, et al. (2021) Early Readmission following NICU Discharges among a National Sample: Associated Factors and Spending. American Journal of Perinatology 21287.
2. Lee JH, Chang YS (2015) Use of medical resources by preterm infants born at less than 33 weeks’ gestation following discharge from the neonatal intensive care unit in Korea. Journal of Korean Medical Science 30: S95–S103.
3. Petrou S (2005) The economic consequences of preterm birth during the first 10 years of life. BJOG: An International Journal of Obstetrics and Gynaecology 112: 10-15.
4. Bapat R, McClead R, Shepherd E, Ryshen G, Bartman T (2016) Challenges, successes and opportunities for reducing readmissions in a referral-based children’s hospital NICU. Journal of Neonatal-Perinatal Medicine 9: 433-440.
5. Hannan KE, Hwang SS, Bourque SL (2020) Readmissions among NICU graduates: Who, when and why? Seminars in Perinatology 44: 151245.
6. Stephens AS, Lain SJ, Roberts CL, Bowen JR, Nassar N (2016) Survival, Hospitalization, and Acute-Care Costs of Very and Moderate Preterm Infants in the First 6 Years of Life: A Population-Based Study. Journal of Pediatrics 169: 61-68.
7. Rubinos LH, Foster CC, Machut KZ, Snyder A, Simpser E, et al. (2022) Risk factors for hospital readmission among infants with prolonged neonatal intensive care stays. Journal of Perinatology 42: 624-630.
8. Braun D, Braun E, Chiu V, Burgos AE, Gupta M, et al. (2020) Trends in neonatal intensive care unit utilization in a large integrated health care system. JAMA Network Open 3: 1-12.
9. Ray KN, Lorch SA (2013) Hospitalization of Early Preterm, Late Preterm, and Term Infants During the First Year of Life by Gestational Age. Hospital Pediatrics 3: 194-203.
10. Lorch SA, Passarella M, Zeigler A (2014) Challenges to measuring variation in readmission rates of neonatal intensive care patients. Academic Pediatrics 14: 47-53.
11. Kuzniewicz MW, Parker SJ, Schnake-Mahl A, Escobar GJ (2013) Hospital Readmissions and Emergency Department Visits in Moderate Preterm, Late Preterm, and Early Term Infants. Clinics in Perinatology 40: 753-775.
12. Sociedade Portuguesa de Neonatologia (2018) Consenso Clínico “Restrição de Crescimento Fetal.
13. Sociedade Portuguesa de Pediatria (2012) Prescrição Pré-natal de Corticoides para reduzir a Morbilidade e Mortalidade Neonatais.
14. Volpe J (2018) Preterm Intraventricular Hemorrhage. In Neurology of the newborn 25: 3-11.
15. Sociedade Portuguesa de Pediatria (2009) Consenso Clínico “Displasia broncopulmonar – prevencao e tratamento na UCIN.
16. Secção de Neonatologia da Sociedade Portuguesa de Pediatria (2014) Consenso Clínico Procedimento no recém-nascido com risco infeccioso.
17. Bell MJ, Ternberg JL, Feigin RD, Keating JP, Marshall R, et al. (1978) Neonatal necrotizing enterocolitis. Therapeutic decisions based upon clinical staging. Annals of Surgery 187: 1-7.