Emily G. Marshall , PhD 1 , Barry S. Clarke , MD, CCFP, COE 1 , Nirupa Varatharasan , MSc 1 , Melissa K. Andrew , MD, PhD, FRCPC 2
1 Department of Family Medicine, Dalhousie University, Halifax;
2 Division of Geriatric Medicine, Dalhousie University, Halifax NSDOI: http://dx.doi.org/10.5770/cgj.18.122
Background
Most older adults living in long-term care facilities (LTCF) are frail and have complex care needs. Holistic understanding of residents’ health status is key to providing good care. Comprehensive Geriatric Assessment (CGA) is a valid assessment method which aims to embrace complexity. Here we aimed to study a CGA that has been modified for use in long-term care (the LTC-CGA) and to investigate its acceptability and usefulness to stakeholders and users.
Methods
This mixed methods study, conducted in 10 LTCFs in Halifax, Nova Scotia, reviewed 598 resident charts from pre- and post-implementation of the LTC-CGA. Qualitative methods explored stakeholder perspectives (physicians, nurses, paramedics, administrators, residents and families) though focus groups.
Results
The LTC-CGA was present in 78% of LTCF charts in the post -implementation, period though it did not appear in acute care charts of transferred residents, despite the intention that it accompany residents between care sites. Some items had suboptimal completion rates (e.g., Advance Directives at 56.4%), though these were located in other sections of the LTCF chart (98.2%). Nevertheless, qualitative findings suggest the LTC-CGA describes a clinical baseline health status which enabled timely and informed clinical decision-making.
Conclusions
The LTC-CGA is a useful resource whose full capacity may not yet have been realized.
Key words: long-term care , geriatric assessment tool , mixed method evaluation , interdisciplinary care , comprehensive geriatric assessment
Most long-term care facility (LTCF) residents require medical care because they are frail; a majority have cognitive impairment or dementia, and many are nearing the end of their lives. As such, they have complex care needs. Comprehensive assessment of health and functional status is key to planning and providing good care for frail older adults who live in LTCF. Comprehensive Geriatric Assessment (CGA) is a validated method for assessing frail older adults in ambulatory and acute care environments which embraces this complexity.(1,2) CGA is a holistic assessment which documents an older patient’s health status, including cognition (e.g., dementia, delirium), mood, mobility, function, appetite, weight, bowel and bladder function, medical conditions, and medications.(1)
An adapted CGA, the “LTC-CGA”, has been modified and validated for use in LTCF.(3) Modifications to better suit the LTC setting include documentation of behavioural disturbances common in dementia, foot and dental care requirements, skin integrity, whether a legal next of kin has been appointed (including name), and goals of care (e.g., whether resuscitation is to be attempted or hospital transfer for acute illness). The LTC-CGA also includes a frailty measure which is a focused version of the CSHA Clinical Frailty Scale.(3–4) Since no LTCF residents would be expected to be on the low/non-frail end of the Frailty Scale (i.e., the “very fit”, “well”, “well, with treated comorbid disease”, and “apparently vulnerable” categories, all of which have no dependence for Instrumental or Basic Activities of Daily Living), these categories were not included on the LTC-CGA tool. The LTC-CGA currently exists in a one-page paper format on the resident’s chart (Figure 1).
|
|
||
|
FIGURE 1. LTC-CGA Tool |
||
As part of a larger “Care by Design” program which aimed to improve the quality of care and health outcomes for frail older residents of LTCF, the LTC-CGA was implemented in LTCF within Capital District Health Authority (CDHA) in Nova Scotia in June 2011. Family physicians caring for LTCF residents are mandated to complete the LTC-CGA for every resident, and to keep it updated every six months and after any significant change in health status. The tool is meant to have a prominent place in the LTCF chart in order to guide care in that setting, and should accompany residents who are transferred between facilities and to acute care to ensure clear communication of baseline health and function. The LTC-CGA has the potential to be a powerful tool for assessment and communication in LTCF; however, its uptake (i.e., is it completed fully, is it accompanying residents who are transferred to acute care settings?) and acceptability (i.e., is it easy to use, useful, meeting needs?) to end-users are not known.
