Utilization of Resources and Characteristics of Patients with Recurrent Readmissions to Children’s Hospitals

Hospital readmission is increasingly considered an indicator of quality care and a factor in rising healthcare costs. Children routinely cared for at children’s hospitals are at risk for readmission if they have complex chronic conditions (CCCs); however, to date, data that examine the impact of rehospitalization trends and the clinical characteristics of children experiencing recurrent readmissions are lacking.

Researchers conducted a retrospective cohort study of the Pediatric Health Information System (PHIS), a database of hospitalization data from 37 pediatric hospitals in the United States. The analysis included 317,643 patients (579,504 admissions) admitted in 2003 with follow-up across multiple admissions through 2008. The study describes inpatient utilization of resources, examines demographic and clinical characteristics of patients, and evaluates reasons for readmission.

The results of the study were published in the Journal of the American Medical Association [2011; 305(7):682-690].

Data collected to assess the utilization of hospital resources included the number of admissions, days spent in the hospital, and total hospital charges for each patient. Hospitalizations for psychiatry and psychology services, inpatient rehabilitation, and routine obstetric and newborn care were excluded due to difference in availability of these services in the 37 hospitals. Chemotherapy admissions were also excluded, as they are frequently protocoldriven, scheduled recurrent admissions.

Readmission intervals were not limited by calendar-year intervals but defined as any 365-day interval (1460 full 365-day intervals) during the followup period. Each child was classified into 1 of 5 categories: 0, 1, 2, 3, and ≥4 readmissions during any interval.

Analyzed demographic characteristics of the children included age, sex, race/ethnicity, and insurance type. Chronic diagnoses were categorized into 2 groups—CCCs and technology assistance—and assignments to these categories were based on diagnosis and procedure codes provided by each hospital. Ambulatory care–sensitive conditions (ACSCs), conditions that may be treatable without hospitalization through proper ambulatory care, were assessed by evaluating the diagnosis code for each hospitalization. Readmissions were further evaluated to determine if the recurrent admissions were for the same reason as a prior admission, and therefore potentially avoidable.

Using median and interquartile range (IQR), hospital resource utilization was summarized for patients in each of the 5 frequency categories. The consumer price index for hospital and related services was used to convert accrued costs to 2008 dollars. In analyzing the likelihood of a patient characteristic being associated with an incrementally increasing number of recurrent admissions, attributes with a P value >.05 were eliminated and a P<.001 was considered statistically significant.

Of the 317,643 patients admitted to the 37 children’s hospitals during 2003, 69,294 patients (21.8%) had ≥1 readmission within 365 days of a prior admission, and 9237 patients (2.9%) had ≥4 readmissions. These patients account for 18.8% (109,155) of all admissions and 23.2% ($3.4 billion) of total inpatient charges for the whole cohort throughout the follow-up period. The median time between admissions was 37 days (IQR, 21-63), indicating that the majority of the readmissions would not have been recorded using the traditional 30-day frequency interval.

As readmissions increased from 0 to ≥4, the prevalence increased for CCCs (from 23.3% to 89%; P<.001), technology assistance (from 5.3% to 52.6%; P<.001), public insurance use (from 40.9% to 56.3%; P<.001), and non-Hispanic black race (from 21.8% to 34.4%; P<.001). Readmissions associated with ACSCs decreased (from 23.1% to 14%; P<.001). Of the 2.9% of patients experiencing ≥4 readmissions, 28.5% were rehospitalized for a problem in the same organ system.

Study limitations cited by the researchers include reliance on PHIS inpatient administrative data versus clinical data, a lack of outpatient data, unavailability of psychological factors, and the inability to identify children who died in a setting other than a PHIS hospital or were readmitted to a non-PHIS hospital.

In conclusion, researchers said that the study suggests that a small group of pediatric patients accounts for a disproportionate amount of inpatient resource utilization and expenses due to recurrent readmissions; “Nearly one fourth of all inpatient bed-days and charges were attributable to 3% of admitted patients who experienced 4 or more readmissions within a 1-year period,” they commented.