Geriatric Grand Rounds Visit web sites: Tuesday, May 26, 2009 12:00 noon Dr. Bill Black Auditorium Glenrose Rehabilitation Hospital In keeping with Glenrose Rehabilitation Hospital policy, speakers participating in this event have been asked to disclose to the audience any involvement with industry or other organizations that may potentially influence the presentation of the educational material. Disclosure will be done both verbally and using a slide or handout. for handouts, poster, schedule, subscription: http://www.ualberta.ca/~geriatri/ggr/. for on-demand archive of previous presentations: http://www.beamtelehealth.ca Conducting research in long term care settings in Alberta: The potential of RAI data Anne Sales Faculty of Nursing University of Alberta anne.sales@ualberta.ca Disclosure I have no relationship that could be perceived as placing me in a real or apparent conflict of interest in the context of this presentation. Thanks to: Leadership and staff at four long term care facilities for their enthusiastic participation Knowledge Brokering Group, CHSRF and AHFMR for funding and support
Overview Resident Assessment Instruments Overview Pilot work to develop feedback reports using RAI-MDS 2.0 data Methods and findings Current project: DICE (Data for Improvement and Clinical Excellence) What we're doing What we're finding The Resident Assessment Instruments (RAI) The RAI tools are developed by the interrai group http://www.interrai.org The RAI instruments are a suite of assessment instruments for use in chronic care across settings The Minimum Data Set (MDS) was the first one developed for use in nursing homes Version in use in Canada is RAI-MDS 2.0 Most current version is interrai-ltc, not yet being used in Canada Other instruments include RAI-HC for home care (mandated in Alberta) interrai-al (Assisted Living) interrai-mh (Mental Health focuses on populations requiring chronic mental health services both in community and institutions) interrai-ca (Contact Assessment for intake across the long term care spectrum) interrai-ac (Acute Care for use with frail older people in hospital settings) And others Standardized data with common data elements to permit following people over time and over settings Initial impetus came from the OBRA 87 legislation in the United States Goal of improving care in nursing homes through standardized assessments, documentation, and care planning Fully mandated in all US nursing homes that receive federal funding Public reporting of key quality indicators through a public web site (www.medicare.gov/nhcompare) Coupled with other requirements including minimum staffing and other measures Suite of instruments has been implemented and/or tested across several countries and jurisdictions Creates a standard data stream which permits relatively easy access to a lot of information about residents and some information about facilities or settings Assessments cover many areas Functional assessment Activities of daily living Gait, stability, mobility Psychosocial assessment Engagement with environment Depression screen Activities Cognitive assessment Cognitive Performance Scale Pain, health conditions, medications A total of over 400 possible domains Many common elements across assessment instruments
Scales and measures derived from data A number of validated scales and outcome measures have been derived from RAI data Cognitive Performance Scale Validated against MMSE and other cognitive assessment tools Pain Scale Validated against other pain measures including Visual Analog Scale Depression Rating Scale Depression screener (not diagnostic) that has been validated against other screening tools Other scales including quality of life, risk of hospitalization or other major episode, screening for risk of institutionalization for home care, etc. How it works Assessments are done at admission/intake and periodically after In LTC facilities (RAI-MDS 2.0), assessments are annual, quarterly, and after major status change In home care (RAI-HC), assessments are annual and after major status change Other intervals depending on setting Assessments are intended to be performed by trained professional staff RNs, rehabilitation professionals, dieticians, social workers, others depending on setting Requires significant training to obtain standard data Most settings input data directly into vendor provided software Can be rolled up to regional, provincial, and ultimately federal databases Canadian Institute for Health Information (CIHI) has Continuing Care Reporting System and Home Care Reporting System; Mental Health Care Reporting System is in development Uses of the data Care planning at the individual resident/client level Triggers built into software products that provide client assessment protocols (CAPs) when risk is detected Falls, pain, hydration, polypharmacy, psychotropic medications, others Assessing quality of care at the unit and facility level Set of quality indicators (QIs) have been developed and are reported on Incontinence without a toileting plan; change in cognitive status; depression; falls risk Comparing across facilities in terms of processes of care and outcomes Determining payment Resource Utilization Groups, version 3 Based on care requirements Used in the United States and other jurisdictions Not yet used in Canada My current research using RAI data Feedback reports for quality improvement in long term care and home care
Research objectives To understand how different groups of providers, including non-professional staff, in long term care settings prioritize different domains for quality improvement To develop feedback reports that will be effective in improving quality of care among residents in facility-based long term care settings To use these feedback reports over a 12 month period to assess impact and uptake Two phases in our pilot work Phase 1: Elicit priority ratings from front line staff in long term care settings Data collection completed September 2007 Initial analysis completed November 2007 Phase 2: Distribute prototype feedback reports and get input Data collection March 2008 Analysis completed June 2008 Phase 1: Survey to elicit priorities Developed 34 items describing domains of data available from RAI-MDS 2.0 Constructed survey asking front line staff in two facilities to rate each item on a scale of 1 (not important) to 7 (very important) Asked them to consider importance in terms of all the residents on their unit as a group, not individually Asked them to consider importance for improving the care they give residents on their unit Example (full survey handout) I am going to read you a list of different kinds of information about residents on your unit and ask you to tell me if you think this information is important for you to know in order to give the best care possible to residents on your unit: How many residents have problems with their ability to think or remember problems with being able to remember things that happened recently or happened a long time ago. Not important Very important (circle number) 1 2 3 4 5 6 7 Any comments? Please specify:
