Final Research Report
Submitted to the Resident Staff Ratio Committee on June 25, 2001
The Resident Staff Ratio Committee is a committee of the Department of Health and CUPE with representation from local unions, employer representatives and NSAHO. This committee’s purpose is to examine workable resident-staff ratios for nursing homes in Nova Scotia. Researchers from CUPE and NSAHO were requested to collect objective data as a foundation for committee deliberations. This final research report represents the research only and should not be interpreted as the report of the Resident Staff Ratio.
- Joseph Courtney, Canadian Union of Public Employees
- Dave Kerr, Nova Scotia Association of Health Organizations
- Lesley Buchanan Larrea, Nova Scotia Association of Health Organizations
Table of Contents
- Executive Summary
- Section I: The Context
- The fundamental questions
- Report methodology
- Section II: Research Findings Nova Scotia comparisons to Canada
- Other available information/data
- Survey says
- Section III: Other Considerations RAI 2.0
- Influencing variables
- Research recommendations
- Appendix A: Cross-Canada survey
- Appendix B: Research contributors
- Appendix C: Tables
- Executive Summary
- a review of the relevant literature on the development of resident/staff ratios in nursing homes;
- comparisons between Nova Scotia and the rest of the Canada in terms of the LTC sector;
- details of a cross-Canada survey on nursing home staffing standards; and
- next steps in the research process and a summary of key recommendations.
Our review of the literature has revealed a lack of empirical research on the development of LTC resident/staff ratios. Furthermore, the research that does exist tends to be anecdotal. Health care researchers are aware of this conundrum. Indeed, efforts are now underway in Canada and the United States to address this dilemma.
Hence, the Task Force’s efforts to develop recommendations for the implementation of LTC resident-staff ratios in Nova Scotia have been frustrated by the paucity of research to date. In response to this dilemma, in Phase Two of the project the Task Force will follow research projects underway in Canada and the United States on the establishment of staffing ratios in nursing homes. The Task Force will also conduct a survey of all seventy (70) nursing homes in Nova Scotia. The objective of the survey is to ascertain current staffing levels, and staff roles and responsibilities, among other factors.
Overall, research indicates that a single resident-staff ratio cannot be universally applied to all care settings. Several factors including the existence of support and other professional staff, and the experience and education of staff, contribute to the determination of appropriate staffing levels to support the delivery of quality client care. Further comprehensive study of the Nova Scotia environment will be appropriate to determine appropriate staffing guidelines.
SECTION I: THE CONTEXT
The research project was initially guided by the following key questions:
- What are the current resident staff ratios for long term care in Nova Scotia and across the country?
- How does Nova Scotia compare with respect to other provinces?
- Are there standards/legislation that determine ratios in other provincial jurisdictions? If yes, then how were the ratios established? What are the criteria? What are the various elements that determine ratio levels?
- What information/data can be derived from other jurisdictions?
A questionnaire titled “Canadian Review: Nursing Home Direct Care Staffing Standards” (see Appendix A) was developed by CUPE and NSAHO researchers and distributed to government officials across Canada. The following is a sample of the survey questions:
- Does your province use various levels of care when classifying long-term care (LTC) residents?
- What are your current provincial guidelines for provision of nursing administration in LTC? For example: 1 FTE / 60 beds
- Does your province currently require 24 hours RN coverage in nursing homes?
- Does legislation in your province allow LPNs to dispense medications?
- Does your provincial legislation for LTC mandate the provision of non-direct care services in nursing homes? For example: Nursing, OT/Physiotherapy, Recreation Therapy, etc. If so, please provide the FTE/Resident ratio specified by your legislation.
- Have universal LTC resident-staff ratios been proposed or implemented in your province?
- Are there standards or legislative frameworks governing LTC resident-staff ratios in your province?
The comprehensive nature of the questionnaire underscores the complexity of developing a potential formula for workable resident-staff ratios in LTC facilities in Nova Scotia.
