Using Quality Improvement Methods to Implement an Electronic Medical Record (EMR) Supported Individualized Home Pain Management Plan for Children with Sickle Cell Disease

Abstract

Objective

Using quality improvement methodology, our goal was to develop and implement individualized home pain management plans (HPMP) that included pharmacologic as well as non-pharmacologic strategies for children with sickle cell disease (SCD). We hypothesized that successfully implemented HPMPs would have an impact on Emergency Department (ED) use, decreasing ED visits for uncomplicated SCD pain episodes.

Methods

A multidisciplinary quality improvement team developed a questionnaire to assess the frequency, location and severity of a patient’s pain during a routine, comprehensive visit in order to help the patient and family develop an effective pain management strategy using both pharmacologic and non-pharmacologic actions. Using plan do study act cycles (PDSAs), this team was able to build this process into the daily workflow for all SCD patients age 5 years to 21 years of age. Patients with comprehensive visits scheduled from January 2012 to May 2013 were included (N=188) in the intervention.

Results

By May of 2013, 88% of eligible patients had an individualized HPMP in place. There was a concomitant reduction in the percentage of SCD patients seen in the ED for uncomplicated SCD pain (6.9% vs. 1.1%).

Conclusions

Using quality improvement methods, an individualized HPMP intervention was incorporated successfully into the daily workflow of a busy outpatient SCD clinic. This intervention has the potential to improve patient outcomes by decreasing avoidable ED visits as well as reducing overall healthcare costs.

Sickle cell disease (SCD) is one of the most common genetic disorders in the United States, affecting approximately 1 in 400-500 African-American infants each year[1]. The genetic mutation that causes SCD results in the production of an abnormal hemoglobin molecule (HbS) in the red blood cells (RBC). Under low oxygen conditions, the HbS polymerizes and causes the RBCs to elongate into a sickle form (crescent shape) and decreases the life span of the RBC. Additionally, RBCs with HbS are more “sticky”, adhering to vessel walls and limiting blood flow and oxygen delivery to many tissues and organs in the body. The resultant tissue ischemia causes progressive organ injury as well as episodes of pain (vaso-occlusive crisis).

Problem/Issue

SCD pain episodes are quite unpredictable and recurrent and are the hallmark of this disease. They account for the majority of Emergency Department (ED) visits as well as hospitalizations. High quality outpatient care can reduce acute care and ED visits and decrease hospitalization rates in patients with SCD [2]. Additionally, ensuring that patients have a home pain management plan and understand how to assess and reassess their pain may improve health outcomes for the patient[3]. Data from our own population of children with SCD indicate that 40-50% of ED visits in 2011 were for uncomplicated pain episodes (no concomitant medical issues such as fever, increased respiratory rate, wheezing, worsening pallor). If these pain episodes had been effectively managed at home, the ED visits might have been avoided.

Intention/Objectives

In an effort to reduce these potentially preventable ED visits and subsequent hospitalizations, the Comprehensive Sickle Cell Center at Cincinnati Children’s Hospital Medical Center (CCHMC) assembled a quality improvement (QI) team to partner with patients and their families to develop an individualized home pain management plan (HPMP) that incorporated both pharmacologic and non-pharmacologic pain management strategies. We also sought to identify and remove barriers to the successful use of an HPMP, such as not having enough analgesics at home or not allowing enough time for analgesics to work before presenting to the ED. We ensured that medications were refilled. Finally, we documented the plan in a standard location and format in the electronic medical record (EMR) making it available to all CCHMC heath care providers. Ultimately, we hoped that this intervention would improve the management of pain at home and reduce the need for ED visits for uncomplicated pain and any resulting hospitalizations. This paper describes the development, refinement and testing of an individualized HPMP protocol and related outcomes.

