Lapses in Appointment Attendance Among Children with Sickle Cell Disease in the First Five Years of Life

Introduction

Children with sickle cell disease (SCD) are at risk of negative health outcomes in early childhood, especially related to bacterial infections. Regular engagement with SCD disease management and preventive care from an early age is vital to reducing morbidity and mortality. The American Academy of Pediatrics recommends that children with SCD be seen at least annually by a multidisciplinary SCD team (AAP, 2024). There is a lack of knowledge around how frequently this care recommendation is adhered to, especially in the very young population. Children’s Healthcare of Atlanta (CHOA) has the largest pediatric sickle cell program in the country, seeing over 2,000 patients a year. We sought to understand the prevalence of and characteristics associated with lapses in appointment attendance in children with SCD through their first five years of life.

Methods

This retrospective chart review was approved by the CHOA Institutional Review Board. A cohort was identified of children with SCD who were born between 2013 and 2017 and seen in a CHOA SCD clinic within their first year of life. SCD clinic visit attendance, including telehealth visits, up to the child’s 6^th birthday was measured. A patient was considered to have lapsed attendance (LA) if they had over 365 days without being seen in clinic. Non-clinic encounters (ED visits, hospitalizations) were not considered in assessing attendance. LA patients who never returned to any CHOA facility by the end of the follow-up period were presumed to have moved out of the service area and excluded from further analysis. Patients who died or received curative therapy during follow-up were censored at that event. Demographic characteristics and healthcare utilization were compared between patients who had LA during the follow-up period versus those who did not. Time to the first instance of LA was analyzed using Kaplan Meier curves.

Results

A total of 501 children with SCD were born between 2013 and 2017 and attended SCD clinic in their first year of life. Ninety-two patients never returned to a CHOA facility after an incidence of LA, leaving a final cohort of 409 patients. The majority of the cohort identified as Black or African American (96.3%), were male (54.0%), and had a sickle cell anemia (SCA) genotype (HbSS or HbS/β⁰ thalassemia, 59.7%). In the final cohort, 39.9% of patients (n=163) had an instance of LA. Twelve patients were censored prior to having any lapse in attendance due to death (n = 6) or receiving curative therapy (n = 6). Of the LA patients, 11.7% (n = 19) had their first instance of LA after their first clinic visit. The median time to return to clinic was 501 days (IQR 423.0, 705.8), and 37.4% (n = 61) of LA patients had a subsequent episode of LA in the follow-up period. By 24 months of age, the probability of having LA was 24% (95% CI: 20%, 28%), and by 48 months the probability had increased to 37% (95% CI: 33%, 42%).

LA was associated with having a non-SCA genotype, with 57.6% of non-SCA patients experiencing LA compared with 27.9% of SCA patients (p <0.001). LA was also associated with older age at first clinic visit (p = 0.006), not having received a pneumococcal polysaccharide vaccine (PPSV) by age 3 (p = 0.012), and having a no-show clinic appointment in the year prior to LA (p =0.032). Among SCA patients, LA was associated with not being started on hydroxyurea (p <0.001), not having a history of chronic transfusions (p = 0.025), and not having a TCD before the age of 3 (p <0.001). Demographic factors (birth year, sex, race, primary language, insurance type, distance of home address from SCD clinic) and other utilization history (ED encounter, admission, or cancelled appointment in the year before the last visit) were not associated with LA.

Conclusion

At our Center, over a third of young children with SCD experienced a lapse of at least a year between clinic visits in their first five years of life. We found that there was an association between LA and having a non-SCA genotype, older age at first appointment, and a previous no-show clinic appointment. LA was also associated with not having received recommended prevention measures, and almost 40% with LA had a subsequent episode of LA. These data highlight the need for more effective programmatic approaches to reduce LA and suggest potential high risk populations to focus efforts on. Further research is needed to better understand barriers to SCD clinic attendance and outcomes associated with these lapses in care.