1 INTRODUCTION

Sickle cell disease (SCD), a set of genetic erythrocyte disorders affecting persons of African, Mediterranean, or Asian descent, was first described in 1910.¹ As a result of dramatic improvements in care,^{2, 3} the vast majority of patients survive to adulthood so that there are currently more adults than children living with SCD in the United States.⁴ Despite this, adult SCD care in the United States is sorely lacking. Studies of access to care and utilization for adults with SCD compared with children using administrative databases find that transition-age patients and young adults have both higher emergency department (ED) utilization rates and unstable access to primary and subspecialty ambulatory care.⁵ A lack of adult SCD providers nationally has been blamed. Disparities in the availability of SCD care have been decried by national authorities^{6, 7} and have been blamed for the increased morbidity and mortality during the transition-of-care years, when adolescents and young adults (AYA) leave pediatric care and seek adult care. Further, morbidity and mortality continue at higher rates during adulthood⁸ again presumably due to lack of SCD specialty providers.

The quality of care (QoC) and the success of transition from pediatric to adult care for patients with SCD also lags, compared with the QoC for well children or those with other special health care needs.^9-11 QoC indicators for SCD have only recently been proposed, published¹² and/or measured, either for children^{13, 14} or for young and older adults.^15-19 The first paper that outlined the necessary components of dedicated adult sickle cell treatment centers was only recently published⁶ and descriptions of patient reported experiences regarding QoC have only recently been published.²⁰

We therefore were interested in comparing how adolescent and adult patients already in established SCD centers perceived access and QoC. We fielded a survey to test whether access to SCD care and satisfaction with SCD care were comparable when both pediatric and adult SCD specialty providers were available. We were interested in perceptions regarding timeliness of care, interpersonal aspects of QoC, and ratings of pain management, and whether they differed among adult versus pediatric centers. Further, we were interested not only in whether differences in these perceptions were reported, but also why they might be reported. We hoped to determine whether underlying structural differences in care were driving any reported differences in access and satisfaction by clinic type, versus differences in biopsychosocial determinants of perceived need for care. We were aware that disease severity and psychosocial need, both of which could drive patient perceptions, tend to increase in older patients and in those who require contact with the medical system more often, those more likely to be seen in adult clinics.

2 METHOD

3 RESULTS

A total of 224 patients aged 15 years and older were enrolled in the SHIP-HU study, but six were inappropriately randomized and were excluded. Another 34 provided no ASCQ-Me QOC data at baseline (n = 25) or were missing MHC (n = 9), leaving n = 184 for this analysis—54 treated at pediatric clinical sites and 130 treated at adult clinical sites. Generally, patients 18 and younger were cared for at pediatric sites, and patients older than 21 were seen in adult sites. However, young adults between the ages of 18 and 22 years were evenly split between pediatric and adult sites. Table 1 shows there were no statistical differences between pediatric versus adult site patients with regards to gender, genotype, HU use, social support, or number of ambulatory care visits. However, there was a statistically significant difference between pediatric versus adult site patients in SCD comorbidities and complications: patients treated in adult sites had a higher mean MHC score of 2.2 versus a mean score of 1.1 among those treated in pediatric sites (p < .0001) and more ED visits (for example, 23.4% at adult clinics vs. 41.5% at pediatrics had no ED visits, and 28.9 vs. 9.4% had four or more ED visits). In addition, patients treated in adult sites had a lower (worse) mean emotional impact scale score of 51.6 versus a mean score of 56.2 among those treated in pediatric sites (p = .0006).

Regarding overall satisfaction with care, shown in Figure 1, on a scale of 0–10, satisfaction was significantly higher for patients treated in pediatric versus adult sites (mean 8.75 vs. 7.46, p < .0001). While 41.5% pediatric site patients reported their overall satisfaction with care was 10, only 17.8% of adult site patients scored overall satisfaction of 10. No pediatric site patients reported any value lower than 5, while 7.4% of adult site patients reported values of 0–4. When controlling for social support, ASCQ-Me MHC and emotional impact, and number of primary care and ED visits, the mean overall satisfaction values were similar to unadjusted values (8.51 vs. 7.55, p = .0059).

