Hospices' Preparation and Practices for Quality Measurement
Article Outline
Abstract
Context
Hospice and palliative care organizations need to measure and analyze quality of care, in response to national palliative care practice guidelines and new hospice regulatory requirements. Little is known about hospices' readiness to meet this new mandate.
Objectives
We analyzed data from a national survey of hospices to describe preparation and practices for quality measurement and research and to examine associated organizational characteristics.
Methods
Web-based survey of hospice staff responsible for quality of care.
Results
Survey respondents represented 652 National Hospice and Palliative Care Organization (NHPCO) member hospice organizations; 52% were participating in the NHPCO Quality Partners program. Most of these hospices involve clinical providers in decisions to change care practices (69%) and participate in quality improvement projects (64%), but research participation is uncommon (16%). Many hospices collect data about staff certification and training (76%) and use family surveys to measure care quality (70%). A minority of hospices have clinical data in electronic format (13%–29%). Large size, multiple sites, government ownership, and presence of a change leader in the organization were the characteristics associated with greater preparation for quality improvement and research.
Conclusion
Current organizational activities, data collection rates, and use of electronic data systems may limit hospices' preparation and practices related to quality improvement and research participation; larger size and designation of a change leader are associated with greater capacity. Hospices may need technical assistance and training to provide for meaningful measurement of quality of care.
Key Words: Hospice, quality measurement, quality improvement
Introduction
Hospice and palliative care services are expanding in the United States at a time when new research is informing and reshaping best practices. Although hospice providers have traditionally enjoyed a high level of consumer satisfaction with services, maintaining high-quality care is both desirable and challenging during times of rapid change. These health care organizations will need the ability to collect and analyze data as a necessary tool to facilitate quality improvement efforts, permit benchmarking across organizations, and test best practices that improve patient outcomes.1
Data collection and quality improvement have been mandated for years in other health care settings but may be relatively new to hospice and palliative care. Providers of hospice and palliative care have recognized these trends and taken major steps to equip organizations to measure and improve quality of care. In 2004, representatives of leading hospice and palliative care organizations endorsed the first Consensus Clinical Practice Guidelines for Quality Palliative Care.2 In 2006, the National Quality Forum reviewed and endorsed these guidelines and a list of 38 preferred practices to provide an operational definition of high-quality care.3 In November 2008, the National Quality Forum announced the formation of the National Priorities Partnership, a group of 28 partners with significant influence in American health care. Palliative and end-of-life care was named as one of the six priority areas; specifically, “The Partners envision healthcare capable of promising dignity, comfort, companionship, and spiritual support to patients and families facing advanced illness or dying, fully in synchrony with all of the resources that community, friends, and family can bring to bear at the end of life.”4
Measuring and improving hospice quality has now become a regulatory requirement. In June 2008, the Centers for Medicare and Medicaid Services (CMS) published new conditions of participation for hospice organizations, thereby requiring a new approach to quality assessment and performance improvement.5 Hospices are now required to use a structured, data-driven approach to understand and demonstrate the quality of care they deliver and to develop strategies to improve processes and outcomes of care. However, quality measurement in hospice and palliative care may be limited by lack of experience with quality improvement and by lack of staff training and preparation for translating research evidence into practice innovation.
In anticipation of these needs, the National Hospice and Palliative Care Organization (NHPCO) launched its Quality Partners program in 2006. To understand the readiness of hospice organizations to measure and improve quality of care, a national survey of NHPCO member hospices was conducted. We analyzed these survey responses, aiming to 1) describe current hospice practices relevant to quality measurement and research; 2) describe current hospice preparation and practices in data collection; and 3) examine characteristics of hospices associated with preparation for quality measurement and research.
Methods
In October 2007, the NHPCO national office conducted a web-based survey of member organizations. Electronic notice about the survey, and its relevance to hospice organizations, was sent to NHPCO's member hospice and palliative care organizations with valid e-mail address connections to NHPCO. The electronic notice requested completion of a 10-minute web-based survey by the individual in the hospice organization who was most knowledgeable about quality measurement and quality of care improvement efforts. The introductory letter that was signed by NHPCO leadership included a link to the web survey. Nonrespondents received three prompts to complete the survey, which remained open for a three-week period.
