Abstract
Although SCD-related mortality has improved over the last 3 decades, mortality from acute pulmonary causes remains relatively unchanged.(Payne, 2017)Acute chest syndrome (ACS) occurs in patients with SCD of all ages and influenza infection is one of its cause. Although influenza vaccination is recommend by the CDC for patients with SCD, and was proposed as a quality of care indicator(Wang CJ, 2010), its use is not mentioned in 2014 SCD NIH guidelines. This study examined influenza vaccination in patients with SCD as a potential marker of quality of care delivery. The study population included black individuals aged 1 to < 60 years, continuously enrolled on Medicaid from January 1 2008 until July 31 2009 in Georgia (CMS Data Use Agreement: 25585) to identify disparities in influenza vaccination uptake among Medicaid enrollees with a diagnosis of SCD. Patients with one inpatient claim or two outpatient claims with an ICD-9 code for SCD were considered to have a diagnosis of SCD. The association between influenza vaccination and health care utilization was measured by logistic regression models. Outcomes included: outpatient visits (<median outpatient visits versus >=median outpatient visits), emergency department (ED) visit or any hospitalization and influenza specific hospitalization during the study period. The fully adjusted models included the following covariates: age, gender, rural status, residing in two index counties (Fulton, Rockdale; a state sponsored SCD center is located in Fulton county), eye examination (as per SCD NIH guidelines), Elixhauser and a SCD comorbidity index. For the SCD comorbidity index, a score of one was given for each of the following 5 groups of sickle cell disease complications: CNS complications (>=1 diagnosis of stroke, TIA or epilepsy/recurrent seizures); cardio pulmonary complications (>=1 diagnosis of chronic pulmonary heart disease, pulmonary hypertension, malaise and fatigue, edema, chest pain or hypoxemia); kidney disease (>=1 diagnosis of chronic kidney disease/renal failure, proteinuria), avascular necrosis or ulcer of lower limbs (adapted from Afenyi-Annan, 2008, Candrilli. 2011, Elmariah. 2014). A total of 1544 patients with SCD were included in the study. Having influenza vaccination (aOR=1.77, p-value=0.004) and eye examination (aOR=1.74, p-value=0.0004) were significantly associated with higher number of outpatient visits. There was no statistically significant association between receipt of influenza vaccination with ED visits and hospitalizations. Hospitalizations for influenza were also not significantly associated with vaccination status, possibly due to ineffectiveness of the vaccine during the 2009 H1N1 pandemic. Women were more likely to be seen as outpatients or to be hospitalized when compared to men. No gender differences were observed in ED visits. Residence in a poor or rural county had no significant association with health care utilization. A modest decrease in ED utilization was noted among patients not residing in the two index counties (OR=0.71, p-value=0.045). Both comorbidity indices were strongly associated with increased healthcare utilization. Strength of association between the Elixhauser index and hospital admissions was greater than with outpatient visits, whereas the reverse was noted with the SCD index (Elixhauser index score >=2 outpatient aOR=6.8, p-value <.0001, ED aOR=4.07, p-value <.0001, ED aOR=23.65, p-value <.0001; SCD index score >=1 outpatient aOR=4.20, p-value <.0001, ED aOR=2.82, p-value <.0001, ED aOR=2.57, p-value <.0001). Although adult age was strongly associated with utilization in unadjusted models, it was not significant in the fully adjusted models. This suggests that increases in utilization observed when children transition to adulthood may be driven by increasing disease severity over time. In conclusion, outpatient-based candidate quality indicators of care examined (influenza vaccination and eye examination), were associated with increased outpatient utilization, but not acute visits or hospitalizations. Co-morbidity indexes have significant confounding effects on outpatient and hospital utilization. Administrative data based quality indicators of care such as influenza vaccination, as well as SCD specific comorbidity indexes warrant further studies as these may help better understand optimal allocation of health care resources for patients with SCD.
No relevant conflicts of interest to declare.