Research materials gathered to date have come from a variety of sources including Internet databases and web sites, and telephone conversations with and email requests to stakeholders and research experts in health care in Canada. Materials have been supplied by a number of individuals and organizations (see Appendix B)
There exists many interpretations of what research is and what it is not, just as there exists the belief that “good” research is by definition objective and is therefore divorced from the opinions, beliefs and value judgments of those involved in the production of knowledge. There are two basic types of research: primary and secondary. Primary research takes as its object the production of new knowledge through, for example, surveys and other types of research designs (e.g., participant observation and semi-structured interviews). Alternatively, secondary research involves the systematic collection and analysis of existing sources of information such as that contained in literature reviews, academic journals, newspapers, magazines, books, surveys, data bases, etc.
This project employs the techniques of secondary research. It is important to realize that the research enterprise is inherently biased. The research process can never be absolutely objective because researchers, as human beings, are not objective. Researchers approach their work with a particular agenda, assumptions and biases. This is why researchers scrutinize each other’s work. “Perfect research”, if it existed, would require no such treatment.
With respect to this report, there is a scarcity of empirical research in the area of LTC resident-staff ratios in Canada; for example, there exists no national database of LPN information. Existing evidence is descriptive and anecdotal as opposed to empirical. Much more rigorous research is required in this area. Indeed, health care researchers are aware of this conundrum. There is growing interest by researchers to tackle this dilemma with a view to filling this gap in the health care literature.
Trends in the data
Tables one through six provide a historical review of RN and LPN volumes and employment patterns between 1982 and 1998. The following trends are worth noting:
- While showing a marked increase during the 1980s, the number of RNs in Nova Scotia has steadily declined between 1992 and 1998. By contrast, other Atlantic provinces have shown an increase in RN volumes over this period.
- Only New Brunswick and Newfoundland showed an increase in LPN staffing between 1992 and 1997. Nova Scotia experienced a 3.0 percent decline; however, this rate was among the lowest in Canada.
- The RN to population ratio in Nova Scotia has fallen to 112:1 in 1998 from 101:1 in 1992. All other provinces except New Brunswick and Newfoundland have shown similar declines.
- Between 1987 and 1997 all provinces except New Brunswick, Prince Edward Island and Newfoundland experienced declines in their LPN to population ratios.
- In all of Canada since 1985, nearly 22,000 RNs have begun working in nursing home or community care settings, while over 3,000 have left the acute care environment.
In 1997, Nova Scotia ranked 7th in its RN to LPN ratio at 2.7:1 - below the Canadian average of 3.0:1.
Despite the current dearth of solid research on resident-staff ratios in Canada, efforts are underway to remedy this dilemma.
Advisory Committee on Health Human Resources
The Advisory Committee on Health Human Resources (ACHHR), Nursing Strategy for Canada, (October 2000), points to the need for more empirical research on RNs, LPNs and Registered Psychiatric Nurses (RPNs). To this end, the report calls on the federal, provincial and territorial governments to establish a Canadian Nursing Advisory Committee (CNAC). One of the main objectives of the CNAC will be to improve nurses’ quality of work life through improved nurse/patient ratios to address workload concerns; reduction in non-nursing duties; and reduced “casualization” (ACHHR, 2000: ES 3). Specifically, the ACHHR has recommended the following strategy as a means to improving nurses’ quality of work life with a view to enhancing nursing retention:
- Address appropriate nurse/patient ratios;
- Utilize an efficient and appropriate nurse mix;
- Reduce non-nursing duties;
- Prevent workplace injuries and illness;
- Reduce casualization and increase permanent positions;
- Implement improved flexibility/family-friendly scheduling options and customized work arrangements;
- Reintroduce/enhance clinical leadership at the bed/ward/unit level; and
- Ensure appropriate opportunities for continuing education and practice development (ACHHR, 2000: 31-32.).
