TRANSITION CARE CONTINUITY PROMOTES LONG-TERM RETENTION IN ADULT CARE AMONG YOUNG ADULTS WITH SICKLE CELL DISEASE

Abstract

Background/Objectives:

Care continuity prevents increased health care utilization and mortality during transition from pediatric to adult care. Our program employs a co-located care delivery model, in which pediatric provider involvement continues during young adulthood. We tested the hypothesis that individuals who participated in the co-located model have greater retention in adult care compared to those who only received pediatric transition services.

Methods:

This study consisted of 311 youth with SCD (51.4 % male; 63.0% HbSS/HbSβ⁰-thalassemia) who transferred to adult care from 2007–2017. Retention was defined as continuation with an adult provider for ≥12 or ≥24 months post-pediatric care. Logistic regression estimated the association between co-location status and retention at 12 and 24 months. Logistic regression and t-tests were used to evaluate potential predictors of retention in adult care.

Results:

Individuals who participated in the co-location model were 1.9 times more likely to remain in adult care 12 (95%CI: 1.01, 3.47) and 24 (95%CI: 1.01, 3.70) months post pediatric care compared to those who did not participate. Individuals with HbSS/HbSβ⁰-thalassemia were 1.9 times more likely to be retained at 12 months compared to those with HbSC/HbSβ⁺-thalassemia/HbS/HPFH (95%CI: 1.12, 3.09). For every clinic encounter in the last 2 years of pediatric care, the odds of being retained at least 24 months after initiating adult care increased 1.1 times (95%CI: 1.02, 1.13).

Conclusions:

Continuity of providers from pediatric to adult care may increase long-term retention in adult care. Longitudinal monitoring of adult outcomes is critical to identifying the efficacy of transition services.

Meeting Abstracts:

Transition Continuity Promotes Long-Term Retention in Adult Care Among Young Adults with Sickle Cell Disease. American Society of Hematology Annual Meeting. Poster presentation. November 13, 2019. https://ashpublications.org/blood/article/134/Supplement_1/4676/428681/Transition-Continuity-Promotes-Long-Term-Retention

INTRODUCTION

Sickle cell disease (SCD) is a group of inherited blood disorders which affects nearly 100,000 people in the United States.¹ Further, attributable to newborn screening and numerous advancements in pediatric treatment, >95% of American children with SCD are expected to survive into adulthood.^2–4 However, young adults with SCD experience increased acute healthcare utilization and early mortality during the transition period from pediatric to adult care.^5–8

Health care transition programs have become essential for young adults with chronic diseases. The goal of health care transition is to successfully prepare, transfer, and integrate patients with chronic disease into adult-oriented healthcare in a way that is developmentally appropriate and uninterrupted.⁹ However, there are several barriers to transition which include patient-perceived lack of knowledge by practitioners, resistance of pediatric teams to initiate transition, and challenges in care coordination.¹⁰ In 1996, the American Academy of Pediatrics recommended that pediatric and adult providers co-manage care for patients with chronic illnesses for up to two years to increase the patients’ comfort level during transition.¹¹

In light of this recommendation, our institution designed a transition program for individuals with SCD that starts at age 12 years, transfers to adult care at age 18 years, and uses a co-located care delivery model for adult care integration between ages 18 and 25 years.¹² In our co-located model, pediatric providers (physicians, neuropsychologists, and nurse educators) co-locate care with adult providers (physicians, social workers, advance practice practitioners), in the adult clinic. In prior work, we showed that individuals who participated in this co-located model of care did not experience the expected increased acute healthcare utilization after transfer to adult care.¹³

Previous studies have demonstrated that the course of disease is worsened if therapies such as hydroxyurea and transfusion are interrupted during health care transition.^10,14–16 This highlights the importance of continuity of care during adult care transition to preserve health. Unfortunately, there is a paucity of interventions shown to improve adult care retention in SCD. We tested the hypothesis that individuals with SCD who participated in the co-located transition health care delivery model had greater retention in adult care 12 or 24 months after engaging in adult care compared to those who only received pediatric transition services. Additionally, we explored the study cohort for predictors of retention in adult care including sex, SCD genotype, year of transfer, death status, economic health index, and pediatric outpatient history.

