When children with sickle‐cell disease become adults: Lack of outpatient care leads to increased use of the emergency department

Abstract

Young adults with sickle-cell disease have increased emergency department (ED) utilization and increased risk of mortality for unclear reasons. Emergency Department Reliance (EDR) differentiates excessive ED use due to increased need in chronic illness from increased use due to limited access to outpatient care. A higher EDR has been used to define excessive reliance on the ED and thus access to care issues leading to increased ED utilization. We conducted a retrospective cohort study of sickle-cell disease patients within the Wisconsin Medicaid database over a 5-year period to examine EDR during the transition period from childhood to adulthood. The study population included four distinct groups: (1) children, (2) patients transitioning from pediatric to adult providers, (3) young adults, and (4) adults age 31–45. Rates of visits per year were calculated for ED visits and outpatient visits for all diagnoses and sickle-cell disease-related diagnoses. Overall, we found increased EDR among the transition group and young adults compared to children and adults for sickle-cell disease-related diagnoses. These findings suggest access to care issues play a significant role in the increased ED utilization seen during the transition period from pediatric to adult providers in sickle-cell disease. Sickle-cell disease is the most common inherited blood disorder in the United States, affecting ∼90,000 Americans [1, 2]. Complications of sickle-cell disease begin in early childhood and historically have contributed to a significantly shortened life expectancy [3-5]. With advances in clinical care, however, the life expectancy of children with sickle-cell disease has increased significantly, and, now, the majority of individuals with sickle-cell disease live well into adulthood [6]. With improved survival, the need for transition from pediatric centered care to adult-centered care for patients with sickle cell disease has garnered increased attention in the recent years. Much of the literature has focused on patient and provider perceptions of this transition, determination of patient readiness, or components of a good transition program [7-11]. Recent studies have demonstrated that young adults 18–30 years have increased emergency department (ED) utilization [12]. In addition, young adults have a higher risk of mortality during this transition period [6, 13]. The reasons for increased utilization and mortality are not entirely clear. The transition of adolescents with chronic illnesses to adult providers is associated with many challenges, including loss of a medical home, decreased access to ambulatory care providers, and loss of insurance coverage [14]. Adolescents with sickle-cell disease face similar challenges as they transition from pediatric care, and such issues may impact their clinical course. Alternatively, a worsening of disease pathology may itself lead to more severe complications in this age group. The relative role of these factors in relation to utilization patterns in patients with sickle-cell disease is not known. Emergency Department Reliance (EDR) has been used to differentiate increased ED use due to need for care from increased ED use secondary to access issues [15]. EDR is defined as the number of ED visits divided by the number of ED and outpatient visits and therefore views ED visits in relation to all ambulatory visits. Patients with more severe disease, who utilize the ED more, should also have more outpatient visits, while those without adequate outpatient clinic access simply use the ED more without a rise in outpatient clinic visits, thus increasing EDR. The objectives of our study were to examine health care utilization patterns and reliance on ED care among patients with sickle-cell disease transitioning from pediatric to adult care. We hypothesized that patients transitioning from pediatric to adult centered care would have increased reliance on the ED for acute care as reflected by an increased EDR during this time period. Our final study population included 687 patients: 345 children, 65 patients in the transition group, 139 young adults, and 138 adults age 31–45. The study population had a mean of 47 months of continuous Medicare enrollment. Ambulatory visits (32,258) have occurred over the 5-year study period including 20,418 outpatient visits and 11,840 ED visits. The most frequent diagnosis for all groups was sickle-cell disease unspecified (for outpatient visits) and sickle-cell disease with crisis (for ED visits) in both the sickle-cell-related and all diagnoses categories. Significant differences were found among the study groups for outpatient and emergency department visit rates for all diagnostic categories (Table 1). EDR was highest in the transition and young adult groups across all diagnoses categories. Differences in EDR were significant between the four age groups for any diagnoses (P < 0.001) and for SCD-related diagnoses (P < 0.001) (Fig. 1). Upon pairwise comparisons, for sickle-cell disease-related diagnoses the transition group had a significantly higher EDR compared to children [mean (SE) of 0.46 (0.03) vs. 0.38 (0.01), P = 0.013] as well as older adults [mean (SE) of 0.46 (0.03) vs. 0.33 (0.02), P < 0.001]. Similarly, for sickle-cell-related diagnoses, the young adult group had a significantly higher EDR than both children [mean (SE) of 0.47 (0.03) vs. 0.38 (0.01), P = 0.002] and adults [mean (SE) of 0.47 (0.03) vs. 0.33 (0.02), P < 0.001]. A similar