Implementation of a Standard Order Set at a Sickle Cell Acute Care Observation Unit

INTRODUCTION:

Acute painful vaso-occlusive crises (VOC) in sickle cell disease (SCD) are the leading cause of emergency department (ED) encounters and frequent hospital admissions. For patients presenting with an uncomplicated VOC, acute care observation units (ACOU) have previously been shown to reduce admission rates and length of stay. We wished to evaluate if implementing a standardized acute care order set (ACOS) at the University of Illinois Hospital Sickle Cell ACOU would decrease the time to first dose of analgesic medication, inpatient hospital stays, and subsequent admissions to the ACOU.

METHODS:

The ACOS includes standard orders for laboratory tests to monitor severity of sickle cell hemolysis, intravenous fluids, and analgesic medications including opioids. We conducted a retrospective analysis to evaluate if the ACOS enhanced workflow and improved the timeliness of treatment in patients experiencing a VOC. The ACOS was created in April 2017 and we compiled data from the three months before (January-March 2017) and after (May-July 2017) ACOS creation. We collected data on the time it took to administer the first opioid dose, admission rate, length of stay, number of acute care visits, ED visits, and inpatient hospitalizations in a three-month span, and demographics including variations in age, gender, and sickle cell disease genotype. Patient data was collected from a pharmacy-generated list of patients who received narcotics in the ACOU during the aforementioned time period. We analyzed the effect of the ACOS on the aforementioned variables. A mixed effects linear model was used to compare time to first dose of opioids and length of stay between data sets. A mixed effects logistic model was used for binary outcomes. Covariates of age (years), gender, and severity of sickle cell hemoglobin genotype (severe: HbSS, HbS beta0 thalassemia; mild: HbSC, HbS beta+ thalassemia) were included in the models. Statistical analyses were carried out in R version 3.4.3.

DISCUSSION:

The pre-ACOS data set contains 291 patient encounters for 76 patients with a median age of 37 years (interquartile range [IQR], 30-47 years), 66% female, and 71% with severe genotypes. The post-ACOS data set contained 289 patient encounters for 80 patients with a median age of 32 years (IQR, 27-45 years), 80% female, and 73% with severe genotypes. Implementation of an ACOS was associated with decreased time to pain management by 3.7 minutes (p=0.077) in patients presenting with uncomplicated VOC and with fewer repeat visits to the ACOU in the studied 3-month period [OR 0.35 (95% CI 0.13-1.00), p=0.049]. The median number of opiate doses received by patients in both data sets was 3. Using 3 as a cutoff, the implementation of the ACOS was also associated with more patients receiving >3 doses of opiates [OR=1.84 (95% CI 1.05-3.19), p = 0.033].

We demonstrate that in SCD patients experiencing VOC, a standardized ACOS was associated with a trend to reducing time to receiving pain management, with increased total opioid doses during the ACOU admission (suggesting better pain control), and subsequently with a statistically significant reduction in the number of repeat ACOU visits in the studied 3-month period. We have shown that a standardized ACOS that streamlines workflow in an ACOU may play an important role in delivering timely and quality care to patients with SCD.