We had 161 participants Evenly split between NH1 (82) and NH2 (79) By provider group 15 RNs and Care Managers 16 LPNs 71 Health Care Aides 22 Allied Health Professionals 36 Other Included housekeepers, maintenance staff, food service, hairdresser, unit clerks, laundry service Very experienced group overall Worked on same unit average of 6 years Overall, top 3 domains were in 2 major groups 1 Mood and Behaviour Patterns 2 Physical Functioning and Structural Problems 3 Physical Functioning and Structural Problems Are sad, depressed, or anxious 6.27 Need help to avoid falling 6.26 Need help transferring 6.24 Some things to note Different provider groups had different priorities But across five groups, there were commonalities among the top three priorities Most of the top-rated domains were in the Physical Functioning category Assistance with activities of daily living Psychosocial Depression or anxiety Health Conditions Pain Phase 2: Feedback reports Developed feedback reports Understandable convey information easy and quick to read One for each different provider group Total of five different types Based on the priorities expressed by each group
Understanding uptake and response Distributed feedback reports to staff Ask them about their reactions to the reports Did they understand them? Do they think that the information would cause them to change how they treat residents or what they do during their shift? Did they discuss the reports with anyone else? Ask questions based on the Theory of Planned Behavior http://www.iceberg-grebeci.ohri.ca/ Phase 2: Findings 183 staff responded (102 NH1, 81 NH2) 16 RNs/Care Managers 20 LPNs 84 HCAs 21 Allied Health Providers 41 Other Only 12 reported not receiving feedback report the week before (they received it that day before completing the survey) Most people read the reports and said they found them useful Everyone reported reading and understanding the report Usefulness 2% overall found it not at all useful 33% found it somewhat useful 49% found it useful 16% found it very useful 56% RNs/CM 55% LPNs 45% HCAs 43% AHCs 54% Other
Most people wanted more information Yes: 75% Information about reasons for why things are the way they are in the report: 26% Information about best practices for specific kinds of care: 43% Information about how other units have addressed these issues: 39% Information about whether things are changing over time: 41% Information about best practices for specific kinds of care was endorsed as most valuable additional information by 41% of those responding to query (N=54) Beyond the pilot DICE Test the feedback reports in a larger sample across more facilities DICE: Funded by CHSRF and AHFMR Trial in 4 facilities over 12 months beginning January 2009 Extend to home care in 2010 Probably use home care offices in Edmonton, Calgary and perhaps Lethbridge Social networks and their impact on uptake Funded by CIHR Will be done as part of DICE Planning early fall addition to current DICE project Not planning to discuss today Data for Improvement and Clinical Excellence Conducted surveys in the four facilities to assess context during four month period before January 2009 Began distributing feedback reports in four facilities (9 units) on a monthly basis January 2009 Obtain MDS data every month 4 month time lag In February 2009 we can report October 2008 data Issues of complete assessments, when a period closes, other organizational issues Feedback reports focus on three issues Pain Uses Pain Scale Depression Uses Depression Rating Scale Falls Risk Uses items from Sections G and J on the MDS No validated falls risk scale in MDS Working on this at the moment We also report falls rate over the last 31-180 days on each unit
Findings from first month 123 staff responded (18 NH1, 27 NH2, 32 NH3, 9 NH4, 28 non-unit based) 17 RNs/Care Managers 14 LPNs 56 HCAs 15 Allied Health Providers 21 Other Only 14 reported not receiving feedback report the week before (they received it that day before completing the survey) Most people read the reports and said they found them useful Most reported reading and understanding the report (94% read it, 91% said they understood it) Usefulness 5% overall found it not at all useful 40% found it somewhat useful 42% found it useful 6% found it very useful About half discussed the report with another staff person 23% wanted to see what the other person thought about the report 18% wanted advice about the report 24% wanted to know if the other person had ideas about how to improve care Most people wanted more information Yes: 64% Information about reasons for why things are the way they are in the report: 20% Information about best practices for specific kinds of care: 40% Information about how other units have addressed these issues: 28% Information about whether things are changing over time: 25% Information about best practices for specific kinds of care was most often cited (N=49)
How might you change care? Some variability in how many residents they expect to assess for pain 63% said the report gave them information they would use to change care 27% said they might change how they assess residents 33% might change how they assist residents with ADLs 20% might change residents' daily schedule 32% might change residents' activities 23% might change policies 10% specified a different kind of change 40 35 30 25 20 15 Number of residents out of next 10 1-10 10 5 0 1 2 3 4 5 6 7 8 9 10 Ongoing work Have just completed fifth feedback cycle (out of 13) Just beginning to get data corresponding to the beginning of the feedback intervention period in January Will complete 13 cycles of feedback reports in long term care Begin feedback reports in home care settings in 2010 At least two home care offices (professional staff) At least one Designated Assisted Living facility (professional and non-professional staff) Compare uptake of feedback reports, intent to change behavior, and resident outcomes measured through RAI data Significant care and research opportunities Alberta Health and Wellness has several initiatives aimed at implementing and using RAI data throughout long term care and home care settings in the province Best practice leadership and support Consultation and advice to facilities and providers about conducting assessments, using the data for improving quality of care to individuals and groups of residents Aligning Continuing Care Standards with RAI data sources Continuing Care Desktop As implementation proceeds and data begins to become standard and widespread, enormous opportunities for research