Methods

Setting

The project was conducted at CCHMC, a non-profit pediatric medical center with 587 inpatient beds in Ohio providing acute and chronic care for children in Southern Ohio, Northern Kentucky and Southeastern Indiana. CCHMC’s Comprehensive Sickle Cell Center provides comprehensive care to approximately 280 children with SCD in the region from birth to 21years of age. CCHMC is the only major pediatric inpatient facility in the tri-state area. Greater than 75% of the SCD patients at our center live within a 15 mile radius of CCHMC, therefore essentially all ED visits and hospitalizations occur at CCHMC for our patients.

Participants

Patients were eligible for the intervention if they met the following criteria: 1) a diagnosis of SCD documented in the EMR (Epic Systems), 2) 5 years to 21 years of age, and 3) attendance of an outpatient visit in the CCHMC Comprehensive Sickle Cell Center on or after January 2012. We did not target the children less than 5 years of age because data on our patients’ ED use showed that the majority of the younger children with SCD did not have ED visits for pain only (only 6 of 132 visits in a 12 month period of time).

Team

The core QI team consisted of multidisciplinary health care providers with experience caring for patients with SCD, including three SCD nurse care managers, two physicians, two PhD psychologists, four nurse practitioners, a QI consultant and a data analyst. Additional support and suggestions were received from other SCD team members (e.g. social workers, school interventionists). The core QI multidisciplinary team met weekly to design and test the intervention and implementation process.

HPMP Intervention

The HPMP intervention comprised the following elements: 1) pre-visit review to identify eligible patients needing a new or updated HPMP; 2) family completion of a pain assessment tool; 3) review of pain assessment tool by SCD team; 4) development of collaborative HPMP with family and the medical team; 4) integration of non-pharmacological strategies into the HPMP (developed with the psychologist); 5) printed copy of HPMP for family to take home; 6) documentation of HPMP in the EMR (see Figure 4); and, 7) a follow-up phone call for eligible patients with ED or urgent care visits for uncomplicated SCD pain by the nurse care manager.

Figure 4.

Home Pain Management Plan Template

The pain assessment tool consisted of 13 items and measured the recent illnesses or transfusions, the patient’s pain location, intensity, associated symptoms, potential triggers, and the impact of the pain on quality of life (missed days of school/work). In addition, the patient’s current pain management strategies, perceived effectiveness of those strategies, and analgesics available at home was recorded.

HPMP Implementation

Each week the data analyst generated a list of eligible patients with ICD-9 diagnostic codes for SCD using SQL (structured query language) to extract the data from the EMR (Table 1). The SCD nurse care managers reviewed the list and notified the team of those patients needing a pain assessment and updated HPMP during the daily pre-clinic patient review rounds each morning.

Table 1.

Eligibility Criteria for SCD Patients Needing a Home Pain Management Plan (HPMP)

Criteria for Identifying Sickle Cell Patients Eligible for Home Pain Management Plan (HPMP)
MEASUREMENT: The number of sickle cell patients that have a documented home pain management plan in their EPIC problem list.
I. Description and Rationale In January 2012 the Department of Hematology introduced a Home Pain Management plan as an intervention with a goal of reducing the number of Emergency Department (ED) visits made by patients with Sickle Cell disease which were for pain only.
II. Population Definition (Inclusions/Exclusions)
The patients included in this measure have to meet the following criteria:
Have Sickle Cell disease as identified by the following ICD9 Codes in their EPIC problem list:
282.60 - Sickle-cell disease, unspecified
282.42 - Sickle-cell thalassemia with crisis
282.61 - Hb-SS disease without crisis
282.62 - Hb-SS disease with crisis
282.68 - Other sickle-cell disease without crisis
282.41 - Sickle-cell thalassemia without crisis
282.63 - Hb-SC without crisis
282.64 - Hb- SC with crisis
Are at least 5 years old (>=60 months) during the reporting period.
Do not have a Home Pain Management Plan documented in their EPIC problem list.
Have an office visit scheduled in the Department of Hematology during the reporting period.
Exclusions
Patients who are known to have transferred their care to a provider outside CCHMC are excluded from this measure.
Baseline Period: Prior to January 2012, Home Pain Management Plans did not exist. Therefore, the baseline is zero.
III. Data Source(s)
Anderson Center Condition Specific Measures Data Mart (Data Mart) and HPCE Clarity (Custom Tables)
IV. Sampling and Data Collection Plan A list of patients who do not have a Home Pain Management Plan (HPMP) and have a scheduled office visit in the Department of Hematology during the upcoming reporting period (week) is generated and sent to the improvement team. At the end of the weekly reporting period, the patients appearing on the list are reviewed to see if they received a HPMP during the office visit.
Report Format: The statistic will be expressed as the number of patients that received a HPMP divided by the total number of patients who were identified as needed a HPMP.
The statistic will be calculated weekly and displayed in a chart.
V. Calculation
Numerator: Total number of Sickle Cell Patients eligible for the measure that have documented home pain management plan in their EPIC problem list.
Denominator: Total number of Sickle Cell Patients eligible for this measure.
VI. Analysis Plan and Frequency of Reporting Run chart displaying the percent of SCD patients have a documented home pain management plan
Frequency: Weekly
VII. Reporting Venues Monthly Report Quarterly SIP Scorecard

