The Nationwide Emergency Department Sample (NEDS) database was utilized as the data source for this study. NEDS is comprised of a sample of roughly 20% of hospital-owned ED and inpatient units in the US, and is provided by the Healthcare Cost and Utilization Project (HCUP). ^, It is the largest ED database in the US, encompassing ∼30 million annual pediatric ED visits from 990 hospitals spanning 37 states. It provides appropriate sample weights to obtain national estimates. NEDS data specifically includes data on ED visits that result in admission, transfer, or discharge. The database utilizes encounter-level data rather than patient-level data; thus, a single patient may be represented by multiple ED visits within the database. An ED to ED transfer is counted as a single encounter. In order for an ED encounter to be captured, they must be included in the NEDS sample. Admissions linked to an ED encounter were included in our analysis. De-identified data for this study were obtained from HCUP. This analysis utilizes a de-identified data set, and as such, this study was exempt from IRB review in accordance with the Common Rule (45 CFR 46.102(f)).

This is a multicenter retrospective cohort study utilizing the NEDS database to evaluate outcomes of ED encounters in pediatric patients (age ≤18 years) with SVT. All ED encounters between 2006 and 2018 for patients age ≤18 years with a primary diagnosis code for SVT ( International Classification of Diseases Clinical Modification, Ninth Revision [ICD-9-CM] code 427.0, and Tenth Revision [ICD-10-CM] code I47.1) were included in this analysis. Encounters for patients with structural heart disease (with the exception of patent ductus arteriosus), including hypertrophic cardiomyopathy, were excluded (Supplementary Table S1 for excluded ICD-9 and ICD-10 codes). To identify diagnoses, ICD-9 codes were used for encounters between January 2010 and September 2015, and ICD-10 codes were used for encounters starting in October 2015. Applying a previously described methodology, hospitals were classified as pediatric if more than 70% of total visits were accounted for by children.

The primary outcome was hospital admission (further classified as hospital admission, transfer, and a composite of admission or transfer). Secondary outcomes included inpatient length of stay; mortality (encompassing both ED and in-hospital mortality); and SVT-related procedures (including cardioversion, electrophysiology study (EPS), and catheter ablation).

The primary exposure variable was insurance status (government, private, other, or missing/unknown). In NEDS, insurance payor is specified as one of the following: Medicare, Medicaid, private insurance, self-pay, no charge, other, missing, invalid, or unavailable from source. For this study, government payor was defined to include Medicare and Medicaid; private payor to include private insurance; other to include self-pay and no charge; and missing/unknown to include other, missing, invalid, or unavailable from source.

The following patient characteristics for ED encounters were evaluated: age in year, sex, patient residence (urban versus rural), and national quartile of median household income for patient ZIP code. Hospital characteristics included geographic region (Northeast, Midwest, South, West) and teaching status (metropolitan/teaching, metropolitan/nonteaching, or nonmetropolitan). We also evaluated for associated secondary diagnosis codes for arrhythmia conditions, including Wolf-Parkinson-White syndrome (WPW), other arrhythmias (ie, ventricular fibrillation, ventricular tachycardia, cardiac arrest, atrial fibrillation, atrial flutter) and other cardiac comorbid conditions (ie, dilated cardiomyopathy and heart failure). Unfortunately, patient race and ethnicity were not available within the NEDS database during our study period, which creates a limitation in our analysis, as these can have confounding effects on our findings.

All analyses performed utilized appropriate survey weights given NEDS’ survey design. ^,, Survey weights are used to produce national and regional estimates from the given data sample. Each record is assigned a weight that is derived by a sampling design that adjusts for region, trauma center designation, urban/rural location, teaching status, and hospital ownership status. The survey weights are then calculated to account for unequal probabilities from the above stratifications, and then adjusted again to correct for nonresponse/participation. The last adjustment accounts for matching the weights to population totals from the U.S. Census Bureau to accurately reflect the national population. Descriptive statistics are reported as weighted frequencies and percentages for categorical variables, and as median and interquartile range (IQR) for continuous variables. National estimates of ED visits for SVT among patients age 0 to 18 years across the years 2006 to 2018 were calculated based on primary insurance payor. We further sub-stratified by age group (<1 year, 1-4 years, 5-8 years, 9-12 years, 13-18 years). Chi-Square analysis was performed for ratio comparisons between insurance payor groups. We assessed for associations between primary insurance payor and other patient and hospital characteristics to guide our logistic regression analysis. We evaluated the association between insurance payor and outcomes using survey-weighted logistic regression, including the covariates that had a significant interaction with primary insurance payor, which included patient age, hospital teaching status, hospital region, patient residence location, and national quartile of median household income for patient ZIP code. We performed an additional analysis excluding the age <1 group from the model. The relationship between hospital SVT volume and SVT encounter disposition (admission or transfer) was evaluated with Pearson correlation testing. To calculate survey weighted frequencies and execute statistical tests, we utilized SAS’s PROC SURVEYFREQ. The default assumption within this package is that missing values are missing completely at random (MCAR), however, we utilized the not missing completely at random (NOMCAR) option as it provides more conservative variance estimates. All analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC), and a 2-sided P <.05 was considered statistically significant.

