BRIEF OBSERVATION

Quality Improvement Process in a Sickle Cell Infusion Center Lauren N. Whiteman, MPH,a Sophie Lanzkron, MD, MHS,a Rosalyn W. Stewart, MD, MS, MBA,a,b Carlton Haywood, Jr, PhD, MA,a,c John J. Strouse, MD, PhD,a,b Leonard Feldman, MDa,b Departments of aMedicine and bPediatrics, Johns Hopkins University, Baltimore, Md; cBerman Institute of Bioethics, Johns Hopkins University, Baltimore, Md.

ABSTRACT BACKGROUND: The American Pain Society recommends that individuals experiencing sickle cell crisis receive parenteral pain medication within 30 minutes of assessment. We examined factors affecting achievement of this standard at the Johns Hopkins Sickle Cell Infusion Center. METHODS: Baseline patient care time intervals and data on variables affecting the ability to achieve the American Pain Society goal were measured. Time to first parenteral opiate administration was modeled using simple and multivariable linear regression. RESULTS: Mean time from initial assessment to first dose was initially 41 minutes. Increased nurse to patient ratio decreased time to first dose. CONCLUSIONS: Of the factors associated with improved times to first dose, only nurse to patient ratio is amenable to process change, suggesting it as a potential target for future interventions. Ó 2015 Elsevier Inc. All rights reserved.
The American Journal of Medicine (2015) 128, 541-544 KEYWORDS: Acute care; Quality improvement; Sickle cell disease

Sickle cell disease is a life-threatening inherited red blood cell disorder affecting approximately 100,000 individuals in the United States.1 Episodes of acute “pain crises” are the most common clinical manifestation and are linked to increased mortality.2 Care for pain crises is typically sought in emergency departments and is often inadequate.3 The American Pain Society guidelines regarding pain management for sickle cell disease recommend that parenteral analgesics for acute pain crises should begin within 30 minutes of arrival.4,5 This recommendation is habitually unmet,3 with the median time to first dose of analgesic ranging from 77 to 107 minutes.3 From January to June 2012, the

Funding: LNW, RWS, JJS, and LF were funded by Health Resources and Services and Administration Award 5 U1EMC17183-03-00. SL was funded by a Career Development Award from the National Heart, Lung, and Blood Institute (K23HL083089). CH was funded by a Career Development Award from the National Heart, Lung, and Blood Institute (1K01HL108832-01). Conflict of Interest: None. Authorship: All authors had access to the data and played a role in writing this manuscript. Requests for reprints should be addressed to Lauren N. Whiteman, MPH, 601 N. Caroline St, Suite 7143, Baltimore, MD 21287. E-mail address: [email protected] 0002-9343/$ -see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.amjmed.2014.11.020

mean time to first dose at the Johns Hopkins Adult Emergency Department was 166 minutes.6 Infusion centers can model how to manage patients with sickle cell disease who are experiencing acute pain crises by demonstrating a reduction in time to pain relief and hospital admissions.7 The Johns Hopkins Hospital Sickle Cell Infusion Center opened in 2008. The Sickle Cell Infusion Center is open 7 days a week for 8 hours and provides individualized care for persons with sickle cell disease experiencing a pain crisis. Staffing at the Sickle Cell Infusion Center includes 1 to 2 registered nurses, a physician assistant, a medical office coordinator, a clinical technician, and a social worker. It contains 4 infusion beds and 1 examination room, and averages 4.1 patients daily. All staff have extensive training in acute and chronic sickle cell disease management. When the Sickle Cell Infusion Center is open, patients arrive from home or as an emergency department transfer. Patients arriving from home first call the clinic to assess whether their symptoms can be accommodated in the Sickle Cell Infusion Center. If patients presented to the emergency department while the Infusion Center was closed and require continued outpatient treatment when the Sickle Cell Infusion Center opens, they may be transferred.