METHODS

Participant selection

Patients with SCD were included if they were followed by our pediatric program and completed transfer to adult care (i.e., fulfilled a first adult visit) during the years 2007–2017. Demographic information, including sex, SCD genotype, mortality status, and healthcare utilization from both pediatric and adult facilities were extracted from the participants’ electronic medical records. The participant’s last known address from their pediatric care was used to determine their economic hardship index. The economic hardship index (EHI) is a standardized score of six variables (unemployment, dependency, education, per capita income, crowded housing, and poverty) used to indicate economic conditions.¹⁷ It ranges from 1 to 100 with higher scores indicating worse economic hardship. Each participant’s score was standardized to the scores of the county (Shelby County, TN). Participants were retrospectively followed through December 31, 2017. All study activities were approved by St. Jude Children’s Research Hospital’s Institutional Review Board. The data that support the findings of this study are available from the corresponding author upon reasonable request.

All participants received transition services through the St. Jude Transition to Adulthood program during childhood¹², but were given the choice to transfer to the partner adult care facility where they received co-located care or transfer care to a different adult care facility. Those who transferred to the partner adult care facility were in the co-location model, while those who transferred to a different facility were in the non-co-location model. Adult care retention was defined as maintaining adult care, as indicated by attending the adult SCD facility at least once 12 or 24 months after the first adult care visit.

Statistical considerations

Chi-square tests were used to assess the association between demographics and co-location exposure status to inform the regression models. Multivariable logistic regression was used to assess the association between co-location exposure and adult care retention. Models were tested for effect modification and confounding by demographic variables such as sex, SCD genotype, year of transfer, and age at transfer. Forward selection was used for confounder assessment. Simple logistic regression was used to assess the association between demographic factors, such as sex, SCD genotype, year of transfer, economic hardship, death status, and retention status at 12 and 24 months after initiation of adult care. T-tests and simple logistic regression were used to assess how pediatric outpatient encounters might predict adult care retention status at 12 and 24 months after initiation of adult care. Pediatric outpatient encounters were analyzed two ways: first, the mean number of pediatric encounters per year from their complete pediatric history and, second, the mean number of pediatric encounters per year from their last two years of pediatric care.

As a sub-analysis, the sample was restricted to patients who transferred from pediatric to adult care in 2012 or later to account for the partner adult care facility opening. All analyses were repeated with the restricted sample. All statistical analyses were performed in SAS 9.4 with an alpha of 0.05 for significance.

RESULTS

Participants’ characteristics and co-location health care delivery model participation

Of the 469 youth who completed pediatric care at our institution between the years 2007 and 2017, 364 patients (53% male, 62% HbSS/ HbSβ⁰-thalassemia) completed transfer to adult care (Figure 1). However, participants were excluded from the 12-month analysis if they transferred in the year 2017 (n=53) and from the 24-month analysis if they transferred in the years 2016–2017 (n=91) because they did not have opportunity for the full 12 or 24 months of observation in adult care. After exclusion of those without opportunity for complete follow-up, there were 311 participants in the 12-month analysis and 273 participants in the 24-month analysis (Table 1). In both analyses, 52% were male, 63% had a severe SCD genotype (HbSS or HbSβ⁰-thalassemia), and 95% were living at the end of follow-up. Of the 14 participants who died during follow-up, 10 (71%) died due to SCD-related causes and 4 (29%) died due to non-SCD-related causes.

FIGURE 1.

Participant follow-up at 12 and 24 months

TABLE 1.