There were a total 3,425 unique hospitals included in NEDS over the study period; approximately 1% (37/3425) of hospitals met the aforementioned definition for pediatric classification (Supplementary Table S2). Of 378,250,593 ED encounters in NEDS during the study period, 38,896 (0.01%) were for SVT in patients ≤ 18 years old. Of these encounters, 52% were in female patients and 53% with private insurance. Adolescents (age 13-18 years) comprised 46% of these encounters. The median age at these encounters was 11.5 years (IQR 6-15 years). Most patients lived in an urban area (78%) and presented to metropolitan teaching hospitals (47%). Of note, 1434 (3.7%) had missing/unknown insurance payor status. Demographic data are summarized in Table 1 .

The proportion of pediatric SVT-related ED encounters to total pediatric encounters was consistent (0.005%-0.008%) from 2006 until 2015, after which the proportion increased significantly to 0.021% to 0.024% ( Table 2 ). Of note, the relative number of pediatric hospitals represented in the NEDS database remained constant between these years at 0.84% (8/953).

A secondary diagnosis of WPW was observed in 4.3% of SVT encounters. Otherwise, clinically significant arrythmia and cardiac comorbidities were rare among SVT-related encounters, with a prevalence of <1%.

When comparing government and private insurance payor groups, there were significant differences in patient and hospital characteristics, including patient residence ( P <.0001), patient income ( P <.0001), hospital region ( P <.0001), and hospital teaching status ( P < .001) ( Table 3 ). There were also differences in patient age by insurance payor. In the government insurance group, there was higher proportion of younger age groups relative to private insurance; in contrast, there was a higher proportion of patients age 13 to 18 in the private insurance group (48.1% vs 41.5% in government, P <.001). Supplementary Table S3 depicts these differences when including all insurance payor groups.

Table 3

Comparison of patient and hospital characteristics by primary insurance payor for SVT-related ED encounters

Characteristic	Government (N = 14942)	Private (N = 20563)	* P -value
Sex			.13
Male	7,394 (49.5%)	9,752 (47.4%)
Female	7,548 (50.5%)	10,801 (52.5%)
Missing	0 (0%)	10 (0.1%)
Age, median (IQR) Age group	10.8 (4.5, 14.7)	11.8 (6.5, 15.1)	<.001
<1 year	1,575 (10.5%)	1,636 (8.0%)
1-4 years	1,961 (13.1%)	1,820 (8.9%)
5-8 years	2,166 (14.5%)	2,884 (14.0%)	<.001
9-12 years	3,039 (20.3%)	4,339 (21.1%)
13-18 years	6,201 (41.5%)	9,884 (48.1%)
Region of hospital			<.0001
Northeast	421 (2.8%)	1,381 (6.7%)
Midwest	641 (4.3%)	1,678 (8.2%)
South	1,110 (7.4%)	1,467 (7.1%)
West	629 (4.2%)	1,289 (6.3%)
Missing	12,141 (81.3%)	14,748 (71.7%)
Teaching status of hospital			.0009
Metropolitan-nonteaching	4,757 (31.8%)	7,436 (36.2%)
Metropolitan teaching	7,149 (47.8%)	9,630 (46.8%)
Nonmetropolitan	3,036 (20.3%)	3,497 (17.0%)
Patient residence			<.0001
Urban	11,267 (75.4%)	16,682 (81.1%)
Rural	3,650 (24.4%)	3,819 (18.6%)
Missing	25 (0.2%)	62 (0.3%)
† Income	1.4 (1.0, 2.3)	2.4 (1.4, 3.2)	<.0001

Only gold members can continue reading. Log In or Register to continue

Share this:

• X
• Facebook

Related

Related posts:

1. Design and rationale of the GUARD-OAC: A randomized controlled trial evaluating proton pump inhibitor cotherapy for gastrointestinal protection in patients requiring direct oral anticoagulants
2. Low levels of atherogenic lipoproteins and incident atherosclerotic cardiovascular disease: A pooled cohort primary prevention study
3. Comparative performance of large language models on cardiovascular certification simulation exam
4. Evidence supporting the role of longitudinal NT-proBNP assessment for risk stratification in adults with congenital heart disease

Stay updated, free articles. Join our Telegram channel

Join