Survey items were developed specifically for this study, with input from the investigators and members of the NHPCO leadership, research, and quality teams. The survey included questions about respondent and organizational characteristics, including the question, “Does your organization have a ‘champion’ or ‘change leader’ whose primary job is to identify opportunities and strategies to improve quality of care?” Twelve items asked about the frequency of current practices and activities related to quality improvement and research. These items used 5-point Likert-scaled responses to describe the frequency of specific activities from “always” to “never.” A second group of survey items asked respondents about current use of data to measure quality, data sources, and whether or not data reports were electronic. A final group of survey items asked respondents to consider potential uses of quality-of-care data for hospice organizations—in quality improvement projects, public reporting, pay for performance, or research. For each use, they were asked to select three National Consensus Project domains with measures appropriate to this use.
Web-based survey responses were transmitted directly to staff at the NHPCO national office, where data were deidentified and transmitted to The Carolinas Center for Medical Excellence. All individual items were reported using simple frequency distributions. Likert-scaled responses were grouped into three categories—“rarely or never,” “sometimes,” or “always or often.” Chi-square was used to test for trend in correlations between items measuring current practices and activities in quality improvement or research, and hospice organization characteristics of census size, multiple vs. single location, presence of a change leader, urban/rural location, and tax status. Respondents were dropped from this analysis if they failed to answer items reporting practices.
Results
The web survey was received and opened by 772 unique hospice representatives, and 684 surveys were completed. Of these surveys, 32 were duplications from the same organization; the second respondent was excluded, leaving 652 analyzable surveys. Surveys were completed by 84% of those who opened the weblink, and completed surveys provide data from 29% of NHPCO's 2279 member organizations with a valid e-mail connection. Compared with data on all CMS Medicare-certified hospices (n
=3288), responding organizations were more often nonprofit (Table 1). Half of the responding organizations served a mixed urban/rural geographic area, and 39% reported an average daily census of more than 100 patients. Fifty-two percent of survey respondents reported that their hospice had signed on as an NHPCO Quality Partner; this is significantly higher than the 8% of NHPCO member organizations that had signed on at the time of the survey, which may indicate greater investment in training in quality measurement and quality improvement methods.
Table 1. Characteristics of Hospice Providers Participating in Survey as Compared with CMS Medicare-Certified Hospices
| Characteristics | Respondents (n=652) | CMS Provider of Services (n=3288) |
|---|---|---|
| % (n) | % | |
| Agency type | ||
| Freestanding | 56 (365) | 63 |
| Hospital based | 18 (116) | 17 |
| Home health agency hospice | 17 (113) | 19 |
| Nursing home based | 1 (8) | 1 |
| Palliative care program | 1 (9) | |
| Other | 6 (36) | |
| No data | 1 (5) | |
| Multiple locations | 47 (304) | |
| Tax status | ||
| Nonprofit | 74 (483) | 38 |
| Proprietary | 21 (137) | 50 |
| Government | 2 (13) | 6 |
| No data | 3 (19) | 6 |
Survey respondents were most often lead administrators, such as executive directors or chief executive officers (28%). Other common respondent roles were quality improvement coordinators (20%) or directors of nursing (19%). Fifty-eight percent reported having a full-time change leader or other individual responsible for quality measurement and quality improvement within their organization.
Hospices' Current Practices to Measure Quality and Improve Practices
Among responding hospices, there was varied adoption of activities related to quality measurement, quality improvement, and research (Fig. 1). Most responding hospice organizations involve clinical providers in decisions about changes in care practices (69%), participate in quality improvement projects (64%), and support clinicians' participation in meetings to learn about best practices (60%). Responding hospices were least likely to engage in activities related to research, including discussing research findings at meetings (39%) or participating in research projects (16%).
Fig. 1
Current practices for quality measurement and research.
Hospices' Current Preparation and Practices for Data Collection
Hospices currently collect some data of potential use in research or quality measurement (Fig. 2). They most commonly collected data about staff, such as certification and training (76%), and often assessed the patient experience using family surveys (70%). Respondents indicated that most data were collected on paper rather than electronically, limiting data reporting and analysis. Staff data were usually summarized and reported on an annual basis, and other types of data were typically reported quarterly.
Fig. 2
Types of hospice data collected.
Domains of Care for Which Hospices Collect Data
Respondents were also asked to specify domains of care about which they currently collect data (Fig. 3). Respondents most often collected data on satisfaction with care (72%) and least often collected data on cultural care (31%). In every domain, except physical symptoms, family survey was the most common and patient interview the least common form of data collection.