Participant demographics and co-location status

	12 months Analysis (n=311)			24 months Analysis (n=273)
	Co-Location N (%)	Non-Co-Location N (%)	p-value	Co-Location N (%)	Non-Co-Location N (%)	p-value
Sex			0.3807			0.5130
Male	80 (50)	80 (50)		67 (47)	75 (53)
Female	83 (55)	68 (45)		67 (51)	64 (49)
Genotype			0.3807			0.2169
HbSS, HbSβ0-thalassemia	99 (51)	97 (49)		79 (46)	92 (54)
HbSC, HbSβ+thalassemia, HbS/HPFH	64 (56)	51 (44)		55 (54)	47 (46)
Economic Hardship Index			0.7733			0.7592
Low (<36.6)	81 (53)	72 (47)		67 (50)	68 (50)
High (>36.6)	79 (51)	75 (49)		64 (48)	70 (52)
Missing	3 (75)	1 (25)
Year of Transfer			<0.0001			<0.0001
2007	2 (6)	29 (94)		2 (6)	29 (94)
2008	6 (15)	33 (85)		4 (11)	33 (89)
2009	0 (0)	14 (100)		0 (0)	14 (100)
2010	7 (19)	29 (81)		5 (15)	29 (85)
2011	9 (53)	8 (47)		9 (53)	8 (47)
2012	40 (91)	4 (9)		40 (91)	4 (9)
2013	20 (65)	11 (35)		20 (66)	10 (33)
2014	27 (87)	4 (13)		27 (87)	4 (13)
2015	32 (80)	8 (20)		27 (77)	8 (23)
2016	20 (71)	8 (29)		N/A	N/A
Death Status			0.6049			0.5919
Living	154 (52)	142 (48)		126 (49)	133 (51)
Deceased	9 (60)	6 (40)		8 (57)	6 (43)
	Mean (SD)	Mean (SD)	p-value	Mean (SD)	Mean (SD)	p-value
Age at Transfer	18.15 (0.7)	18.27 (0.5)	0.0722	18.17 (0.7)	18.27 (0.5)	0.1926

Sex, SCD genotype, and death status were not found to be significantly associated with participating in the transition co-located model (Table 1). Those who transferred to adult care between the years 2012 and 2016 were 18.7 times more likely to participate in the co-location model compared to those who transferred between 2007 and 2011 (95%CI: 10.51, 32.25).

Predictors of adult care retention

Participation in the co-location transition model resulted in 80% 12-month retention compared to 64% 12-month retention among those who did not participate in the co-location model (Figure 2). Youth who participated in the co-location model were 1.87 (95%CI: 1.01, 3.47) times more likely to remain in adult care 12 months post-pediatric care compared to those in the non-co-location model after adjusting for year of transfer. Additionally, participation in the co-location transition model resulted in 75% 24-month retention compared to 58% 24-month retention among those who did not participate in the co-location model. Those who participated in the co-location model were 1.94 (95%CI: 1.01, 3.70) times more likely to remain in adult care 24 months post pediatric care compared to those in the non-co-location model after adjusting for year of transfer. These analyses classified participants who had died in the first 12 or 24 months (n=6 and n=7, respectively) as patients who voluntarily abandoned care. Exclusion of the deceased participants from the analyses yielded similar results (Supplementary Table 1).

FIGURE 2.

Proportion of participants retained in adult care at 12 or 24 months

In the 12-month analysis, those with more severe genotypes (HbSS/HbSβ⁰-thalassemia) were 1.86 (95%CI: 1.12, 3.09) times more likely to be retained in adult care compared to those with a less severe genotype (HbSC/ HbSβ+-thalassemia/HbS/HPFH) (Table 2). Additionally, those who transferred to adult care during the years 2012–2016 were 1.69 (95%CI: 1.03, 2.79) times more likely to be retained in adult care compared to those who transferred during the years 2007 to 2011. In the 24-month analysis, those who transferred to adult care during the years 2012–2016 were 1.81 (95%CI: 1.05, 3.12) times more likely to be retained in adult care compared to those who transferred during the years 2007 to 2011.

TABLE 2.