The present study is part of a larger study of the “Care by Design” model and its impact on health and health-care outcomes (notably ambulance transfers to acute care).(2) Here we evaluate the use and benefits of the LTC-CGA using a mixed methods approach. Specifically, the objectives of this study were to evaluate the implementation of the LTC-CGA and determine its utilization and utility, with both process and outcome goals. Process goals were to determine the up-take of the LTC-CGA (completion rates and completeness), as well as acceptability to users. Outcome goals aimed to evaluate the efficacy of the LTC-CGA as a tool for improving the care for older LTC residents. We focused specifically on transfers to acute care and Emergency Departments, as one of the reasons the tool was implemented was with to have it accompany residents who were transferred to acute care, thus improving completeness and accuracy of communication between care settings. We also sought to determine whether the LTC-CGA was useful in helping to clarify goals of care at end of life.
We used a mixed methods approach including chart reviews in both LCTF and acute care for quantitative measures and focus groups for qualitative inquiry. Full details of the study methods have been published elsewhere.(2)
Ten of the 12 LTCF in the Halifax Regional Municipality participated in the study, representing a total of 1,482 beds. Two LTCFs were excluded because they represented variations on the model of primary care and are, therefore, not representative.
Here we report results from the “pre” and “post” phases of the LTC-CGA implementation. Pre-phase data were collected retrospectively for the time period September 1, 2008 to February 28, 2009; post-phase data came from the period September 1, 2011 to February 29, 2012, allowing time for the LTC-CGA to have been fully implemented.
In the post-implementation time period LTC charts were reviewed to identify completion rates for the LTC-CGA (i.e., the proportion of charts with an up-to-date LTC-CGA), as well as completeness of items within the tool. Acute care charts were reviewed to see whether the LTC-CGA accompanied residents who were transferred to the Emergency Department. Individual items were compared between the “pre” and “post” phases to give an indication of whether the item in question would have been easily locatable in the LTC chart prior to the implementation of the LTC-CGA.
Prior to initiating the chart reviews we performed a power calculation to determine the required sample size. A sample of 171 charts would be required to estimate the proportion of completed LTC-CGAs of N = 1,482 LTCF beds with a 99% confidence interval.*(5) In keeping with our overall study’s focus on residents for whom emergency (“911”) calls had been made, charts of all residents for whom a 911 call was made were included in our data collection, along with a sample of 100 non-911 involved residents for each time period. This yielded a total sample of 598 reviewed charts.
A total of 11 focus groups (FG) were conducted with key stake-holder groups: Care by Design physicians (1 FG), nurses (3 FG), administrators (1 FG), extended care paramedics (1 FG), care assistants (2 FG), residents and family members (3 FG), for a total of 75 participants. Each FG was limited to a single stakeholder group for balanced participation and to encourage open expression of perceptions and experiences. Facilitators were trained in conducting focus groups, with specific attention to research bias, professional dress, and ethics. All focus groups were conducted in private settings and were digitally audio-recorded to ensure accuracy of transcription. Participants were theoretically sampled;(6) with recruitment continuing until saturation was achieved.(7) Audio recordings were then transcribed verbatim by experienced research transcriptionists. Transcribed qualitative data were subject to rigorous data quality checks.(6) Transcriptions were entered into Atlas.ti qualitative data analysis software.(8) Data were coded using an agreed-upon coding scheme that was developed by the entire research team. Framework analysis(9,10) was conducted of narrative responses to open-ended questions about participants’ knowledge of the LTC-CGA, its acceptability, its impact on clinical care, and its usefulness in defining goals of care.
The study was approved by the Capital District Health Authority Research Ethics Committee and individual Research Ethics boards of participating LTCF, where these required.
70% of the LTC residents whose charts were reviewed were women. The mean age was 83.0 (SD 12.1) and 60% had a diagnosis of dementia (Table 1). A total of 598 residents were included in the study (both “pre” and “post” time periods), of whom 460 (77%) had emergency 911 calls and 306 (67%) were transferred to acute care at least once.