542

The American Journal of Medicine, Vol 128, No 5, May 2015

Historical data from the Sickle Cell Infusion Center pain reassessment and discusses with the provider the next suggest that this model achieves outcomes closer to the steps in treatment. The process map (Figure) outlines the American Pain Society’s guidelines than most emergency steps involved in pain management and mean times for departments.3,6 By using a slightly different outcome meaeach step. sure than the American Pain Society, we measured time from triage to first parenteral opioid instead of time from regisProcess of Care Outcomes tration to first parenteral opioid. and Predictors of Interest The National Initiative for ChilCLINICAL SIGNIFICANCE In February 2013, the Sickle Cell dren’s Healthcare Quality recomInfusion Center staff designed a mended this modified outcome
A refined process for acute care mandata collection sheet to measure measure. agement will ensure that all patients timing of care (Figure). Outcome Quality improvement relies on receive safe, effective, efficient, and variables included time of arrival; continuous and methodical apequitable care. registration time; initial assessment proaches to measurable improveby clinical technician, nurse, and
Alternative routes of administration for ment in the health of individuals prescribing provider; provider or health care services.8 Quality parenteral analgesic or technology orders; first dose of opiate improvement efforts can focus on improving time to intravenous access are analgesic administration; and care processes and patienta possible solution to improve time to pain reassessment. Potential centered outcomes. The objecfirst parenteral analgesic. predictor variables were order of tives of our quality improvement patient arrival, type of access
Improving the timeliness of pain control effort were to (1) provide a (peripheral or central access, or is a patient-centered approach to comprehensive description of a access obtained in emergency variety of process variables for the improve outcomes. department), daily patient census, Sickle Cell Infusion Center and and nurse to patient ratio. (2) examine determinants of the The daily patient census was retrieved from the Johns process variable, time to first dose. Ultimately, we wanted to Hopkins electronic medical record. The number of nurses determine target areas for process improvement to achieve was based on nurse schedules. the American Pain Society’s guidelines.

MATERIALS AND METHODS

Data Analysis

Infusion Center Operating Procedure On arrival to the Sickle Cell Infusion Center, the medical office coordinator registers the patient. The clinical technician assesses the patient’s vital signs and begins peripheral intravenous line placement. The clinical technician attempts to place a peripheral intravenous line up to 3 times. If unsuccessful, a nurse places the line. Only a nurse can access implantable central venous access devices. The patient is assessed by the nurse and provider; the provider then places orders. After medication administration, a nurse conducts

Figure

Analyses were conducted using Stata version 12 (StataCorp LP, College Station, Tex). We modeled time to first parenteral opiate using simple and multivariable linear regression with adjustment for clustering of observations within patients. For our bivariate analyses, we conducted simple linear regression on all variables examining unadjusted association of potential predictors and time to first dose. Any variable with a statistically significant result (P  .05) was subsequently included in a multivariable linear regression model.

Mean times process map.

Whiteman et al

Quality Improvement in a Sickle Cell Infusion Center

RESULTS

Table 2

We analyzed data from 77 visits from 54 unique patients over 5 weeks. Visits with missing data (time of vitals, time of first dose, or if the patient did not receive parenteral opioids) were excluded. The mean age of patients was 33.9 years, 59.3% were female, and 25% had central venous access (Table 1). The care variables are listed in Table 2 and include time from vitals to intravenous access, time to obtain intravenous access, time from initial assessment to first dose of parenteral opiate analgesics, and time to pain reassessment. The order that patients arrived (ie, first patient of the day, second patient of the day) had no effect on time to first dose (95% confidence interval [CI], 0.95 to 2.6; P ¼ .4), even when comparing the first patient of the day with all other patients (95% CI, 10.9 to 2.3; P ¼ .2). Patients arriving simultaneously (defined as registered within 10 minutes of one another) did not affect time to first dose (95% CI, 6.1 to 13.7; P ¼ .45). The total number of patients seen for the day did not affect time to first dose (95% CI, 1.2 to 3.1; P ¼ .4). When comparing the 3 types of intravenous access—peripheral, central, or transfer from the emergency department with access—the only statistically significant difference in time to first dose was between peripheral and central access (95% CI, 2.1-18.9; P ¼ .016). Arrival from the emergency department with access versus requiring access on arrival, either peripheral or central, likewise was statistically significant (95% CI, 16.2 to 0.5; P ¼ .04), suggesting that one barrier to achieving the American Pain Society’s guidelines is obtaining intravenous access. In simple analysis (Table 3), 3 factors were associated with decreased time to first opioid analgesic dose: increased nurse to patient ratio, patients who arrived from home and required intravenous access, and patients who arrived from the emergency department with intravenous access already secured. For the nursing to patient ratio, we found that increasing nurse to patient ratio from 1:4 to 1:2 decreased time to first dose by 33 minutes (95% CI, 60.6 to 5.4; P ¼ .020). Patients who arrived from home and required peripheral intravenous access, versus central access, received medication 10.5 minutes faster (95% CI, 18.9 to 2.0; P ¼ .02). Patients arriving from the emergency department with intravenous access had a reduced time to first dose by 8 minutes (95% CI, 16.2 to 0.5; P ¼ .04) compared with those requiring intravenous access (peripheral or central).

Demographics

Mean age (y) Percent female Access type Peripheral line Central Accessed in ED ED ¼ emergency department.

33.9 59.3% 36% 25% 37%

Care Variables

Time from. (in minutes)

Interquartile Standard Mean Median Range Deviation

Vitals to access Start to finish of IV line placement Vitals to medication administration Medication administration to pain reassessment

12 11

9 10

8 7

11 5

41

38

21

15

36

33

16

16

IV ¼ intravenous.