Table 2 outlines differences between pediatric site patients versus adult site patients with respect to satisfaction with ED care. On a 1–4 Likert scale, patients treated at pediatric sites reported statistically higher means on overall satisfaction with care in the ED (3.4 vs. 2.88, p = .0068), items querying whether their ED doctors or nurses cared about them (3.33 vs. 2.74, p = .0038; 3.33 vs. 2.77, p = .0053 respectively), whether their ED staff believed their pain (4.32 vs. 3.63, p = .0038) and whether their ED staff was able to help with their pain (4.26 vs. 3.67, p = .0131). The same relationships held when controlling for social support, emotional impact, MHC and number of ambulatory visits (3.35 vs. 2.74, p = .0047; 3.29 vs. 2.79, p = .0175, 4.38 vs. 3.61, p = .0015; 4.29 vs. 3.66, p = .0064, respectively) (Table S1).

The majority of patients treated at both adult (68%) and pediatric (85.7%) sites reported that they were always satisfied with ambulatory sickle cell care (Figure S1). Table 3 outlines satisfaction with ambulatory care. Patients generally felt that their medical providers listened carefully, understood how sickle cell impacted the patient personally, and treated them with courtesy and respect. There were no statistically significant differences in any of these measures based on treatment site. However, patients cared for at pediatric sites reported significant higher overall satisfaction with scheduled appointments (3.70 vs. 3.43, p = .0399) as well as time spent with patient (3.80 vs. 3.56, p = .0281). These latter two significant relationships though were no longer significant after controlling for social support, emotional impact, MHC, and number of visits (3.58 vs. 3.47, p = .4346 and 3.77 vs. 3.58, p = .0933, respectively) (Table S2).

Access to care in both the ED and ambulatory settings was better for patients in pediatric sites versus adult sites. Pediatric site patients were more likely to be able to get timely ambulatory appointments when necessary (p = .004) (Figure 2). There was no difference between sites with regards to provider continuity at each visit, with approximately 95% of all patients reporting that they would see their usual provider when making ambulatory appointments regardless of clinical site. Access to ED care was measured by reported wait times (Figure 3). Fully 47.7% of patients in adult sites reported their longest wait times of >1 h, whereas only 19.3% of patients at pediatric sites reported wait times of >1 h (p = .0048). Of note, while 66.4% of adult site patients and 40.7% of pediatric site patients reported ever delaying or avoiding ED care, only 1.5 and 1.6% respectively of those reported delayed care due to concerns related to insurance, suggesting that insurance was not a significant barrier to care in this population.

To explore differences in site controlled for age, we analyzed the population aged 18–22 years as a separate group. These patients were evenly divided between pediatric and adult care sites (Table S3). Differences between patients seen at pediatric versus adult sites were more similar for MHC, social support, emotional impact, and utilization than for the full sample. We found no obvious differences between either access to care or satisfaction with care based on treatment site, with the data showing no numerical trend toward improved scores for patients seen at pediatric sites (Tables 4 and 5). Generally, differences between sites were diminished. Because the sample size was small, which would imply lack of power to detect differences, statistical inference was not performed (thus no p values presented).

4 DISCUSSION

We found, among patients with established care at pediatric or adult SCD clinics, differences in reported access to care and satisfaction with care, using a validated QOC instrument. Patients treated in pediatric clinics reported higher overall satisfaction with care as well as higher satisfaction with ambulatory and ED care. Similarly, patients treated in pediatric clinics reported better access to care, both ED and ambulatory care, than patients treated in adult clinics. The most striking differences seen between pediatric versus adult clinics were in access to ED care and satisfaction with ED care.

We controlled our results for the relatively worse reported comorbidity and complexity of SCD patients cared for in adult clinics based on the SCD-MHC score, for social support, and for emotional impact of SCD. ED utilization is known to be higher for patients treated at adult clinics⁵ and was reported higher for patients seen in adult clinics in our sample. This higher utilization, plus worse comorbidity in patients seen in adult sites in our sample, raised the hypothesis that SCD satisfaction with care and access to care in our sample declined simply as a function of perceived need, which was higher among older, more complicated patients. Controlling for MHC, social support, and emotional impact did not confirm this hypothesis, suggesting that the decreased access and satisfaction were instead related to differences in the care delivery systems themselves.

Findings of differences in SCD pediatric and adult access and quality usually blame two major barriers: lack of adult sickle cell providers and lack of appropriate insurance coverage. We acknowledge these barriers are important and impactful and must be addressed. However, in this study, few patients cited lack of insurance as a barrier to ED care.

Further, although increased age was likely associated with decreased access and QoC in both populations, the sample we studied had adequate structural access. All had an established relationship with an adult SCD expert who was part of a larger health care system. Limiting our investigation to this study sample forces us to explain our findings by postulating factors other than access to care. We conclude our findings are best explained by differences in the quality of systems of adult versus pediatric SCD care at the studied sites. Differences in comprehensive care, rather than differences in the outpatient hematology clinics or the SCD providers where patients received their specialty care, were reported by study patients.