Human Resources Development Canada
Human Resources Development Canada (HRDC) is currently examining human resource planning in nursing. The first phase of the report consists of a literature review by Dussault et al., 1999. The second phase of this project consists of a national nursing sector study, which is now underway.
United States of America
American empirical research exists that establishes a direct link between low nurse staffing levels (e.g., RN’s, LPNs and Nurse Aides/Nursing Assistants) and inadequate and even harmful resident outcomes. The U.S. Health Care Financing Administration’s (HCFA) recent report to the U.S. Congress is a prime example. The 800 page study was mandated by law and published in the summer of 2000. The report examines “the analytic justification for establishing minimum nurse staffing ratios in nursing homes” (HCFA, 2000: ES 1). Essentially, the research “establishes a clear and irrefutable link between low staffing levels and poor health outcomes for residents including avoidable hospitalizations, a high incidence of pressure sores, and weight loss” (HEU, 2001: 1; see also HCFA, 2000, ES 3). The authors of the report also determined that adequate staffing levels are:
- important for the provision of adequate care levels necessary to avoid serious harm to residents
- important for both improving the health outcomes and quality of life for residents; and
- cost effective in human and financial terms, especially by reducing costs associated with certain acute care expenditures and by lowering the rate of staff injuries (HEU, 2001: 1).
A complex research methodology was employed in the study and consisted of the following elements:
- Consultations with experts in long-term care, nursing economists, stakeholders, consumer advocates, nursing home industry officials and labour organizations with a view to reviewing the literature on staffing;
- Empirical determination employing multivariate analysis
- Time-motion studies were employed to determine the amount of time required to perform certain tasks (e.g., changing wet clothing, toileting, exercise, feeding, morning care, etc.). A simulation analysis was adopted using six categories of residents with different functional limitations and care needs.
Major Findings of the HCFA Report:
- The multivariate analysis and time motion studies indicate a strong relationship between staffing and quality of care.
- Minimum staffing levels may reduce the likelihood of quality of care problems; however, higher “preferred minimum” levels exist above which quality was improved across the board.
- The minimum staffing level associated with reducing the likelihood of quality problems is approximately 2.0 hours per resident day for nurse aides, regardless of facility case mix.
- The preferred minimum staffing levels for RN and total licensed staff in which quality was improved across the board are .45 and 1.0 hours per resident day, respectively.
- Using a time-motion derived standard, the minimal nurse aide time necessary to provide optimal care in delivering five specific daily care processes is 2.9 hours per resident day. The five daily care services include repositioning and changing wet clothes, repositioning and toileting, exercise encouragement/assistance, feeding assistance, and ADL independence enhancement (morning care).
Report to the US Congress: HCFA Study, July 2000
Direct Care Hours Per Resident Per 24 Hour Day
Source: Adapted from Hospital Employees’ Union (HEU), British Columbia (February 2001). Backgrounder: Staffing in BC’s Long-Term Care Facilities Fall below Minimum Levels set Out in US Congress Study. (p. 6).
Canadian Reaction to HCFA Report:
Upon reviewing the Report to Congress study, researchers with the Hospital Employees’ Union (HEU) of British Columbia concluded that current staffing levels in BC’s long term care facilities jeopardizes residents’ health and safety. The unions that represent 60 thousand health services and support workers in BC “are proposing contract language that would establish staffing ratios that are safe for residents and staff. Under the unions’ proposal, a joint union/employer committee would set staffing ratios for long-term, multi-level and extended care facilities for full implementation within one year, followed by staffing ratios for home support workers in the second year” (HEU, February 2001: 7). The proposal will be on the table during this year’s round of collective bargaining.
Seven of nine provinces contacted responded to the cross Canada survey on nursing home staffing standards developed by researchers with NSAHO and CUPE (see Appendix A). The survey provided the following points of comparison between Nova Scotia and the rest of Canada (see tables 7, 8a and 8b in Appendix C):
- Nova Scotia does not have formal care levels which equated to the Level I and IV as described in the survey.
