309 - Increased risk of complicated pneumonia in children with higher deprivation and rurality

Poster Number: 309
Publication Number: 309.602

Background: Complicated community acquired pneumonia (CCAP) is defined as the presence of parapneumonic effusion, empyema, necrotizing pneumonia or abscess in children with pneumonia. Risk factors for CCAP, including those associated with social determinants of health (SDoH), are incompletely defined.

Objective: To determine SDoH risk factors for development of CCAP.

Design/Methods: This was a retrospective, case-control study of children aged 3 months to 19 years hospitalized with a diagnosis of pneumonia between January 2012 and December 2022. Cases included children with parapneumonic effusion, empyema, necrosis, or cavitation. Controls were those with community acquired pneumonia without complication. Primary exposures included smoke exposure, childcare attendance, area of deprivation index (ADI), presence of segregation and degree of rurality/urbanicity. Covariates also included sex, race, ethnicity, primary language, payor, readmission and length of stay. Children with multilobar pneumonia, discharge or admission within the last 30 days, possible aspiration, or chronic disease predisposing to respiratory infections were excluded. Routine statistical analyses and multivariable logistic regression was performed with SAS 9.4. Cases and controls were matched at a 1:2 ratio by age, BMI or weight for length Z-score, hours of fever prior to admission and malnourished status utilizing propensity score matching.

Results: In total, 380 controls and 220 cases were included after matching (mean age 60 months, mean fever duration prior to admission of 96 hours for cases and 72 hours for controls and mean BMI or weight for length Z-score of 0.71 and 0.59 respectively after matching). In bivariate analyses, cases were more likely to have a longer length of stay (5.0 vs 2.0 days, p< 0.001) and longer duration of fever after admission (33 hours vs 4 hours, p< 0.001). Multivariable logistic regression demonstrated increased odds of cases having a higher state ADI (OR 1.10, p 0.004) and living in a less urban area (OR 1.87, p 0.004). There was no significant association seen with sex, race/ethnicity, readmission, primary language, payor, smoke exposure or daycare/school attendance.

Conclusion(s): We identified a small increased risk of CCAP compared to controls in children with a higher state ADI rank and living in more rural areas. CCAP was also associated with longer length of stay and post-admission fever duration. The factors leading to increased risk of CCAP in rural children is an area for further study.