Abstract
Abstract
Background:
Benefits of palliative care have been extensively described; however, reports on adherence to national quality indicators are limited.
Objectives:
This study focuses on describing the characteristics of patients who were seen at an urban academic hospital and their care team's adherence to 5 out of 10 Measuring What Matters (MWM) quality indicators.
Design:
Retrospective chart review
Setting/Subjects:
Patients seen by inpatient palliative care service from January 2014 to December 2015 in an urban academic hospital.
Measurements:
Patient age, gender, ethnicity, disease category, discharge end point, life-sustaining preferences, surrogate decision-maker documentation, and initial palliative assessment were analyzed using descriptive, parametric, and nonparametric statistics.
Results:
During two years, 1272 patients were seen by the inpatient palliative care service. Fifty-one percent of patients were male, with an average age of 68 years. The majority were Caucasian (57%) and African American (41%). Life-limiting illnesses included were cancer, complex chronic illnesses, and gastrointestinal illness. Adherence to comprehensive palliative care assessment was measured at 64%; initial visit assessment for physical symptoms was 38%; code status preference was 99%; care consistent with preference in vulnerable elders was 99%; and surrogate documentation was noted at 33%. Compared to hospital patients discharged without hospice, patients discharged with hospice care had consults with higher adherence to comprehensive assessment and surrogate documentation quality standards (p < 0.05).
Conclusions:
Adherence to MWM measures was variable. Subjects discharged with hospice services were more likely to receive comprehensive assessment within 5 days of admission and surrogate documentation compared to those subjects without hospice care.
Introduction
R
In light of these advances in palliative care, it is important to examine adherence to national quality indicators and the feasibility of measuring MWM indicators. This study focuses on describing the characteristics of a palliative service and adherence to 5 out of 10 MWM quality indicators of patients who were seen at Saint Louis University Hospital (SLUH) inpatient consult services over a period of two years. The study explores differences in adherence to standards by patient characteristics and correspondence between adherence and discharge end points.
Methodology
Study design
This is a retrospective study of patients seen by the palliative consult service at SLUH from January 2014 to December 2015. Data from the SLUH palliative care consult service were collected prospectively as part of routine clinical care quality assessment at the SLUH inpatient consult service. The SLU Institutional Review Board approved this project.
Setting
SLUH is a 356-bed adult hospital. The inpatient consultation service consisted of two nurse practitioners, a rotating medical student or house staff, and a physician. During this study period of 2 years, the inpatient team saw on average 52 consults per month (24–80).
Definition of MWM measures adherence
This study assesses clinician adherence to 5 of the 10 MWM items. The remaining five MWM items were not routinely documented. 12 For this analysis, an adapted methodology was used to operationalize MWM measure adherence as described by Kamal et al.13,14 (Table 1).
Measuring What Matters Measure Definition
DPOA, durable power of attorney; MWM, Measuring What Matters.
Data collection
Nurse practitioners maintained records of consultations initially in paper summary forms; then later, the facility adopted Epic electronic health record (EHR). Reports were automatically generated for hospital administrative purposes. Secondary chart review of the EHR was used to complete discharge disposition and other incomplete variables in the computerized consult spreadsheets. All data collection was based on the first point of palliative care contact.
Demographic and patient characteristics extracted were age, gender, ethnicity, and disease category based on admission diagnosis. This includes cancer, complex chronic, gastrointestinal, neurological, or other disease categories. The complex chronic disease category entailed more than three active disease categories. Discharge end point was classified as expired in the hospital, home with hospice or palliative services, home with home health, nursing home with or without skilled nursing, and other end point.
Data analysis
Demographic data were examined using descriptive statistics. Parametric and nonparametric statistics were used to analyze MWM adherence and univariate associations with sample characteristics. Data analyses were performed using Stata 14.2 (College Station, TX).
Results
There were 1272 patients seen in the inpatient palliative consult service between 2014 and 2015. Table 2 shows patient demographics and sample characteristics. The patients were predominantly Caucasian (57%) and African American (41%) male (51%) with an average age of 68 ± 14 years. Most patients had a noncancer-related (58%) primary diagnosis, and among living patients 25% were discharged with hospice or palliative care.
Sample Characteristics
Other ethnicities include Hispanic/Latino, Asian, Pacific islander, and others.
Other disease categories include cardiac, infectious, metabolic, pulmonary, renal, trauma, vascular, and musculoskeletal. Complex chronic disease category entailed more than three active disease categories.
Other end point includes long-term acute care or rehabilitation facility or transfer to federal or outside hospital, psychiatric unit, shelter, rehabilitation center, team signed off, or patient left against medical advice.
Decreased time to consult was related to age, p < 0.001.
SLU, Saint Louis University.