TABLE 1.
Description of resident characteristics
The LTC-CGA was present on 78% of the LTCF charts in the post-implementation phase (2011/12). For residents who had a LTC-CGA, the items most likely to be filled in were Influenza vaccination status (Yes or No; 100%), Speech (Within Normal Limits vs. impaired; 93% complete), Current Frailty Score on the Clinical Frailty Scale(5) (91% complete), functional abilities in basic Activities of Daily Living (87%–90% complete), mobility transfers and balance (90%), and cognitive status (88%). Items describing the resident’s background were frequently missing (e.g., only 28% listed occupation, 23% documented educational attainment, and 48% rated family stress). Bowel continence was noted in 85%. Presence or absence of falls was documented in 78%, though the number of falls was missing in half. Advance directives were listed on the LTC-CGA in only 56%, while resuscitation wishes were documented on 68% of LTC-CGA tools.
When items were not optimally completed in the LTCCGA, they may have been present in other areas of the chart (Table 2). This was the case with advance directives, which were present in 86 of 88 charts (98%), but were only documented on 56% (chi-square p < .0001) of the LTC-CGA forms.
TABLE 2.
Clinical data from Long Term Care Facility chart or LTC-CGA
In the comparison between the pre- and post-implementation time periods, while cognition as measured by the Mini Mental State Examination (MMSE(11)) was documented on the LTC-CGA in 73% of charts, it was possible to ascertain through detailed chart review in only 56% of the “pre” time period charts (chi-square p < .0001). Likewise, balance was also relatively under-recorded (or difficult to locate) in the “pre” charts (present in 75%), while it was documented on the LTC-CGA in 90% of cases (chi-square p < .0001). Some items were reported less frequently in the “post” time period, including marital status, which dropped from 98% to 70% (chi-square p < .0001), and bowel and bladder continence which were 97%–98% in the “pre” period and 85% in the “post” period (chi-square p < .0001).
The LTC-CGA did not appear in archived ED charts of any of the residents who had been transferred to acute care during this time period.
Three key themes emerged from the focus group data analysis: 1) Benefits of the LTC-CGA; 2) Challenges which remain with the LTC-CGA; and 3) Recommendations for the LTC-CGA. (Supporting quotes for all themes can be found on Table 3 Qualitative Findings).
Benefits included: a) that stakeholders found the LTC-CGA encourages proactive care and increased opportunities to review residents care plans and needs; b) that the LTC-CGA was a useful reference for providers who are less familiar with the resident, such as on-call physicians and nurses or clinical care aids learning about a new resident; c) that the LTC-CGA provided information continuity during transitions in care between providers and facilities.
Benefits of the LTC-CGA were identified by different stakeholder groups. Physicians and nurses described the LTC-CGA as being a useful tool to review previous care decisions, especially when questioning why care decisions were made or why residents were prescribed particular medications. Physicians found the tool assisted them in tracking systems of changes made in care, including medication adjustments, and adverse events. The LTC-CGA was found to be a useful tool for end of life care planning.
Various care team staff members stated the LTC-CGA was valuable as a reference tool when providing care to a resident who was new to them, especially on-call physicians or extended care paramedics. Nurses and clinical care aids found the tool oriented them to the needs of a new resident. The LTC-CGA was also found to provide information continuity within the LTCF and for care providers outside of the LTCF.
Challenges which remain with the LTC-CGA include a) timing and concern that the form was out of date; b) financial compensation wanted to complete the form; c) inconsistency on which health professional is responsible for completing the form; and d) the form is perceived to only be as good as the person filling it out.
The topic of financial compensation came up numerous times in the focus groups. In particular, registered nursing staff care team members expressed concern that physicians were additionally compensated for completing the LTC-CGA, though it was often a team process to complete. Physicians felt restricted in being able to bill only twice a year for completing the LTC-CGAs, causing concern as to whether the tool was up-to-date by a variety of stakeholders.
