UCLA

Introduction: Due to advances in medical knowledge and technology, life expectancy has increased for many child-onset complex chronic conditions including sickle cell disease (SCD). As a result of living longer and reaching age-dependent cut-offs for insurance eligibility, a greater number of young adult SCD survivors may experience insurance loss, loss of access to health care and consequently have higher adverse health events and greater utilization of hospital services. This is a possible consequence for individuals dependent on public programs such as the California Children's Services (CCS) program in which eligibility ends on the 21st birthday.

Purpose: The primary purpose of this study was to examine variation in hospital and emergency care use between youth (ages 14-17 years) and young adults (ages 18-20 years and ages 21-26 years). Secondarily, the purpose was to examine variations in utilization by social disadvantage as defined by community poverty level, insurance status, and travel distance from home to nearest hospital.

Data and Methods: This retrospective quantitative analysis utilized confidential patient level discharge data (PDD) and emergency department (EDD/ASCD) patient encounter data collected by the Office of Statewide Planning & Development (OSHPD) from 2006-2011. Patients were included in the study population if they had either a primary or secondary ICD-9 code of sickle cell anemia, sickle cell thalassemia with crisis, and sickle cell thalassemia without crisis. Some of these patients had hospitalizations that were not coded for SCD. Hospitalizations with matching patient identifiers to hospitalizations selected with SCD were also included. Patients were categorized according to age group based upon age and relative eligibility to CCS/Medicaid services. Multi-level zero-truncated negative binomial regressions, generalized estimating equations with negative binomial link, and multilevel logistic regressions were performed to assess the association of age and socio-economic factors on count of index hospitalizations, readmissions within 30 days, length of stay (LOS), and count of ED visits after controlling for patient demographics, hospital level characteristics, or zip code level poverty status.

Results: 1,825 patients were identified accounting for 13,257 hospitalizations in the PDD dataset and 27,001 ED visits representing 2,314 patients in the EDD dataset. Twenty-five percent of index hospitalizations were followed by at least one readmission within 30 days of last discharge. The population hospitalized had a mean LOS of 6.2 days. The number of hospitalizations and visits steadily increased between age groups 14-17 and 18-20 with the greatest increase occurring after age 21. Over half of all SCD hospitalizations (56%) and ED visits (66%) during the six year period were for patients ages 21-26. Statistically significant differences in number of index admissions were observed between patients aged 21-26 [IRR=1.60; 95% CI: 1.34-1.95; p value=0.00] and 14-17 when adjusting for individual and contextual factors. Age group 21-26 was associated with higher odds of readmission [OR= 1.14; 95% CI: 0.93-1.40; p value=0.02], longer LOS [IRR=1.02; 95% CI: 1.01-1.03; p value=0.00], and increased number of ED visits [IRR=2.27; 95% CI: 1.88-2.75; p value=0.00] relative to (which age group). Some SES factors were significantly associated with all outcomes. Lack of insurance was associated with significantly lower predicted index hospitalizations [IRR =0.34; 95% CI: 0.27-0.44; p value=0.00], readmission [OR= 0.64; 95% CI: 0.48-0.88; p value=0.05], LOS [IRR= 0.96; 95% CI: 0.94-0.98, p value=0.00], and ED visits [IRR =035; 95% CI: 0.30-0.44; p value=0.00] than those with Medicaid/other government insurance. Residing in zip codes with a higher concentration of poverty was associated with higher odds of a readmission [OR= 0.80; 95% CI: 0.65-0.99; p value=0.05].

Conclusions: Hospitalizations, ED visits and LOS increases with age among SCD patients. Lack of insurance was associated with decreased hospitalizations, ED visits and shorter LOS. Understanding drivers that influence higher hospitalization rates, longer LOS, and greater ED care seeking behavior as youth grow older should be further explored. Drivers may include increasing severity of illness, delays in accessing primary care, growing difficulties in self-management as an adolescent enters into adulthood, and changes in access to care resulting in losing insurance coverage for specialists.