Our results are consistent with those of the Sickle Cell Disease Implementation Consortium who administered the ASCQ-Me QOC to 440 adolescent and adult patients across 7 sites. Their population also scored low with regards to access to emergency care and satisfaction with ED visits and staff. They did not analyze patients by treatment site, but rather by age, and found that satisfaction was significantly higher for patients less than 19 years of age when compared to those older than 19 years of age.²⁰

Our results force us to consider differences in the structure of pediatric versus adult SCD care. In all of the study's participating centers, pediatric sickle cell outpatient and acute care appeared to be more fully integrated such that pediatric SCD patients had a single care team during regardless of the clinical location. Pediatric ED care generally involved the pediatric SCD team early, seeking input and guidance from the providers who already had an established relationship with the patient. If patients were admitted to the hospital, they were cared for by a sickle cell specialist from their known practice group. This was not the case for participating adult sickle cell centers, where acute care in the ED or inpatient environment was largely carried out by providers without any specific sickle cell expertise and with minimal involvement of the SCD care team.

Our results have implications for adult SCD care. Patients with SCD are living longer. Yet there remains a lack of understanding about, comfort with, and acceptance of SCD adults by nonhematologist adult medicine specialists, including ED physicians and primary care physicians.²⁶ Holistic integration of SCD adult care across all facets of the medical center may improve adult SCD care quality. Systems will need to focus on education and training of adult providers²⁷ and must develop coordination-of-care teams that include adult SCD specialists. Additionally, systems need to implement evidence-based practice models to better serve this population. For example, EDs appear inferior to SCD-specific infusion centers or day hospitals to administer care for acute SCD pain.²⁸ Such centers use individualized pain plans, standardized treatment pathways, and rapid analgesic titration algorithms for patients with SCD, and vastly improve the likelihood of timely analgesia and lower the likelihood of admission to the hospital. Additionally, they are staffed by medical providers with training and expertise in SCD. Even when similar evidence-based strategies to improve care are implemented in the ED, care in infusion centers appears to be superior.²⁹ All of the pediatric patients in the SHIP-HU study had access to infusion centers as part of their sickle cell care. Only two of the clinical sites serving adults had an infusion center during the study period. While the tools used to assess satisfaction with care specifically asked about experience in the ED, this disparity may have contributed to our findings. This is one example of the differences in the pediatric and adult systems of care in this particular study.

Our study has some limitations. Chief among them, the SHIP-HU study used ASCQ-Me for the evaluation of adolescent patients ages 15–18. The ASCQ-Me QOC questionnaire was not validated in this age group. It has been used by other groups to assess QoC.²⁰ However, it cannot be assumed that all ASCQ-Me measurements are appropriate for use in adolescent patients. Future work could validate the ASCQ-Me tools in the adolescent population. This could be particularly helpful in defining successful health care transitions in this high-risk population. A second limitation of our study is that it did not include patients seen primarily outside of SCD specialty centers. Extrapolation of these results to patients seen in more remote settings requires further investigation.

Finally, in an attempt to control for age, we analyzed a small group of AYA patients split between pediatric and adult centers finding no numeric difference between accesses to care or satisfaction to care based on treatment site. However statistical power was insufficient to determine whether site differences might exist for this group. We also recognize that the transition age group may not be fully representative, given their unique position. Although many of these patients have established care with an adult sickle cell provider, they are just starting to learn the complex and very different adult system of care. As a result, we cannot fully differentiate between the effect of age and the effect of treatment site on our results for this population.

We conclude that, while future work is necessary to define quality metrics and program structure for adult care, recent calls for improved access to SCD care and improved SCD care quality are valid and apply to all adult providers.⁷ It also highlights a potential fallacy of transition-of-care programs in SCD—the idea that the main goal of the pediatric SCD provider is to educate adolescents and then find a willing adult SCD provider. Simply moving a patient to an adult system of care does not guarantee the quality of that care. Instead, our findings support calls for improved QoC of patients seen in adult SCD clinics and suggest there is much systematic work to be done.⁶ Perhaps such changes would encourage more providers to create centers of excellence for adults with SCD, where comprehensive support systems are in place to ensure excellent quality care.

CONFLICT OF INTEREST

The authors have no conflicts of interest to disclose