The provider seeing the patient that day facilitated the patient and family’s completion of the pain assessment tool. After discussing the results with the team, a medical provider reviewed the findings with the patient and family and developed a plan for pharmacologic pain management at home utilizing a stepwise approach based on the World Health Organization (WHO) analgesic ladder for selecting pain-relief drugs[4, 5] and the American Pain Society Guidelines for Management of Acute and Chronic Sickle Cell Pain[6]. The medication’s method of action, side effects, risks and benefits were reviewed and prescriptions were provided as needed.

During the same visit, patients who reported acute or chronic pain within the last month met with the team psychology provider. The psychology provider educated the patient and family about pain, the mind-body connection and non-pharmacological approaches to pain management that could be incorporated in the HPMP. Following the education, the psychology provider taught the patient at least one relaxation strategy (i.e. diaphragmatic breathing, guided imagery, progressive muscle relaxation) and provided written materials to take home to encourage the practice. At the time of discharge from the clinic, patients and families received a copy of the comprehensive HPMP and any needed prescriptions for analgesics. Families were encouraged to access a copy of their HPMP at home by logging on to MyChart (Epic Systems Corporation), a limited version of the child’s EMR designed for patients and families.

After each ED or urgent care visit for uncomplicated SCD pain, the nurse care manager attempted to call the family within three business days to ask whether the HPMP had been used and determine if it needed to be revised. Medication refills were confirmed via phone follow-up by the nurse care manager at this time. Laminated pocket guides for the care managers facilitated and standardized the follow-up questions. A maximum of three attempts were made to contact the family. Information from the telephone encounter was documented in the patient’s EMR in a standard format and location. This information was then communicated to the SCD provider (nurse practitioner or physician) who modified the plan as needed. If the patient did not have any ED or acute care outpatient visits, the HPMP was reviewed every 6 months at a routinely scheduled comprehensive visit.

The team used multiple plan-do-study-act cycles (PDSAs) to refine the intervention and implementation process. One PDSA involved a focus group consisting of three young adult patients and one parent. Participants were asked if they knew what we were referring to when we used the term “home pain management plan,” what they remembered about their plan, and if they thought we should keep or change the name. All four participants reported that they were familiar with the term and were able to describe aspects of their or their child’s home pain management plan. Although one participant suggested shortening the name; the SCD team had worked to develop a high level of familiarity with the name, so it was retained. Another PDSA was conducted to assess whether the pediatric hematology fellows (post-graduate trainees) were aware of the HPMP and how to access it in the EMR. Eight of the ten fellows responded, and the majority indicated that they were aware of the HPMP; however, only one fellow knew where to locate it in the EMR. This resulted in PDSAs to increase fellows’ awareness and use of the HPMP.

The QI team also completed a failure mode effect analysis to identify (FMEA) potential failures in the clinic flow process. The FMEA helped identify low hanging fruit “quick fixes”, PDSAs and develop process maps. Weekly data guided our PDSAs and allowed us to continuously improve our processes, and team members were accountable for specific weekly action items.