- Care provided at Level II averaged about 1.8 hours/resident. Nova Scotia: approximately 1-1.5 hours.
- Care provided at Level III averaged about 2.6 hours/resident. Nova Scotia: approximately 2-2.5 hours.
- Four of seven provinces have formalized guidelines for nursing administration, whereas, Nova Scotia does not.
- Nova Scotia has a 24 hours RN coverage requirements as does most all other provinces.
- LPNs in Nova Scotia are permitted to dispense medications as is the case with most other provinces.
- Only Ontario and Newfoundland have the provision of other health service providers mandated in LTC legislation.
- Only New Brunswick claims to have universally implemented resident-staff ratio guidelines.
- All provinces except Newfoundland identified “recruiting” and “retention” among the top three issues facing LTC.
(This section has been contributed by Wade Were, Senior Policy Analyst, Nova Scotia Department of Health.)
Minimum Data Set 2.0
The RAI 2.0 was designed to be first and foremost a care planning tool. Central to the RAI 2.0 is the Minimum Data Set (MDS 2.0) which is a standardized data collection form used to collect the minimum amount of information needed to plan the resident’s care. A Registered Nurse coordinates the completion of the form drawing information from the residents records, the resident and/or family, and through communication with other members of the residents care team such as the physician, pharmacist, rehabilitation therapists, activity coordinators, dietary staff, LPNs, and Personal Care Workers. The MDS 2.0 provides a database of coded information that indicates the resident’s level of function or care needs on each assessment item.
In addition to the development of individualized care plans, the MDS 2.0 data supports other applications including case-mix classification, quality indicator reporting, and outcome measurement.
In Canada, the Canadian Institute for Health Information has endorsed the use of the MDS 2.0 data set as the national standard for long-term care settings.
Ontario has mandated the use of the MDS 2.0 since 1996 for its chronic care hospitals. Since 1997 Saskatchewan has been phasing in the implementation of the MDS 2.0 and requires all Health Districts to begin reporting commencing April 1, 2001. Provincial Governments in Manitoba, British Columbia, Alberta, and Nova Scotia have completed pilot testing of the MDS 2.0 or are in the process of conducting tests.
Resident Assessment Protocols
Upon feeding the MDS 2.0 assessment data into a software program, certain resident characteristics will trigger the need for further assessment and care planning guided by the Resident Assessment Protocols (RAPs). Using one or more of these practise guidelines, the caregivers put together an action plan to care for the resident. The RAPs do not replace the clinician’s judgement.
There are 18 RAPs, which have been created by clinical experts, and can be used for both individual care and facility wide programming activities. They cover important areas such as pressure ulcers, falls, communication, vision, cognition, delirium, incontinence, behaviour, etc.
Researchers have developed and validated 24 Quality Indicators (QI) based on the MDS data. They indicate the presence or absence of potentially poor care practices or outcomes. In addition to being useful information for facility quality improvement activities, comparative QI information can be useful to regulators for such purposes as licensing, benchmarking, and the identification of best practices.
MDS 2.0 Quality Indicators
Accidents - fractures, falls
Clinical Management - behaviour affecting others, symptoms of depression, depression with no an-depressant therapy
Clinical Management - nine or more medications
Cognitive Patterns - cognitive impairment
Elimination & Continence - bladder or bowel incontinence, incontinence without a toileting plan, indwelling catheters, fecal impaction
Infection Control - urinary tract infections
Nutrition & Eating - weight loss; tube feeding; dehydration
Physical Functioning - bedfast residents; late loss ADLs decline; rang of motion decline (no training)
Psychotropic Drug Use - anti-psychotic and no related conditions; anti-anxiety/hypnotic use ore than 2X weekly
Skin care - stage 1-4 pressure ulcers
Quality of Life - daily physical restraints, little or no physical activity
Several Outcome Measurement Scales have also been developed based on the MDS data. These scales have been validated against gold standards in the industry. They facilitate evaluation of interventions and provide evidence for best practice.