Fig. 3
Percent of hospices monitoring domains of care.
Recommendations for Use of Hospice Data
Most respondents recommended measuring care for physical symptoms and satisfaction with care in quality improvement, public reporting, and pay for performance (Fig. 4). Care for physical symptoms was the only domain endorsed by more than half of the respondents for use in quality improvement, public reporting, or pay for performance. All other domains were endorsed by less than half of the survey respondents for any of these uses. Research was recommended over these applied clinical uses for the domains of psychological care, spiritual care, cultural needs, ethical and legal issues, and quality of life.
Fig. 4
Recommended uses for quality measurement by domain.
Hospice Characteristics Associated With Quality Improvement and Research
We examined associations between hospice organizational characteristics and respondents' statements regarding preparation and practices in quality improvement and research (Table 2, Table 3). Smaller hospice organizations and those with a single rather than multiple locations reported lesser capacity for activities related to research and quality improvement. Government-owned hospices reported higher rates of quality improvement training and experience. The presence of a designated change leader within the organization was consistently and strongly associated with greater preparation for research and quality improvement.
Table 2. Hospice Characteristics Associated With Research Practices
| Characteristics | Always or Often Discuss Research (%) | Always or Often Change Based on Research (%) | Always or Often Participate in Research (%) |
|---|---|---|---|
| Average daily census | |||
| Less than 50 | 44 | 40a | 12a |
| 50–200 | 44 | 54 | 17 |
| Greater than 200 | 49 | 51 | 32 |
| Multiple locations | |||
| Yes | 50a | 53a | 26a |
| No | 41 | 44 | 11 |
| Change leader in agency | |||
| Yes | 51a | 55a | 23a |
| No | 35 | 37 | 10 |
| Location | |||
| Primarily urban | 43 | 46 | 15a |
| Primarily rural | 46 | 46 | 13 |
| Mixed urban and rural | 45 | 50 | 22 |
| Tax status | |||
| For-profit | 58a | 56 | 19 |
| Voluntary/nonprofit | 42 | 47 | 18 |
| Government | 44 | 40 | 20 |
aP |
Table 3. Hospice Characteristics Associated With Quality Improvement (QI) Practices
| Characteristics | Always or Often Use a Test of Change (%) | Always or Often Train Clinicians in QI (%) | Always or Often Participate in QI Projects (%) |
|---|---|---|---|
| Average daily census | |||
| Less than 50 | 41a | 59 | 69 |
| 50–200 | 55 | 57 | 75 |
| Greater than 200 | 60 | 63 | 80 |
| Multiple locations | |||
| Yes | 58a | 63 | 77 |
| No | 45 | 55 | 70 |
| Change leader in agency | |||
| Yes | 58a | 68a | 83a |
| No | 40 | 44 | 59 |
| Location | |||
| Primarily urban | 50 | 55 | 69 |
| Primarily rural | 46 | 61 | 72 |
| Mixed urban and rural | 51 | 59 | 77 |
| Tax status | |||
| For-profit | 46 | 66a | 68 |
| Voluntary/nonprofit | 52 | 67 | 75 |
| Government | 50 | 80 | 100 |
aP |
Discussion
Expanding demand for hospice services, combined with internal and external interest in demonstrating and advancing high-quality care, has challenged hospice organizations to measure quality of care and to translate research evidence into improved practices. In this nationwide survey of NHPCO member hospices, experience with quality improvement exceeds research-related activities. Although most of the respondents report collecting data about staff training and family perspective on care quality, fewer hospices collect clinical or patient survey data. Large, multisite hospices with a defined leader for quality-of-care initiatives reported more preparation and practices related to quality improvement and research. Half of the hospice organizations responding to the survey also had joined NHPCO Quality Partners, suggesting that this group overrepresents early adopters of quality improvement methods. It is likely that respondents in this survey provide an overestimate of the preparation of hospice organizations for quality improvement and research. Smaller, single site hospices may need technical assistance to adopt these new practices.