Whose role it was to complete the forms was unclear and some nurses felt it was extra work for them. Inconsistency emerged between different facilities, and even within facilities, on how much of the work was done by nurses versus physicians depending on the physician involved in the care.
Numerous recommendations emerged from the focus group conversations. Physicians discussed the importance of having a reminder system in place that would provide them with cues as to when the LTC-CGA is to be updated (i.e., upon admission, every six months and after an adverse event). Care team members also expressed concern regarding the duplication of information and wanted policies to minimize the doubling of efforts. For example, nurses stated that, when preparing for residents to be transferred to the Emergency Department, they would have to copy several other clinical summary documents along with the LTC-CGA, creating extra work for them. Nurses also indicated they had not received training on the tool or information on why it was being implemented. Care team members also expressed the importance of consistency in the completion of the tool and the tool’s location in the chart.
If the LTC-CGA tool is to be used to its full capacity, it is important that outside care providers read it and take the information provided on it into consideration when deciding on a care path for the individual. Nurses provided examples of sending the tool with a resident during a transfer to the Emergency Department and then finding the unopened envelope being sent back with the resident. It is clear that it is a tool that is valued for providing information between providers, but not necessarily in the day-to-day work of the nursing staff, or in some outside care environments.
We found that the LTC-CGA was acceptable to users, with most health-care provider stakeholders finding it to be a useful summary of a resident’s health status. Completion rates were good, as it was identified on 78% of the LTCF charts in the post-implementation period. Completeness varied by item; some items had high completeness (e.g., Influenza vaccination, ADL function, sensory impairments) and others lower completeness (e.g., education, occupation and family stress). The LTC-CGA was not located in any of the acute care charts when residents had been transferred to hospital.
Our study has certain strengths. Its mixed methods design allows us to both quantify and understand the use of the LTC-CGA. Our sample size was large. The chart review was done in both “pre” and “post” implementation time periods, allowing for comparison of documentation of important items with and without the use of the standardized LTC-CGA. Our study also has certain limitations. The LTC-CGA was implemented in June 2011, along with the fee code reimbursing physicians for its completion. While we tried to time data collection allowing for physicians to “catch up” with completing the LTC-CGA, it is possible that completion rates may have been higher had we done our chart reviews following a longer time lag. It is possible that our chart reviewers may not have been able to locate a LTC-CGA on the chart despite it having been filled out, although we tried to minimize this possibility with standardized chart review protocols. The LTC-CGA was not identified in the acute care charts when residents were transferred to the ED. It is possible that it may have accompanied residents when they were transferred to acute care but not scanned into the permanent electronic chart record. (Anecdotally, LTC-CGAs for residents transferred from LTCF have been encountered by MKA, an author on our study who works as a geriatrician seeing consults in the ED). Thus, ensuring that it is included in chart documentation would be an important area for future improvement in its use, in order to maximize its usefulness as a tool for improving health care for frail older residents of LTCF. One specific way to ensure that the LTC-CGA is included in the permanent hospital record and not discarded in the scanning process would be to ensure that it is labelled with the proper barcodes to direct its inclusion in a specific section of the acute care chart when the paper chart is sent to medical records for scanning and assembling into the permanent computerized record. As a further recommendation, moving to an electronic medical record with integration between LTCF, Emergency Health Services, Emergency Departments, and acute care settings would ensure that the usefulness of the LTC-CGA for continuity of care is optimized.
The LTC-CGA is but a single part of the larger “Care by Design” model of care.(2,12) Care by Design includes primary care reform with a dedicated primary care physician per ward or unit, guaranteed weekly family physician visits for rounds, participation in multidisciplinary care conferences, and a clear system of on-call coverage. Care by Design also includes an Extended Care Paramedic program and ongoing team education and evaluation. Clearly other factors also play a role in implementation of Best Practices in LTCF, including staffing complement and ratios and the physical layout of the facilities.