Participant demographics and odds of adult care retention

	12-Month Retention N=311			24-Month Retention N=273
Variable	OR (95% CI)		p	OR (95% CI)		p
Gender
Male	1.02 (0.62, 1.67)		0.9506	0.96 (0.56, 1.64)		0.8897
Female	Ref			Ref
Genotype
HbSS, HbSβ0-thalassemia	1.86 (1.12, 3.09)		0.0164	1.64 (0.95, 2.82)		0.0750
HbSC, HbSβ+thalassemia, S/HPFH	Ref			Ref
Year of Transfer
2007–2011	Ref			Ref
2012–2016	1.69 (1.03, 2.79)		0.0391	1.81 (1.05, 3.12)		0.0314
Economic Hardship Index
Low	1.29 (0.78, 2.13)		0.3231	1.12 (0.67, 1.86)		0.6661
High	Ref			Ref
Death Status
Living	1.80 (0.62, 5.22)		0.2788	1.53 (0.50, 4.72)		0.4597
Deceased	Ref			Ref
Pediatric Outpatient Visitsc
Complete pediatric history	1.05 (0.99, 1.12)		0.1195	1.06 (1.00, 1.13)		0.0591
Last 2 years of pediatric care	1.06 (1.01, 1.11)		0.0303	1.08 (1.02, 1.13)		0.0088
Pediatric Outpatient Visitsc,d	Retained Mean (SD)	Not Retained Mean (SD)	p	Retained Mean (SD)	Not Retained Mean (SD)	p
Complete pediatric history	6.25 (4.3)	5.36 (4.9)	0.1181	6.27 (4.4)	5.21 (4.3)	0.0572
Last 2 years of pediatric care	7.73 (6.6)	5.92 (6.3)	0.0308	7.84 (7.1)	5.68 (5.0)	0.0037

^cDefined as mean outpatient encounters per year of pediatric care,

^dcomparison made using T-test.

The median EHI value was 36.6 (IQR: 31.2, 43.5) and participants were dichotomized into low EHI (≤36.6) and high EHI (>36.6). There was no association between EHI and adult care retention at 12 or 24 months (Table 2).

Those who were retained at 12 and 24 months attended, on average, nearly one more outpatient visit per year throughout their complete pediatric history. However, we found that those who were retained at both 12 and 24 months attended, on average, nearly two more outpatient visits per year in the last two years of pediatric care compared to those who abandoned care prior to 24 months after adult care initiation (Table 3). For each additional outpatient visit attended in the last two years of pediatric care, the odds of being retained at 12 months increased by 1.06 (95%CI: 1.01, 1.11) and the odds of being retained at 24 months increased by 1.08 (95%CI: 1.02, 1.13).

To account for the opening of the adult clinic, we restricted the analysis to patients who transferred out of pediatric care in 2012 or later (Supplementary Table 2). Patients who participated in the co-location model had 2.45 (95%CI 1.10, 5.48) times the odds of retention at 12 months and 4.19 (95%CI 1.73, 10.17) times the odds of retention at 24 months (Supplementary Table 3). Additionally, patients with a more severe sickle cell genotype (HbSS/HbSβ⁰-thalassemia) were 3.30 and 3.64 times more likely to be retained in adult care for 12 months (95%CI: 1.58, 6.87) or 24 months (95%CI: 1.55, 8.55) compared to those with a less severe genotype (HbSC/ HbSβ+-thalassemia/HbS/HPFH). For each additional outpatient visit attended in the last two years of pediatric care, the odds of being retained at 12 months increased by 1.14 (95%CI: 1.03, 1.26) and the odds of being retained at 24 months increased by 1.17 (95%CI: 1.06, 1.28).

DISCUSSION

The co-located model of transition bridges the gap between pediatric and adult care by creating a partnership between pediatric and adult providers during the transitional period and the present study demonstrated that it promotes greater retention 12 and 24 months after initiating adult care compared to those who only received pediatric transition services. Among individuals with chronic illnesses undergoing health care transition, retention in adult care is critical to maintaining adherence to medications and therapies that preserve health quality.^18,19 These results are comparable to those found in a similarly designed care delivery model. Among individuals with Type 1 Diabetes, a care delivery model utilized a transition coordinator throughout the transition process and pediatric and adult providers were present at the last visit in pediatric care and the first visit in adult care.¹⁸ It was found that those who received this structured model of care had a shorter transfer duration, improved glycated hemoglobin values, and had better clinic attendance compared to those who received the unstructured model of care.¹⁸

Continuity of care in SCD has been previously shown to decrease acute healthcare utilization and acute complications such as stroke and acute chest syndrome because the patients are able to maintain treatment therapies such as hydroxyurea or transfusions^10,14–16. Not only does retention in adult care allow for opportunities to receive regular care, but it also allows access to newly developed treatments, including curative therapies for SCD, if care is received at specialized centers.