Table 3 summarizes adherence to five MWM quality indicators. In general, there was moderate-high adherence (more than 60%) to comprehensive assessment within 5 days of admission and documentation of code status preferences compared to lower adherence to surrogate documentation and initial visit assessment for physical symptoms (less than 40%). Specifically, care team adherence to documented assessment of distress was 100%, but consult within 5 days occurred only in 64% of patients with a mean of 6.4 ± 8.4 days. Univariate analysis demonstrated that older age and non-White gender (African American and others combined) were associated with shorter time to palliative consult (p < 0.001, MWM #1). In addition, patients discharged to hospice were more likely to have a comprehensive assessment (68% vs. 62%, p = 0.049, MWM #1) and to have surrogate decision-maker documentation (38% vs. 31%, p = 0.006, MWM #7) measures compared to those who were not discharged to hospice.
Measure Feasibility and Relation to Hospice Discharge
Discussion
This study demonstrates moderate-high adherence to measures of palliative comprehensive assessment within 5 days of admission and documentation of treatment preferences. This was true regardless of gender and disease category. Age and non-White ethnicity were related to shorter time to palliative consult. This may be reflective of differences and higher palliative care needs among elderly minority patients compared to Caucasians. However, there was lower adherence to measures dealing with pain assessment and documentation of surrogate. Low completion rates for these measures have been a consistent issue since the inception of SLUH inpatient palliative service (e.g., similar completion of code status (96% in 2009–13 vs. 99% in 2014–15) and durable power of attorney (DPOA) (42% in 2009–13 vs. 33% in 2014–15) documentation). 15 In many instances, patients and families were overwhelmed by acute illness and delayed DPOA discussion or the primary team requested not to discuss.
Similar to our study, LeBlanc et al. described lower adherence to assessment of pain among patients with hematologic malignancies compared to patients with solid tumors. This may be a reflection of other differences among patients with hematologic malignancies, such as care setting or later referral to palliative care amid more serious illness, which may change clinical priorities as to which items to address. 14 In our study, lack of 100% adherence may be because these quality indicators do not adequately describe every encounter and the culture of palliative care practice in this urban hospital is not aligned with documenting MWM. 13
This is the first study that investigates the relationship between adherence to MWM measures and discharge end point. Among five discharge end points, only one, hospice/palliative care, was related to adherence to MWM #1. Inpatients discharged with hospice/palliative services were more likely to receive comprehensive assessment within 5 days of admission and surrogate documentation compared to those inpatient subjects without hospice/palliative care end point (p < 0.05). This may reflect current practice that surrogate decision maker is documented as part of the hospice referral process. While it is not surprising that palliative care intervention results in higher hospice discharge,7,8 this result suggests that a palliative care intervention with adherence to MWM quality standards has a higher impact in hospice discharge, likely because palliative care involvement helps patients and providers understand their prognosis, establish goals of care, and formulate discharge plans in line with those goals. 16 However, hospice discharge is not the only end point for palliative care interventions that meets patient and family needs; further outcomes are worth exploring.
Ryan et al. 17 describe impediments to attainment of quality end points, and in our study, at the time of data collection there were both provider and patient related factors that prevented high-quality palliative care as measured by MWM. Sometimes palliative consultants were specifically requested not to discuss code status or the patient or family declined palliative discussion because they were overwhelmed by acute illness. 17 Similar to other studies, we were unable to consistently collect 5 out of 10 MWM measures12,18,19 from the EHR system because of insufficient time and infrastructure. While there is room for improvement, our study demonstrates a way to integrate EHR into the quality assurance process using auto-generated monthly reports. Such reports need to integrate the components of MWM measures that were missed (MWM # 3–6 and10), which can be supplemented with point-of-care data (e.g., a doc flow sheet that feeds data into a registry).
Limitations
This study has several limitations. First, the nature of our study was to describe current practices in one institution, therefore generalizability is limited. Second, monthly reports were not built to capture all MWM quality indicators but rather to capture employee productivity. Therefore, not all measures were accurately available in the dataset. Third, these data can neither tell us why a particular measure was not met nor what happened after the first point of contact. Monthly reports were based on first point of contact rather than the continuum of care by the palliative team. In many instances, time is needed to build relationships and trust so that complex decision making can be achieved.
Conclusion
Two years of palliative medicine service in an urban academic hospital demonstrated variable adherence to five MWM quality care standards. Compared to patients discharged without hospice, patients discharged with hospice had higher adherence to comprehensive assessment and surrogate documentation quality standards. Future studies are required to explore better ways to monitor quality care standard adherence integrated with EHR and more reflective of care delivery, as well as to examine its relationship with outcomes.
Footnotes
Acknowledgments
The authors are thankful to Joshua Stancil for editions to the article, to the staff in the St. Louis University Hospital in St. Louis, MO for their collaboration in data collection, and for the support of the Cambia Health Foundation Sojourns Scholarship Leadership Program.
Author Disclosure Statement
No competing financial interests exist.