Thorough medical assessments for older adults in LTCF take time. Many family doctors balance patient care in LTCF with busy office practices, leaving limited time to provide care in LTCF. One benefit of the LTC-CGA is it streamlines the assessment process, allowing multidisciplinary teams to address all important medical and functional issues in a focused and standardized fashion. The LTC-CGA is based on the philosophy of “Comprehensive Geriatric Assessment” in which a holistic view of a patient’s medical, functional, and social issues is taken. CGA is therefore a structured approach to creating a unified view of many different parts of a clinical assessment and will, in effect, include consideration of a patient’s cognition, mood, sensory function, mobility, function in Activities of Daily Living, medical problems, medications, care plan, and social history. In this way, one might argue that it is pulling together pieces from other assessments, but we would suggest that going through a formal CGA process on a regular basis (on admission to LTCF, every six months, and as needed following significant changes in clinical status the LCT-CGA) enables comprehensive and dedicated clinical thinking to be applied to the resident’s care. This is supported by our qualitative findings (that clinicians in different disciplines find it useful), and also by quantitative findings demonstrating differences in the frequencies of completion of the various elements. Other models of completion could be also explored, including using a purpose-designed care partner CGA which would be completed by a family member or other care partner who knows the prospective resident well.(13)
Under “Care by Design”, residents do not keep their previous family physician when they move into LTCF; instead, primary care is provided by a dedicated physician per floor or unit.(2) This may explain the observed decrease in documentation of occupation, education, and historical biographical information which may have been well known to community family physicians prior to moving to LTC. An important recommendation would be to formalize a process of transfer of care from the previous community provider to the Care by Design team. Ideally the community physician would complete a “transfer of care” LTC-CGA, which would help preserve continuity of information.
As part of the implementation of the LTC-CGA, education and training on the tool were provided to physicians and staff, though as we heard in the focus groups that more education could have been helpful. The low completion of the code status and advanced directives on the LTC-CGA is a concern. A way to maximize the utility of the tool would be to provide further education on the importance of fully completing the form as a summary resource in times when informational continuity between providers is critical and timely.
The audience for the LTC-CGA includes care providers who are new to residents. The LTC-CGA provides an easy-to-find, standardized resident summary. This information is important in assessing status and change in critical situations for physicians-on-call, paramedics, consultants, acute care providers off-site, and new LTC care staff. When a critical incident occurs it can be put in perspective in relation to the baseline captured on the last date of completion of the LTC-CGA. This, in conjunction with real-time communication with the care team, can assist with a more timely assessment and development of an appropriate management plan.
We found that completion of certain items (e.g., MMSE scores and mobility status) improved using the LTC-CGA. This, along with the fact that the LTC-CGA is meant to represent the single most up-to-date summary of a resident’s health and care needs, is expected to bring benefits for clinical care planning and outcomes. From the qualitative results, clinicians commented on using the LTC-CGA as the go-to place in the chart for determining whether a resident’s observed clinical status was consistent with their prior baseline or whether the current health issue represented a new crisis or decline. This ability to assess in relation to the fuller LTC-CGA assists in decision about whether transfer to hospital would be beneficial or, in contrast, whether a more palliative focus would be appropriate. Although completion rates for Advanced Directives were suboptimal on the LTC-CGA (56.4%), these were easily located in the residents’ charts (98.2%). In this case, it is likely that referring to the standard Advance Directives (AD) form (on which these ADs are documented) in addition to the LTC-CGA (rather than expecting the LTC-CGA to represent the resident’s full AD wishes) is the way to ensure fuller representation of a resident’s Goals of Care, as they may change over time and are difficult to fit into a single tick box item on the LTC-CGA.
Benefits of comprehensive geriatric assessment include identification of issues that affect an older person’s health, provision of a framework within which to plan interventions to address these problems, and establishing a platform for discussing and setting goals of care appropriate to the older person’s level of frailty and comorbidity.(14–16) Until now, CGA has not been a consistent feature of primary care provided in Nova Scotia LTCF. The introduction of changes to the model of primary care delivery affords an important opportunity for Knowledge Translation; knowledge about how to conduct a CGA which is based on years of clinical and research experience can now be applied to a setting in which it has the potential to greatly improve care.
