https://orcid.org/0000-0003-0599-5708
ABSTRACT
Background
Information on the characteristics or long-term outcomes of people with communication support needs post-stroke is limited. We investigated associations between communication gains in rehabilitation and long-term outcomes (quality-of-life [EuroQOL-ED-3 L], mortality) by post-stroke communication support need status.
Methods
Retrospective cohort study using person-level linked data from the Australian Stroke Clinical Registry and the Australasian Rehabilitation Outcomes Centre (2014–2017). Communication support needs were assessed using the Functional Independence Measure™ comprehension and expression items recorded on admission indicated by scores one (total assistance) to five (standby prompting). Multivariable multilevel and Cox regression models were used to determine associations with long-term outcomes.
Results
Of 8,394 patients who received in-patient rehabilitation after stroke (42% female, median age 75.6 years), two-thirds had post-stroke communication support needs. Having aphasia (odds ratio [OR] 4.34, 95% CI 3.67–5.14), being aged ≥65 years (OR 1.21, 95% CI 1.08–1.36), greater stroke severity (unable to walk on admission; OR 1.48, 95% CI 1.32–1.68) and previous stroke (OR 1.25, 95% CI 1.11–1.41) were associated with increased likelihoods of having communication support needs. One-point improvement in FIM™ expression was associated with reduced likelihood of self-reporting problems related to mobility (OR 0.85, 95% CI: 0.80–0.90), self-care (OR 0.79, 95% CI: 0.74–0.86) or usual activities (OR 0.84, 95% CI: 0.75–0.94) at 90–180 days. Patients with communication support needs had greater mortality rates within one-year post-stroke (adjusted hazard ratio 1.99, 95% CI: 1.65–2.39).
Conclusions
Two-thirds of patients with stroke require communication support to participate in healthcare activities. Establishing communication-accessible stroke care environments is a priority.
Introduction
Communication is the exchange of information (spoken, written, nonverbal) that underpins activities of daily living, social relationships, participation in life roles, and employment.Citation1,Citation2 Language (aphasia), speech (dysarthria, apraxia of speech), and cognitive (cognitive-communication disorders) impairments are common causes of communication disability following stroke, impacting between 21%Citation3 and 88%Citation4 of survivors. Prevalence estimates vary considerably, and this is likely attributable to inconsistency in how communication impairment is defined and assessed in stroke services, if it is assessed at all.Citation4
In addition to conditions arising directly from the stroke event, individuals who experience a stroke may require communication support because of pre-morbid conditions (e.g. previous stroke, dementia, hearing, and vision impairment) and/or cultural and language background.Citation4 In the present study, the term communication support needs includes all patients, regardless of etiology, who require support from another person to express, or to understand basic daily needs, such as hunger, self-care, or discomfort.Citation5 This definition aligns with the scoring descriptions of the Functional Independence Measure™ (FIM™) and does not capture the full impact a communication impairment may have on a person’s activities, participation, and social relationships outside of inpatient settings.Citation1,Citation2
Communication support needs have far-reaching implications for stroke recovery. For example, communication disability and diversity can mask capacity and ability to provide informed consent for procedures or research;Citation6–8 increase the risk of adverse events, including inappropriate discharge from care;Citation9 result in unsuccessful patient-staff communication;Citation10 and reduce autonomy during stroke recovery, leading to dissatisfaction and disengagement with healthcare services.Citation11
Clinical registries, integrated outcomes centers, and administrative datasets are used to routinely collect data on patients with stroke.Citation12–14 These data can facilitate investigations of demographic characteristics, quality of care, and outcomes of patients with stroke.Citation15 Prevalence, descriptions of care, stroke scale outcomes,Citation3 hospital costs,Citation16 and associations with mood disordersCitation17 of patients with stroke and subsequent aphasia and dysarthria have been explored in such datasets. Linking datasets may enable analysis of patients’ entire stroke care pathway, from acute admission to long-term outcomes.Citation15,Citation18 In one such study by Mosalski et al.,Citation18 national acute stroke and inpatient rehabilitation registry data were linked. The data linkage facilitated analysis of associations between acute stroke therapies, outcomes during inpatient rehabilitation and self-reported outcomes at 90–180 days after stroke. Greater relative functional gain achieved during inpatient rehabilitation was associated with better longer-term health-related quality of life, greater independence and reduced self-reported hospital readmission.Citation18 However, the influence of post-stroke communication support needs on these outcomes was not investigated.
The aims of this study were to: (i) compare the profiles of people, with and without communication support needs post-stroke, who received inpatient rehabilitation, (ii) understand associations between communication gains in rehabilitation and self-reported outcomes related to quality of life and hospital readmissions at 90–180 days post-stroke, and (iii) determine associations between the presence of post-stroke communication support needs and mortality within one-year post-stroke.
Methods
Study design and population
This study was a retrospective cohort study using the linked dataset established and reported by Mosalski et al.Citation18 The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement guided the reporting of this studyCitation19 (Supplemental file S1). The study cohort comprised patients admitted to both hospital and inpatient rehabilitation between 2014 and 2017, following a stroke. Patients were excluded if they (i) were aged <18 years, (ii) died during initial hospital admission (as documented in the stroke registry data), and (iii) received rehabilitation for an impairment other than stroke or (iv) coded as TIA in the stroke registry or outcomes database.