pattern emerged when computing EDR based on all diagnoses, with the EDR for the transition group being significantly higher than for children [mean (SE) of 0.39 (0.03) vs. 0.29 (0.01), P < 0.001] as well as older adults [mean (SE) of 0.39 (0.03) vs. 0.30 (0.02), P = 0.003]. Again, the young adult group had similar results for all diagnoses with a significantly higher EDR than both children [mean (SE) of 0.41 (0.02) vs. 0.29 (0.01), P < 0.001] and adults [mean (SE) of 0.41 (0.02) vs. 0.30 (0.02), P < 0.001]. There were no significant differences in EDR between the transition group and young adults for either diagnostic category. Mean Emergency Department Reliance for children, the transition group, young adults, and adults for sickle-cell disease related and all diagnoses. Our data show a higher utilization of the ED in patients with sickle-cell disease who are transitioning from pediatric to adult care when compared with children. The EDR was highest for the transition group and young adults across all diagnoses categories, suggesting these patients depend on the ED for care. The high-ED utilization rates are similar to findings of others [12, 13]. However, our study uniquely includes a transition group of patients who made the transition from pediatric to adult providers during the study period. Inclusion of this group allows a direct comparison of the transition group of patients to their older and younger counterparts. Use of the EDR allows rates of ED utilization to be viewed in relation to all ambulatory care encounters. By doing so, the EDR provides additional information on the underlying reasons behind these patterns. The increased EDR for the transition group across all diagnoses categories suggests that the increase in ED utilization is at least in part due to limitations in access to primary care providers. The return of the EDR for sickle-cell disease-related diagnoses in the older adult group to those levels seen in childhood suggests a stabilization of a medical home following the transition period. The increased ED utilization might also be explained by a worsening of disease. However, as disease worsens and the need for more frequent ED visits develops, one would also expect these patients to be followed in clinic on a more regular basis to optimize their medical care through interventions such as hydroxyurea. Increased ambulatory visits would then serve to maintain a lower EDR for this group of patients. The finding of an elevated EDR in the transition and young adult groups for all diagnostic categories therefore suggests that even if their underlying sickle-cell disease is indeed worsening, patients transitioning to adult providers are unable to access adequate ambulatory care outside of the emergency room setting. The return of the EDR in the adult group to those levels seen in childhood suggests a stabilization of a medical home as patients leave the transition period. Although our data suggest access to care or changes in the medical home contribute to increased utilization for patients with sickle-cell disease transitioning to adult care, the mere presence of a comprehensive clinic or other form of medical home for patients during the transition period may itself not be the only answer to these issues. Even if such resources exist, additional challenges such as school, work, and family obligations may make access to care issues not simply about whether a clinic or provider is available. One study examining the correlation between the proximity of a comprehensive sickle-cell disease center to patient's homes found no improvement in access to care based on location alone [16], highlighting the importance of additional barriers to access to care. Further research is needed to identify such barriers and appropriate interventions to improve access to care. Access to care issues and reliance on the ED shown in our study is yet another challenge facing adolescents with sickle-cell disease. This challenge adds to the documented increased risk of mortality during this time period described by others [4]. Whether improving the transition of patients with sickle-cell disease from pediatric to adult providers will decrease ED utilization or lower the risk of mortality in this vulnerable period is an unanswered but critical question in ensuring the best ongoing care for these patients. This study has additional limitations that must be addressed. First, the database is limited to a single state and availability of expert sickle-cell-related care will vary by state and institution. Second, the database includes only individuals cared for within the Wisconsin State Medicaid system, excluding those with private insurance. However, nationally, 67% of patients with sickle-cell disease have public insurance meaning that our database includes the majority of sickle cell disease patients in Wisconsin [17]. Additionally, our analysis relied on ICD-9-CM coding and any coding errors would affect our analysis. However, the method of selecting sickle-cell patients using previously established criteria for large database studies limits the inclusion of many such errors. Last, numerous other factors associated with the transition period such as decreased parental involvement, insurance coverage changes, and evolving work, social, and family commitments may contribute to the decreased use of outpatient resources and therefore affect our analysis. In conclusion, these findings support our hypothesis that for sickle-cell disease patients, the transition of care from pediatric to adult providers is associated with increased reliance on the