Funding for this study was provided by grants awarded by The Canadian Dementia Knowledge Translation Network, The Capital District Health Authority, and Dalhousie University’s Department of Family Medicine.
The authors declare that no conflicts of interest exist.
Long-term Care Comprehensive Geriatric Assessment Form (pdf available as Supplementary Material on the website)
1. Hubbard RE, Eeles EM, Rockwood MR, et al. Assessing balance and mobility to track illness and recovery in older inpatients. J Gen Intern Med. 2011;26(12):1471–78.
2. Marshall EG, Boudreau MA, Jensen JL, et al. A new long-term care facilities model in Nova Scotia, Canada: protocol for a mixed methods study of care by design. JMIR Res Protoc. 2013;2(2):e56.
3. Rockwood K, Abeysundera MJ, Mitnitski A. How should we grade frailty in nursing home patients? J Am Med Dir Assoc. 2007;8(9):595–603.
4. Rockwood K, Song X, MacKnight C, et al. A global clinical measure of fitness and frailty in elderly people. CMAJ. 2005;173(5):489–95.
5. Java applets for power and sample size. University of Iowa. Available from: http://homepage.stat.uiowa.edu/~rlenth/Power/ Accessed Feb 23, 2015.
6. Morse JM, Field PA. Qualitative research methods for health professionals. London: Chapman & Hall; 1995.
7. Holloway I. Basic concepts for qualitative research [dictionary]. Malden M, editor. Oxford: Blackwell Science; 1997. Available from: http://www.ncbi.nlm.nih.gov/nlmcatalog/?term=Basic+Concepts+for+Qualitative+Research%2C+1997
8. Muhr T. User’s manual for ATLAS.ti 5.0. Berlin: ATLAS.ti Scientific Software Development GmbH; 2004.
9. Ritchie J, Spencer L. Qualitative data analysis for applied policy research. In: Bryman A, Burgess R, editors. Anal Qual Data. London: Routledge; 1993. p. 173–94.
10. Mays N, Pope C, editors. Assessing quality in qualitative research, 3rd edition. London: BMJ Books; 2000.
11. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician [Internet]. J Psychiatr Res. 1975;12(3):189–98. Accessed 2013 Nov 25.
12. Marshall EG, Clarke B, Peddle S, et al. “Care by design”: a new model of care in long-term care facilities providing residents coordinated on-site primary and acute care. Can Family Physician. 2014 [In Press].
13. Goldstein J, Travers A, Hubbard R, et al. Assessment of older adults by emergency medical services: methodology and feasibility of a Care Partner Comprehensive Geriatric Assessment (CP-CGA). CJEM. 2014 Jan 1;16(2):1–14.
14. Jones DM, Song X, Rockwood K. Operationalizing a frailty index from a standardized comprehensive geriatric assessment. J Am Geriatr Soc. 2004;52(11):1929–33. Available from: http://www.ncbi.nlm.nih.gov/pubmed/15507074
15. Jones D, Song X, Mitnitski A, et al. Evaluation of a frailty index based on a comprehensive geriatric assessment in a population based study of elderly Canadians [Internet]. Aging Clin Experim Res. 2005;17(6):465–71. Accessed 2013 Nov 25.
16. Rockwood K, Rockwood MR, Andrew MK, Mitnitski A. Reliability of the hierarchical assessment of balance and mobility in frail older adults. J Am Geriatr Soc. 2008;56(7):1213–17. Accessed 2013 Nov 12.
* The confidence interval is of the form p ± ME, where p is the sample proportion and ME is the margin of error: ME = z × sqrt{p × (1−p)/(n−1)} × sqrt{1−n/N}; where z is a critical value from the normal distribution, p is the sample proportion, n is the sample size, and N is the population size. (5) The “worst case” p = .5 was used in this calculation. ( Return to Text )
Canadian Geriatrics Journal , Vol. 18 , No. 1 , March 2015