Case Mix Classification
Case mix systems use combinations of resident characteristics (often available within assessment systems) to identify groups of residents with homogeneous resource requirements. Using the MDS data one can categorize residents into Resource Utilization Groups (RUG-III). The RUG-III algorithm uses over 100 variables from the MDS 2.0 to produce 44 classification levels organized in 7 hierarchical domains. For each of the 44 classification levels, a case mix index has been calculated through extensive time studies carried out in the USA and found to be valid and reliable through international studies including Canada.
A case mix index represents the mean resources used by residents in that group relative to other groups. The time study data coupled with average salary information from nursing, rehabilitation, and auxiliary staff was used to develop the case mix indices.
Case mix provides funders with a system to equitably distribute limited resources. Facilities that care for clients with heavy care needs are provided more resources than facilities that care for clients with lighter needs.
Case mix systems are not financing systems. The case mix system provides information for the equitable distribution of resources, it does not specify the amount of funding needed in the sector. By way of analogy, case mix systems describe how the pie should be divided not how large the pie should be. [NSAHO researchers wish to note that the MDS 2.0 is not a workload measurement tool, and does not automatically provide staffing information.]
It has been consistently found across countries that while the absolute amount of care provided varies widely, the relative resource needs of different groups of residents tend to be stable across cultures even when the resources available through the financing system vary substantially.
It should be noted that the RUG-III (case mix) classification system is fundamentally different from the classification system used in Nova Scotia and many other provinces where classification is used primarily as a tool for making placement decisions rather than resource allocation decisions.
Most provinces in Canada, including NS, classify clients by assessing them prior to placement. The assessor matches the client to one of few classification levels based on their largely subjective interpretation of the assessment data.
The Use of RUG-III in North America
The US Government uses the RUG-III in their prospective payment system for Medicare patients in every state as of July 1, 1998. About 11 US state governments had begum to use a RUG III system or its derivative to reimburse facilities for Medicaid patients prior to 1998. Other US states are expected to follow.
Ontario is scheduled to begin using RUG-III for resource allocation commencing April 1, 2001 in the chronic care hospital sector. Saskatchewan Health is considering using the RUG-III data in its Health District funding methodology.
In Canada, only one other case-mix classification system exists for long-term care, i.e. the Alberta Resident Classification System. ARCS is used in the nursing home and chronic care hospital sector of Alberta as well as in the nursing home sector of Ontario. However, given Alberta appears ready to abandon the ARCS in favour of a new RUG-III based classification system, the future use of ARCS is in serious doubt.
Select Strengths of MDS 2.0 & RUG-III
- Scientifically tested to be valid & reliable.
- InterRai’s continuous development of tools.
- Refined breakdown of levels based on resource intensity.
- Can be used for case mix funding.
- Perceived to be less susceptible to manipulation by assessors.Involves direct assessment of residents so not dependent soley on the quality of charting.
- Built in incentives to provide rehabilitation.
- Primarily assesses client needs, thus, less dependent on facility practices to determine need.
- Able to explain the samller subgropus of patients who are very resource intensive.
- Associated system for care planning and qulaity improvement including outcome measurement.
(see tables 1-6 in Appendix C) The fundamendal questions Report methodology SECTION II: RESEARCH FINDINGS
Other available information/data1 of the relationship between staffing and quality of care. Outcomes examined included avoidable hospitalizations, improvements in the activities of daily living (ADLs), incidence of pressure sores, weight loss, and resident cleanliness and grooming.
SECTION III: OTHER CONSIDERATIONS
In April of 1999, the Canadian Union of Public Employees (CUPE), the Nova Scotia Department of Health (DOH) and Ocean View Manor Nursing Home reached a tentative agreement that includes contract language on the formation of a Task Force to examine workable resident-staff ratios for Nursing Homes in Nova Scotia. The verbatim text of the provision is as follows:
The Department of Health will form a task force before September 30, 1999 for the purpose of making recommendations for implementation of a workable resident-staff ratio for nursing homes in Nova Scotia. This study shall be completed by March 31, 2001. The task force will be comprised of representatives of CUPE and the Department of Health as well as employers and other unions as appropriate in the continuing care sector.
