Pediatric to adult care co‐location transitional model for youth with sickle cell disease

Survival to adulthood for children with sickle cell disease (SCD) in the United States is above 95%, however morbidity and mortality substantially increase following departure from pediatric care.1, 2 Acute healthcare utilization (emergency department visits and hospitalization) is highest during pediatric to adult care transition (ages 18–30).2 Deficits in planning and care coordination as patients move from the pediatric to adult care setting lead to breaks in care continuity and contribute to poor healthcare outcomes. A critical step in ensuring adequate care continuity during healthcare transition is engagement with the new adult provider. Although different service models of healthcare transition have been tested in other chronic diseases, there is insufficient research in the SCD population. Recognizing the need for better integration between pediatric and adult care, we implemented a co-location model of transition to adult care. Our co-location model consists of systematic overlap between pediatric and adult care by creating two levels of intersection at both the nursing and physician levels and offers organized disease education (Supporting Information). We hypothesized that among youth with SCD who received co-located pediatric–adult transitional care, acute healthcare utilization rates would not increase after transfer. Demographics, exposure to hydroxyurea and chronic transfusion therapy, acute care visits, hospitalizations, 30-day readmissions, and health-maintenance visits occurring during pediatric (ages 16–17) and adult (ages 18–19) care were collected. We compared the rate of acute healthcare utilization before and after transfer from pediatric to adult care among youth with SCD who received co-located care, and evaluated the relation between patient knowledge of SCD-related pain with healthcare utilization. Fifty-nine participants established adult care using our co-location transitional model (Supporting Information Table S1). The median interval from leaving pediatric care and matriculation in adult care was two months, and 83% of participants engaged in adult care within the three-months. The overall acute utilization was approximately the same in the pre- and post-adult care transfer periods (Table 1). Acute care visits were 89% higher and hospitalizations were 46% lower, relative to the pre-transfer care period. Thirty-day readmission rate was lower in the post-transfer period relative to the pre-transfer period, but not significant. Outpatient health-maintenance visits decreased by about 46%. Reported national utilization rates for SCD-related overall utilization are 3.61 visits per person-year (95% CI: 3.47, 3.75), and 1.59 (95% CI: 1.50, 1.68) and 2.02 (95% CI: 1.34, 2.10) per person-year for acute care visits and admissions among patients ages 18–30 years, respectively.2 In our cohort of similar age patients, the overall utilization, acute care, and hospitalization rates were all significantly lower (P < .001). Following in-clinic educational sessions, both adolescents and young adults significantly improved their knowledge of SCD pain (Supporting Information Table S2). Pre-education median pain knowledge scores were significantly higher during adult care than during pediatric care, suggesting knowledge retention from pediatric to adult care. There was evidence of decreased hospitalizations in the adult setting when SCD pain knowledge scores increased following the educational session, albeit with an imprecise risk estimate (Supporting Information Table S3). This finding was supported by a significant association between higher post-education pain knowledge scores and lower hospitalization rates (Supporting Information Table S3). Adults with higher knowledge scores following the in-clinic educational session had more health-maintenance visits completed (Supporting Information Table S3). In the pediatric setting, however, a decrease in health-maintenance visits was associated with higher post-educational session scores, although there was no association between pre- to post-score improvement and outpatient visits (Supporting Information Table S3). In our transitional co-location model for SCD, we found no evidence of an overall upsurge in acute healthcare utilization among emerging adults. An increase in acute care visits was balanced by a decrease in hospitalizations, resulting in no change in overall acute healthcare utilization after transfer to adult care. A significant decrease in hospitalizations and an increase in outpatient visits were associated with higher SCD pain knowledge during adult care. There is a paucity of evidence regarding healthcare transition delivery models. Current models include: (1) structured sequential model (no overlap in pediatric and adult care)3; (2) early introduction to adult providers (youth meet with adult provider prior to completion of pediatric care)4; and (3) co-location of care (joint care by pediatric and adult providers in the same setting).5 Some degree of co-location has been successfully implemented among survivors of pediatric cancer, but its efficacy has not yet been examined.6 This is the first report of the co-location transitional model in SCD. Outpatient visits decreased after transfer to adult care (Table 1). Our study was not designed to investigate predictors of outpatient visit attendance, but possible reasons for this decline could include the absence of family support, lack of transportation, and new competing priorities (eg, work or college schedules), all potentially making frequent clinic visits less feasible. Adult test scores were higher than the pediatric scores, suggesting cumulative effect of disease knowledge, resulting in a better understanding of disease and possibly contributing to the decrease in hospitalizations and increased compliance with outpatient visits. While we did not formally measure health literacy, disease knowledge may approximate it by demonstrating understanding of disease complications and possibly higher self-efficacy. Future studies in this population will include formal health literacy and self-efficacy measurements to investigate associations with clinical outcomes. Our sample size was relatively small, limiting our power to detect small differences between pre- and post-transfer of care, however because each participant contributed at least four years of follow-up, our analyses were based on rates of encounters per person-year, increasing our effective sample size. Because data for those youth not receiving co-located pediatric–adult care were not available, these comparisons were not possible. To determine whether selection bias might have affected our results, we compared clinical characteristics between participants and non-participants. We found no difference in the distribution of SCD genotypes, and those who opted for the co-location model were more likely to have been on hydroxyurea or chronic transfusion. In this cohort, initiation of hydroxyurea and chronic transfusions followed national recommendations for SCD complications, ie, monthly transfusions for stroke prevention and hydroxyurea for recurrent vaso-occlusive events.7 These findings indicate that participants were not healthier than the general SCD population and possibly had a more severe phenotype than those not participating in the transitional co-location model. In conclusion, young adults with SCD who utilized the co-location model with disease education for pediatric to adult healthcare transition did not demonstrate increases in acute healthcare utilization. Greater disease knowledge was associated with fewer hospitalizations and an increase in scheduled outpatient visits. The co-location model for pediatric to adult care transition seems to provide benefits among youth with SCD. Future studies validating our findings and potentially randomizing to different healthcare transition service models are warranted. The authors would like to thank Robin Womeodu, MD, Jeff Liebman, DDS, and Mark Yancy, MPH (Methodist Healthcare), for support of transition to adult care model implementation, Brian Winter, PhD and Kerri Nottage, MD, MPH (St. Jude) for support with data analysis and study concept. The authors are humbled by the courage of our patients and families in trusting their care while taking the risks of innovation. This work was supported in part by HRSA U1EMC19331, Baptist Foundation, and ALSAC. The authors have no conflicts of interest to declare. Dr. Nolan conceptualized the study design, conducted the data analyses, drafted the initial manuscript and approved the final manuscript as submitted. Ms. Anderson conceptualized the study design and participated in data collection, and approved the final manuscript as submitted. Dr. Smeltzer conceptualized the study design and aided in data analysis and critically reviewed the manuscript and approved the final manuscript as submitted. Ms. Porter, Ms. Carrol, Mr. Brooks and Drs. Elmagboul and Gurney conceptualized the study design and critically reviewed the manuscript and approved the final manuscript as submitted. Dr. Hankins conceptualized the study design, supervised data collection, provided interpretation of results, and drafted the initial manuscript and approved the final manuscript as submitted. Additional Supporting Information may be found online in the supporting information tab for this article. Supporting Information Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.