The identification of predictors of retention in adult care during health care transition is important because these predictors identify gaps in care for future interventions. Once known, predictors of adult care retention can be used in the development of risk stratification for transition (i.e., risk assessment for successful transition), which would allow providers to identify who is at risk of abandoning care in adulthood and implement interventions to prevent care abandonment in pediatrics or during the transfer period. The present study found that increased compliance with outpatient encounters in the last two years of pediatric care increases retention in adult care 12 and 24 months after adult care initiation. At this young adult stage (ages 16 to 18 years), patients are starting to take control of their own health and beginning to fully understand how to manage their disease, therefore, creating an opportunity for interventions that remove barriers to pediatric outpatient care attendance. Further, the present study did not find economic hardship index to be a predictor of retention in adult care. SCD notably affects a predominantly poverty-stricken population, and, when standardized to our county, there was not enough variability in the sample to demonstrate an association.

This model of in-person, co-located care might not be attainable in other facilities due to lack of resources; however, technology such as telemedicine has been previously used to deliver care and might offer an alternative model of co-located care. Telemedicine has been shown to be effective in increasing encounters while utilizing very few resources.²⁰ This design would allow the pediatric providers to be involved during the transitional stage of adult care without their physical presence in the adult care facilities.

A limitation of this study is that it is retrospective in nature and the co-location participation was not randomized. Despite efforts to locate each participant, there were 105 adolescents eligible for transfer but did not have record of an initial adult care visit. Some may have established care outside the network, which removed them from this study due to inaccessibility to data. There were no significant demographic differences between the ineligible and eligible adolescents. This study found that participants who transferred to adult care between 2012–2016 were more likely to participate in the co-location model and be retained for 12 months compared to those who transferred between 2007–2011. The adult clinic evaluated in this study was established in 2012 and, thus, participation in the co-location model increased significantly at this time. Participants who transferred before 2012 were included in this study because they were given the opportunity to receive care at the onset of the clinic. Models were adjusted for the year of transfer to improve the comparability between those who participated and those who did not participate, and a sub-analysis was conducted to evaluate the participants who transferred after the adult clinic was opened in 2012. After the clinic was created in 2012, patients had greater opportunity to participate in the complete co-location model and the results of the sub-analysis show a stronger association with retention in adult care. However, unmeasured confounding might still be present as we did not have information on additional characteristics that might affect participation in the co-location model, such as parental education, patient neurocognition, severity of disease beyond genotype, patient’s perceived severity, parent’s perceived severity, or parent’s involvement in care. Additionally, we were unable to assess how retention affects acute health care utilization, such as emergency department encounters or hospitalization, because we did not have access to these data. The strengths of this study include the inclusion of 78% of the SCD population of our program who completed pediatric care during the study period, reducing the opportunity for misclassification of co-location model status, as well as a cohort that spans from pediatric to young adulthood years.

Participation in the co-located care delivery model of transition is associated with increased odds of retention in adult care one and two years after pediatric care transfer by approximately two-fold. Additionally, increased compliance with pediatric outpatient encounters is associated with improved retention in adult care. Retention is important for maintaining disease-modifying therapies and other preventive measures, therefore, maintaining and potentially improving patient health. Future studies should investigate longer follow-up as two years might not be enough time to distinguish retention. Additionally, future studies should utilize a randomized, prospective design to further determine the efficacy of the co-located transition model.

Supplementary Material

tS2

SUPPLEMENTARY TABLE 2. Participant demographics and co-location status for patients who transferred from pediatric care between 2012–2016

tS1

SUPPLEMENTARY TABLE 1. Odds of retention excluding participants who died during the first 12 or 24 months of follow-up

tS3

SUPPLEMENTARY TABLE 3. Participant demographics and odds of adult care retention for patients who transferred from pediatric care between 2012–2016

fS1

SUPPLEMENTARY FIGURE 1. Proportion of participants retained in adult care at 12 or 24 months for patients who transferred from pediatric care between 2012–2016

Abbreviations:

SCD

Sickle cell disease

TN

Tennessee

EHI

Economic Hardship Index