Measurement/Analysis

The HPMP implementation process was monitored and tracked using two weekly run charts: one that displayed the percentage of eligible SCD patients who needed a HPMP each week that actually received one and one that showed the overall number of eligible SCD patients with a HPMP (population metric). Run charts provide a graphic display of process performance over time and allow the team to track and monitor process outcomes. The goal was that at least 85% of eligible patients would receive the HPMP intervention by November 2012.

Outcomes were evaluated using a monthly control chart showing the percentage of SCD patients seen in the ED for uncomplicated SCD pain (Figure 3). For the current project, a p-chart was used because ED visits were categorized (see below) and the sample size varied by month. We conducted a retrospective chart review of each ED visit to extract the initial complaint and the final assessment from the ED providers’ notes. ED visits were categorized as follows: 1) fever (with or without other symptoms such as pain), 2) uncomplicated SCD pain only, and 3) other (e.g. trauma, asthma). The goal was to monitor ED visits for uncomplicated SCD pain only to determine if the rate of this type of ED visit decreased after the implementation of the HPMP. Based on the chart review of the 12 months prior to the implementation of the HPMP, the majority of SCD patients seen in the ED had 0-3 ED visits for uncomplicated SCD pain. Only 7 patients had more than 3 ED visits:

• two had 4 ED visits,

• two had 5 ED visits,

• one had 6 visits,

• one had 7 visits,

• one had 13 visits.

Because the patient with 13 visits has complex psychosocial issues that greatly impact the use of the ED and inpatient medical services, this data was excluded from our analyses.

Figure 3.

Percent of ED Visits for Uncomplicated SCD Pain

IRB Approval

The Children’s Hospital Medical Center Institutional Review Board exempted this study from review because it was deemed to be a QI project with the intent to improve care locally and not to develop generalizable knowledge.

Results

HPMP Intervention Implementation

As of May 2013, 100% of eligible patients who needed a HPMP actually received a HPMP during their clinic visit (Figure 1) and this performance was maintained for four weeks. Almost all (98%) of these patients (N=183) met with the team psychology provider. Of the five patients who were referred to the team psychology provider but not seen:

• one declined the service;

• two patients were experiencing acute pain;

• one family was overwhelmed by the amount of medical information received at the visit and asked that the team psychology provider see them at their next visit;

• one patient was asleep.

Figure 1.

Percent of Patients Receving Home Pain Management Plan (Weekly Measure)

We feel our success on this goal was due to regular weekly meetings to develop process maps, complete a failure mode analysis (FMEA) to identify interventions to test in PDSAs and to monitor progress on PDSAs. The goal was that at least 85% of all eligible patients would receive the HPMP intervention by November 2012 (population metric). However, we did not meet 85% until February 2013 due to scheduling difficulties during that time period. By May of 2013, 88% (N=188) of all eligible patients in the population had received a HPMP (Figure 2). Of the 12 eligible patients who did not receive a HPMP:

• six were in the process of transitioning to an adult hematologist and were being seen for the last time in the pediatric SCD center;

• one patient needed to leave and didn’t have time to complete the intervention;

• one patient arrived in acute pain;

• one patient had other treatment issues to discuss and asked to postpone the pain management discussion;

• three patients did not have an obvious reason, but the provider likely did not have time to implement or forgot to implement the HPMP protocol.

Figure 2.

Percent of Eligible Sickle Cell Patients with a Home Pain Management Plan (HPMP)

ED Visits for Uncomplicated SCD Pain

From January 2012 to May 2013 the percent of patients who were seen in the ED for uncomplicated SCD pain decreased from 6.9% to 1.1%. Hence, there was an 84% decrease in this type of ED visit. With respect to return visits to the ED, 1.3% of eligible patients (N= 2) returned to the ED for treatment of uncomplicated SCD pain within a 30-day period of a prior visit for uncomplicated SCD pain.