ED for acute care as reflected by an increased EDR during this time period. These findings that suggest access to care issues play a significant role in the increased ED utilization seen during the transition period from pediatric to adult providers. Further research should focus on the access to care issues of these patients as well as how such access issues affect quality of care and outcomes in sickle-cell disease for patients transitioning to adult providers. We conducted a retrospective cohort study utilizing Wisconsin State Medicaid data from January, 2003 to December, 2007. This database includes 825 patients of any age with sickle-cell disease who are part of the Wisconsin Medicaid system. Consistent with previous literature, having sickle-cell disease was defined by having at least one inpatient hospitalization, or two outpatient visits at least one month apart, with a diagnosis of sickle-cell disease [13, 18]. Although this definition may exclude some patients with sickle-cell disease who are infrequent users of medical care, it also protects against including misdiagnosed or miscoded patients (such as patients with sickle cell trait diagnosis) who do not have sickle-cell disease. All encounters during the study period were extracted from the database and linked by unique anonymous identifiers. Information extracted included age, all diagnosis codes, and classification as ED visit, outpatient visit, or inpatient hospitalization. Provider types that were not representative of acute medical care, such as dentists, physical therapists, and home health agencies, were excluded (see Supporting Information Appendix Table AI for complete details on encounter and provider types). To protect against multiple charges originating from a single patient encounter, only a single visit type (ED, inpatient, or outpatient) on any given day was included in the final analysis. Time of enrollment was defined as the number of months between the first and last encounters in the database. Individuals with a gap in enrollment of 3 months or greater were excluded from the analysis as these individuals likely received significant amounts of care outside the Medicaid system. Similarly, individuals with only a single encounter over the 5-year period were also excluded as their length of enrollment in the database could not be calculated. The study population was divided into three mutually exclusive age groups: (1) children: age ≤ 18 years old for the entire study period, (2) transition group: individuals who turned 19 years old during the 5-year study period; and (3) young adults: age 19–45 years at first encounter. The age of 19 was selected for the transition period as our institutional practice involves transitioning patients to adult providers when they turn 19 years old. Adults age ≥ 46 at first encounter were eliminated from the analysis as the goal was to compare the transition group to the age groups immediately above and below. Patients with only a single encounter in the database, those with a gap of 3 months or greater in enrollment, and those patients over the age of 45 were excluded from the analyses. Encounter data were examined for all diagnoses and for all sickle-cell disease-related diagnoses. The sickle-cell disease-related diagnoses category included all visits with an ICD-9-CM code for sickle-cell disease (28,241; 28,260; 28,261; 28,263; 28,268) or sickle cell with crisis (28,242; 28,262; 28,264; 28,269), as well as ICD-9-CM codes for pain likely related to sickle-cell disease regardless of additional coding (see Supporting Information Appendix Table AII for complete list of ICD-9CM codes included in the analysis). The primary outcomes measured were rates of visits per year and EDR. Rates were calculated for each age group for ED visits and outpatient encounters. Two diagnostic categories based on ICD-9-CM coding were examined for these visits: sickle-cell disease-related diagnoses and all diagnoses. The mean EDR (ED visits/[ED + outpatient visits]) was computed for each age group for both diagnostic categories. A higher EDR value is indicative of higher reliance on the ED for acute care issues, with no increased proportional rise in utilization of a primary care provider in an outpatient setting. Prior studies of ED utilization in infant and general pediatric populations have defined an excessive EDR as >0.33 [15, 19, 20]. For each patient in the database, the rate of visits per year was calculated using the number of visits divided by the number of enrollment years. EDR was calculated for each patient as described earlier. Mean visit rates and mean EDR were compared after log transformation among the four age groups using the analysis of variance and pairwise comparisons were performed between each group using the t-test. The study and a waiver of patient consent were approved by the Medical College of Wisconsin institutional review board. Additional Supporting Information may be found in the online version of this article. 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. Bradd G. Hemker* , David C. Brousseau* , Ke Yan* , Raymond G. Hoffmann* , Julie A. Panepinto* , * Department of Pediatrics, Section of Pediatric Hematology/Oncology/Bone Marrow Transplant, Medical College of Wisconsin, Milwaukee, Wisconsin, Department of Pediatrics, Children's Research Institute, Medical College of Wisconsin, Milwaukee, Wisconsin, Department of Pediatrics, Section of Pediatric Hematology/Oncology/Bone Marrow Transplant, Children's Research Institute of the Children's Hospital of Wisconsin, Medical College of Wisconsin Milwaukee, Wisconsin.