In multivariable analysis, we controlled for nurse to patient ratio, peripheral versus central access, and arrival with access from the emergency department. Only nurse to patient ratio was statistically significant. If the nurse to patient ratio increased to 1:2, a 61.9-minute reduction in time to first dose was determined (95% CI, 107.4 to 16.3; P ¼ .009).

DISCUSSION Within our Sickle Cell Infusion Center, patients receive parenteral pain medication faster with increased nurse to patient ratios. All other factors that decreased time to first dose were not significant in the multivariable analysis. Demonstrating that additional staffing can improve the efficiency of the Sickle Cell Infusion Center is not surprising. Additional staff allows nurses to spend more time with each patient and helps to provide individualized care. We expected that patients who were evaluated previously in the emergency department and already had intravenous access on arrival would receive medication faster; this was not the case. Unexpectedly, the order of patient arrival and daily patient census did not affect time to first dose. There are several potential approaches to achieving the American Pain Society goal of 30 minutes to first dose. Completing steps concurrently rather than serially could improve time to first dose. For instance, while the clinical technician is obtaining intravenous access, the provider could gather the patient’s medical history. However, this may not be perceived as a patient-centered approach. Other possible improvement strategies include using technology to Table 3 Factors to Improve Time to First Dose (Univariate Analysis) Factor

Table 1

543

b (min) P Value 95% CI (min)

Arriving with IV access L8.3 RN to patient ratio L33.0 Central vs peripheral access L10.5 First patient to arrive 4.3 for the day Order of patient arrival 0.8 Total census for the day 0.98

0.04 0.020 0.03 0.2 0.4 0.4

L16.3 to L0.3 L60.6 to L5.4 L19.9 to L1.0 10.9 to 2.3 0.95 to 2.6 1.2 to 3.1

Bold values indicate statistical significance. CI ¼ confidence interval; IV ¼ intravenous; RN ¼ Registered Nurse.

544 obtain peripheral access quickly or providing initial opiate treatment through more readily accessible routes, such as intranasally.9

The American Journal of Medicine, Vol 128, No 5, May 2015 centers throughout the country. This model of care for sickle cell disease is becoming more common, and process improvement for these new centers will be necessary to achieve the best outcomes.

Study Limitations Our study may be limited by the small number of observations and potential confounding by unmeasured variables. In addition, a single site limits generalizability. Our center’s model is similar to others, and we do not believe generalizability is a major limitation. Although the American Pain Society recommends 30 minutes to first parenteral dose of pain medication, there is no published evidence that this improves health outcomes. To justify additional expenses, such as using new technologies to obtain intravenous access or hiring additional nurses, we must demonstrate a positive effect on patient outcomes.

CONCLUSIONS The Johns Hopkins Sickle Cell Infusion Center is close at baseline to achieving 30 minutes to the first parenteral dose of pain medication for sickle cell pain crises. Nurse to patient ratio is amenable to process and system changes. Intravenous placement technology and alternative routes for pain medications delivery may achieve the goal. Further study is needed to determine whether these improvement strategies affect patient outcomes. Although the Sickle Cell Infusion Center is a specialized model of patient care, it is similar to sickle cell infusion

References 1. Hassell K. Population estimates of sickle cell disease in the U.S. Am J Prev Med. 2010;38:S512-S521. 2. Platt OS, Brambilla DJ, Rosse WF, et al. Mortality in sickle cell disease. Life expectancy and risk factors for early death. N Engl J Med. 1994;330:1639-1644. 3. Tanabe P, Myers R, Zosel A, et al. Emergency department management of acute pain episodes in sickle cell disease. Acad Emerg Med. 2007;14: 419-425. 4. Benjamin LJ, Dampier CD, Jacox A, et al. Guideline for the Management of Acute and Chronic Pain in Sickle-Cell Disease. Glenview, IL: American Pain Society; 1999. 5. Benjamin LJ, Dampier CD, Jacox A, et al. Guideline for the Management of Acute Pain in Sickle Cell-Disease Quick Reference Guide for Emergency Department Clinicians. Glenview, IL: American Pain Society; 2001. 6. Feldman L, Lanzkron S, Strouse JJ, et al. Examining emergency department and infusion center wait times to improve the quality of sickle cell care. Proceedings of the Sickle Cell Disease Association of America 40th Annual Convention, Baltimore, Maryland, September 26-28, 2012. 7. Benjamin LJ, Swinson GI, Nagel RL. Sickle cell anemia day hospital: an approach for the management of uncomplicated painful crises. Blood. 2000;95:1130-1136. 8. Massoud MR. Advances in quality improvement: principles and framework. QA Brief. 2001;9:13-17. 9. Lin S, Strouse JJ, Whiteman LN, Anders J, Stewart R. Improving quality of care for sickle cell patients in the pediatric emergency department. [Epub ahead of print]. Pediatr Emerg Care. 2015.