Discussion

Using quality improvement methods, an individualized HPMP intervention was incorporated successfully into the daily workflow of a busy outpatient SCD clinic. The QI team provided critical guidance, organization, and resources for refining the HPMP intervention and implementing it into a very busy outpatient clinical setting. QI methods such as the PDSAs, FMEA, process maps allowed us to continuously improve the intervention and develop an effective implementation process. As a result, we were able to reach our goal of ensuring that 100% of eligible patients received a HPMP during their clinic visit.

Several studies have shown cognitive-behavioral therapies, such as relaxation, imagery, and self-hypnosis, to improve outcomes in children and adults with SCD [7-10]. Therefore, we believe that having psychology providers on our team who could train families in nonpharmacological strategies was critical to the project’s success. Most SCD patients are taught to increase fluid intake and use warm compresses, but few are trained in adjunctive non-pharmacologic strategies while awaiting the effects of oral analgesics. Thus, our multidisciplinary protocol is innovative and future studies may prove the intervention to be more effective than interventions using pharmacological or nonpharmacological strategies alone.

Implementing a comprehensive HPMP intervention into a very busy clinical setting was quite challenging; it required a substantial coordination and communication among the clinical team. Although each member of the team had a well-defined role, we found that our nurse care managers were the drivers of the process during the clinic visit. They ensured the documentation of the HPMP and reconciliation of medications were completed in the EMR, that prescriptions for analgesics were written and educated families to execute the HPMP.

We were able to exceed our goal of ensuring that at least 85% of eligible patients in our population would have a HPMP in place (88% had a HPMP). This is clinically significant as most SCD pain episodes occur at home[11]. Typically, the pain management strategies used by patients and families at home are inconsistent, and several studies indicate that parents may be reluctant to use analgesics for their children, use a dose that is too small, or do not give the medicine often enough[12- 14]. Developing an HPMP with a patient and family allowed for education about distinguishing different types of pain and the appropriate use of medications for specific types of pain.

Challenges to implementation of the HPMP protocol included limited time during clinics visits to integrate the HPMP given competing clinical issues. Some families felt the visit lasted too long and were eager to leave the clinic without further delays. Additionally, the fixed design of the EMR posed some limitations related to documentation, medication reconciliation and updating of the HPMP because different team members could not simultaneously access some parts of the EMR. We also initially overlooked the need to educate other providers in our division about the HPMP, such as fellows who take calls about patients after hours. This has subsequently been addressed via ongoing PDSAs to test processes for making fellows aware of the HPMP and to ensure they use it consistently to coordinate care.

Following implementation of the HPMP protocol, the percentage of ED visits for SCD uncomplicated pain decreased by 84%. These results build on the previous literature which has focused primarily on standardized pain management protocols in the ED [15-17]. However, it makes a unique contribution in that the focus was on systematically teaching families strategies to use at home with the goal of minimizing the need for ED or urgent care intervention. We also learned more about the reasons for some ED visits; there were patients who presented to the ED with presumed acute SCD pain that actually had acute exacerbations of chronic back pain (seen in 8 patients), headaches (5 patients), or abdominal pain due to constipation (12 patients). Each of these is managed differently than acute SCD pain, and the HPMP was not designed for these conditions. In addition, we discovered that a few patients (3 patients) used opiate analgesics for difficulties with sleeping rather than pain further supporting the need for ongoing patient/family education about pain management in pediatric SCD.

We conclude that the HPMP intervention served to empower patients with SCD and their families by providing them with the tools to manage uncomplicated pain events at home thereby reduce utilization of the ED. Hence, the HPMP intervention has the potential to improve patient outcomes by decreasing avoidable ED visits and reducing overall healthcare costs. It is hoped that other clinics or hospitals could use QI methods to implement HPMPs that would allow achievement of similar outcomes. Finally, this paper contributes to the limited literature on both QI and home management in pediatric SCD and addresses a critical gap in the literature: a clinical approach to reducing potentially preventable ED visits and subsequent hospitalizations for youth with SCD. It also serves as the basis for future innovative research examining the relationship between home pain management, healthcare utilization and healthcare costs.