Specifically, the Task Force is comprised of representatives from various labour organizations (i.e., CUPE, Nova Scotia Nurses Union (NSNU), Nova Scotia Government Employees’ Union (NSGEU), Service Employees International Union (SEIU) and Canadian Auto Workers (CAW)), Long-Term Care (LTC) facilities, the Nova Scotia Association of Health Organizations (NSAHO) and the Nova Scotia Department of Health. The main objective of the Task Force is to develop recommendations for workable staffing ratios for nursing homes and homes for the aged in Nova Scotia. Initially, the report of the Task Force, including recommendations, was to be submitted to the Deputy Minister of Health by March 31, 2001, as outlined above; however, CUPE is taking the position that this deadline is flexible.
Researchers from the National Office of CUPE and the NSAHO collaborated on this research document. In our view, this project is very much a work in progress given the complicated nature of the research topic and the relative lack of empirical research that addresses the formulation of LTC resident-staff ratios. Hence, the research presented in this paper will not be conclusive; rather, it will give the reader a sense of the “lay of the land” with respect to the establishment of resident-staff ratios in the LTC sector.
The work of the Task Force commenced at the same time the DOH was piloting a care-planning tool in four of the province’s nursing homes. This care-planning tool is known as the Resident Assessment Instrument or RAI 2.0. The RAI 2.0 was developed by InterRAI, which is comprised of a global team of researchers and clinicians. The instrument is “designed for use in long term care facilities where skilled nursing services are employed (e.g., nursing homes, chronic care hospitals)” (Were, March 2001: 1). Use of the RAI 2.0 entails the collection of data from patients and caregivers to determine residents’ level of functioning and individual care needs. This information is then used to create care plans for residents. The data generated allows for the categorization of residents into Resource Utilization Groupings (RUGs), which is a case mix classification system:
“Case mix provides funders with a system to equitably distribute limited resources. Facilities that care for clients with heavy care needs are provided more resources than facilities that care for clients with lighter needs. Case mix systems are not financing systems. The case mix system provides information for the equitable distribution of resources, it does not specify the amount of funding needed in the sector. By way of analogy, case mix systems describe how the pie should be divided not how large the pie should be” (Were, March 2001: 3).
It is important to distinguish that the RAI 2.0 does not determine or measure “appropriate” staffing levels directly. However, it may be possible to determine staffing levels through the development or use of a bridging tool that would link a particular case mix classification (RUGs score) to the staff required to care for an individual within that classification.
A number of jurisdictions have and are giving serious consideration to implementation of the RAI 2.0. For example, in Ontario the RAI 2.0 has been mandated for use in Chronic Care facilities since 1996 and will be tied to funding in the Chronic Care sector in 2001. The instrument has been mandated for use in Saskatchewan since April 2001 and is being recommended for use in the province of Alberta. The RAI 2.0 has already been piloted in Manitoba, and in British Columbia it will be used in seven regions of the province in 2001. Officials in the provinces of New Brunswick and Newfoundland and Labrador are in the process of examining the instrument’s utility. The Canadian Centre for Policy Alternatives has endorsed the use of the RAI 2.0 in Canada. The instrument is federally legislated in the United States and Iceland and is in use in a total of seventeen countries. A more detailed examination of the RAI 2.0 is provided Section lll of the report.
This report is organized into sections as follows. Section one provides details of key research questions and the methodology employed in the project. Section two outlines data on Nova Scotia comparisons to Canada with respect to RN and LPN employment patterns. Details will also be provided on the results of a new US study and report to Congress that establishes minimum nurse staffing levels for long-term care facilities. Section three discusses other considerations in the quest to develop a resident-staff ratio. This report concludes with recommendations and next steps in the research process.