Children with high blood pressure were not followed up after paediatric emergency department visits despite guidelines

Abstract

The prevalence of high blood pressure (BP) is increasing in children and has been associated with early atherosclerosis, adult hypertension and cardiovascular disease.¹ International guidelines recommend measuring BP at least annually when children reach 3 years old. However, adherence has been low and there has been evidence that high BP has been underdiagnosed in paediatric patients.^{1, 2} Measuring, recognising and diagnosing high BP in children is essential. Although BP is often measured in paediatric emergency departments (PEDs), it is not clear how these levels should be used to screen for high BP.

This multicentre retrospective cohort study focused on all children aged 1–18 years who visited Israeli PEDs run by Clalit Health Services between 1 January 2016 and 30 June 2019. It assessed the frequency of follow-up BP measurements by primary care providers. Children were included if their last BP measurement before their PED discharge was elevated. We did not include hospitalised patients or those with an existing diagnosis of hypertension or antihypertensive treatment. Clalit Health Services is Israel's largest healthcare organisation, which cares for more than 50% of the population³ and has seven PEDs throughout Israel. It also integrates patients' medical records.

High BP was defined as a systolic or diastolic BP of ≥90th percentile for the 5th height percentile for age and sex or >120/80 mmHg for children aged 13 and above. These followed the recommendations in the 2017 American Academy of Pediatrics guidelines for screening high BP.¹ Overweight and obesity were defined as a body mass index (BMI) of the ≥85th and ≥95th percentile for age and sex. The primary outcome was the BP follow-up measurement rate in community clinics 1 and 3 months after the child's PED visit. Variables associated with this outcome were also assessed. Descriptive statistics have been used to summarise the demographic and clinical characteristics of the study population. Multivariable logistic regression was conducted to identify independent variables associated with community follow-up BP measurements. Adjusted odds ratios (aORs) and 95% confidence interval (CIs) are reported. Cohen's d effect size and odds ratio were calculated to mitigate type I errors. The statistical analysis was carried out using R version 4.1.3 (R Foundation, Vienna, Austria). Clalit's institutional review board approved the study and informed consent was not needed.

BP was measured in 163 025 (40.3%) of the 404 846 discharged children and we included the 60 004 (36.8%) with high BP measurements in our study. Their demographic and clinical characteristics are presented in Table 1. Multivariable regression analysis showed that older age (aOR 1.18, 95% CI 1.16–1.21), low socioeconomic status (aOR 1.68, 95% CI 1.34–2.11) and comorbidities (aOR 1.94, 95% CI 1.7–2.22) were associated with community follow-up BP measurements within 3 months.

The frequency of high BP in the PED was 36.8%, compared to 55% in another study.⁴ The frequency in the general paediatric population has been reported to be around 7%,¹ but pain and anxiety are common in PEDs and known to elevate BP. Using the 5th percentile for height may have also contributed to the higher rate.

Our main findings were that only 1.1% and 2.5% of patients with high BP in the PED had follow-up BP primary care measurements within one and 3 months, respectively. We could not find a comparable PED study, but one outpatient study reported that 20.8% of children with high BPs were followed up within 1 month. Our low rate was attributed to under-recognition of high BP, assuming that high BP measurements in children were erroneous and paediatricians' perceptions that high BP thresholds were too low.⁵ We can also suggest a few more explanations for the rates in our study. First, the patient, parent and primary care provider may have been unaware of the high BP if the PED discharge letter did not explicitly state this and recommend follow-up BP measurements. Second, primary care providers have limited time and may only focus on urgent post-PED discharge issues. In addition, they may think that high BP is uncommon in previously healthy children, as older age and comorbidities have been associated with community follow-up BP measurements. Other children may need more frequent BP measurements, not just those with known risk factors for high BP, including obesity, diabetes and kidney disease. We strongly recommend that primary care providers measure the BP of children aged 3 years or more, regardless of the risk factors, as recommended by the guidelines.¹

One study limitation was its retrospective design, which prevented determining causality, but allowed associations. In addition, we could not account for some variables that may influence BP in the PED, such as the main complaint. Finally, using the 5th height percentile, which is recommended for screening, is less specific in determining high BP than using the patient's actual height.

Despite the guidelines, there was a very low follow-up rate for children with high BP levels during their PED visit. While PEDs are not ideal locations for high BP screening, abnormal BP results must not be ignored and significantly higher community follow-up measurements are needed. PED discharge letters need to underline the importance of repeated BP measurements within 1–2 weeks of discharge, but this would require better recognition of high BP by PEDs.

Lebel Asaf: Conceptualization; writing – review and editing; methodology; supervision; writing – original draft. Abu-Ras Muhammad: Writing – review and editing; writing – original draft; data curation. Gilad Chayen: Writing – review and editing. Sireen Sharif: Investigation; methodology; formal analysis; writing – review and editing. Ron Jacob: Conceptualization; writing – original draft; writing – review and editing; methodology; supervision.

None.

None.