Among domains of hospice care, respondents recommended that current quality measurement and quality improvement efforts should examine care for physical needs and family satisfaction. In several published studies in nursing homes, hospitals, and hospices, quality improvement has been effective at changing pain management practices.6, 7, 8, 9, 10 Among survey respondents, the most common form of current data collection is family surveys after death—likely attributable to dissemination of the Family Evaluation of Hospice Care (FEHC) by NHPCO. Less is known about the potential to improve satisfaction outcomes through quality improvement efforts. Satisfaction measures tend to be positively skewed, although FEHC data provide evidence of good distribution in many responses. Unlike chart abstraction, surveys depend on the voluntary participation of both hospice organizations and the recently bereaved family members they contact, and low response rates may distort quality measures.11
Respondents considered several domains of hospice care quality to be priorities for further research, notably, care for psychological, spiritual, communication, and cultural needs; care to improve quality of life; and care during active dying. In these domains, hospice providers may perceive limitations in the science required to define best practices or in the availability of practical measurement approaches. Until more research links specific care practices to improved outcomes in these domains, hospice and palliative care providers may be uncertain about what to measure in quality improvement. Despite acknowledging the need for research, few respondents indicated that their organization had participated in a research project. Research capacity in hospice and palliative care is necessary, both to generate scientific evidence for best practices and enable translation into clinical care.12
Hospice organizations are now required to engage in quality assessment and data-driven performance improvement. Measures derived from staff certification and training data and from family after-death surveys are most often available to adapt to this purpose, but research is needed to permit quality measurement in other aspects of hospice care. Hospice organizations currently make limited use of electronic data analysis and reporting.
The results of this survey should be interpreted cautiously, with attention to the nongeneralizability of the respondent group. Study respondents were more likely than nonrespondents to be nonprofit organizations participating in the Quality Partners initiative. These characteristics suggest that respondents may have the resources of a larger organization and may be early adopters of the preparation and practices surveyed. Further, their responses may have been increased by social desirability bias. Taken together, these limitations suggest that survey results are a likely overestimate of the true hospice preparation and practices in quality improvement and research.
Despite these limitations, this survey has implications for hospice practice, policy, and research. In response to new regulations, hospice quality improvement may begin by focusing on care for physical symptoms or by making use of existing data on staff training or family satisfaction. Even among the responding hospice organizations, use of electronic data systems and chart abstraction on care processes was relatively uncommon. Investment in systems to promote reliable, valid, and easily analyzable data collection is needed to promote future quality improvement or research in hospice. Hospices will need to adapt their staffing and organizational culture, developing strategies to learn about, adopt, and monitor best practices.
Hospice organizations have little current experience in research participation or in use of research data to examine and improve clinical practice. This limited current experience contrasts with their expressed need for research in key domains of care. Some domains of care may lack the evidence base to define best practices. Hospice and palliative care research should expand to examine physical symptoms other than pain, and care to meet psychological needs, spiritual needs, cultural needs, or quality of life at the end of life.
References
- Reforming care for persons near the end of life: the promise of quality improvement. Ann Intern Med. 2002;137:117–122
- The national consensus project for quality palliative care. Available from http://www.nationalconsensusproject.orgAccessed January 30, 2009
- . A national framework and preferred practices for palliative and hospice care quality: A consensus report. Washington, DC: NQF; 2006;
- . Available from www.nationalprioritiespartnership.orgAccessed January 30, 2009
- Federal Register Vol. 73, No. 109. DHHS, DMMS 42 CFR Part 418.58: Medicare and Medicaid programs: hospice conditions of participation final rule. Thursday, June 5, 2008. Available from: http://www.access.gpo.gov/su_docs/fedreg/a080605c.html. Accessed December 5, 2009.
- . Using quality improvement to address pain management practices in nursing homes. J Pain Symptom Manage. 2005;30:271–277
- . A quality improvement intervention to increase palliative care in nursing homes. J Palliat Med. 2005;8:576–584
- Ameliorating pain in nursing homes: a collaborative quality improvement project. J Am Geriatr Soc. 2004;54:1988–1995
- . Institutionalizing effective pain management practices: practice change programs to improve the quality of pain management in small health care organizations. J Pain Symptom Manage. 2006;31:248–261
- . Implementation and evaluation of a quality improvement process to improve pain management in a hospice setting. Am J Hosp Palliat Care. 1998;15:209–216
- . Bereaved family members' evaluation of hospice care: what factors influence overall satisfaction with service?. J Pain Symptom Manage. 2008;235:365–371
- . Palliative care networks, a requirement for evidence-based palliative care: time for coordinated action. J Palliat Med. 2007;10:845–850
PII: S0885-3924(09)01126-9
doi:10.1016/j.jpainsymman.2009.09.003
© 2010 U.S. Cancer Pain Relief Committee. Published by Elsevier Inc. All rights reserved.