Data sources and variables
The linked dataset was created by merging records from the Australian Stroke Clinical Registry (AuSCR) and the Australasian Rehabilitation Outcomes Centre (AROC). The data linkage process and variables collected in the AuSCR and AROC are reported in detail elsewhere.Citation12,Citation13,Citation18 Importantly, the FIM ™ is the primary outcome measure of AROC.Citation13 The FIM™ is a clinician-rated scale, completed based on observations of activities of daily living over a 24-hour period during inpatient rehabilitation.Citation5 In Australia, the FIM™ is administered in rehabilitation settings to assess functional support needs for funding and resourcing purposes.Citation20
The FIM™ comprises 18 items which are rated on a seven-point scale. A rating of one indicates total assistance is always required, and a rating of seven demonstrates total independence in a particular activity.Citation5 The FIM™ assesses two domains of functional independence: motor and cognition. Within the FIM™ cognitive domain (C-FIM™), one expression and one comprehension item are measured. Together, these two items enumerate the support needs of an individual across auditory and written comprehension, intelligibility of speech, and clarity of verbal and/or written language.Citation5 When combined with the social interaction, problem solving, and memory items, the C-FIM™ domain score can be calculated. The domain score ranges from 5–35, where higher scores indicate greater independence in cognitive activities.
Overall, measurement properties of the FIM™ are strong when tested in European and North American populations, including its reliability (Cronbach’s alpha: 0.95),Citation21 validity in detecting functional impairment after stroke (Kappa statistic: 0.92),Citation22 and responsiveness to identify change (Effect size: 0.95).Citation23 Inter-rater reliability scores for individual items in FIM™ were estimated to be between 0.95 and 0.61 for each of the 18 items. Of these, the expression item (0.73) and comprehension item (0.61) demonstrated the lowest agreement consistency.Citation21 Reasons for poor interrater reliability of these items may include scoring biasCitation23 or insensitivity of the FIM™ to document communication needs. However, no known further evaluations specific to the expression and comprehension items have been conducted.
Establishment of the cohort
The study cohort was categorized into two groups according to communication support needs on admission to inpatient rehabilitation (). Group 1: people without communication support needs, defined by FIM™ comprehension and expression admission scores of six or seven; Group 2: people with communication support needs, defined by FIM™ comprehension or expression item admission scores of five and below. Patients with missing data (covariates or outcomes) were excluded from the analysis.
Figure 1. Post-stroke communication support needs (based on AuSCR-AROC cohort).
Outcomes
Communication gains following rehabilitation were defined by FIM™ change scores for the comprehension and expression items. Differences in scores from day of admission to inpatient rehabilitation, to the scores on the day of discharge, were stratified and reported as declined, no change, or improved, consistent with AROC reporting methods.Citation12 Long-term outcomes at 90–180 days post-stroke included self-reported quality of life: Changes in either the dimension values or Visual Analogue Scale (VAS) of the EQ-5D-3 L from time of admission were used to inform gains in quality of life. Hospital readmission: Patients were asked to self-report whether they were readmitted to hospital since the time of their initial stroke event. Self-reported data were categorized as either readmitted or not readmitted to hospital since the initial stroke event. Mortality within one-year post-stroke was ascertained through linkage with the National Death Index, a registry of deaths maintained by the Australian Institute of Health and Welfare.Citation24
Covariates
Covariates included age, sex, type of stroke, stroke severity (defined according to the ability to walk upon initial presentation to hospital, in line with established methodsCitation25), comorbidities (the number of comorbidities impacting treatment), previous stroke, and the presence of post-stroke communication support needs.
Statistical analysis
Descriptive statistics were used to summarize the characteristics and rehabilitation outcomes for patients with and without communication support needs. For the subgroup with data on long-term outcomes, multivariable logistic regression models were used to determine the association between communication gains made in rehabilitation (FIM™ comprehension and expression change scores) and outcomes at follow-up (dependent variables: EQ-5D-3 L domains and hospital readmission). Cox proportional hazards regression model was used to assess the association between the presence of post-stroke communication support needs at admission to inpatient rehabilitation (FIM™ comprehension and expression item admission scores) and mortality. These models were adjusted for covariates collected in the AuSCR and AROC (age, sex, type of stroke, stroke severity, and previous stroke). Analyses were performed using STATA/MP 15.0 for Windows (StataCorp, College Station, USA, 2017), with p values of <0.05 considered statistically significant.
Data availability
Data that support the findings of this study are held by third parties and are not publicly available. Data may be made available by authors upon reasonable request with permission from the Australian Stroke Clinical Registry and the Australasian Rehabilitation Outcomes Centre.
Ethics
Ethical approval for this project was obtained from the Monash University Human Research Ethics Committee (Approval number: 16426). Approvals for the linkage of these data were obtained from the AuSCR Research Task Group and the AROC data governance process.
Results
Patient characteristics
Of the 8,394 patients included, 64% demonstrated post-stroke communication support needs on admission to inpatient rehabilitation. Presence of aphasia, older age (65 years or more), previous stroke, or greater stroke severity (as measured by inability to walk on admission) were more common in those requiring communication support (; Supplementary Table S1).
Table 1. Characteristics of linked AuSCR-AROC patients by post-stroke communication support needs status.
Rehabilitation factors
Rehabilitation outcomes differed between the two study groups (). Patients with communication support needs required longer inpatient stays (mean 14 days longer) and greater frequency of social, nursing, and allied healthcare support at discharge, than those without. Many patients with communication support needs showed improvements on the FIM™ comprehension (53%) and expression (56%) items. Despite this progress, more patients with communication support needs were discharged as a new admission to residential aged care (14%) than those without (3%).
Table 2. Comparison of inpatient rehabilitation outcomes by post-stroke communication support needs.
Long-term outcomes
Follow-up assessment was completed by 50% of all patients at a mean of 139 days post-stroke. Of patients with communication support needs, 47% completed 90–180 follow-up. Overall, gains in communication were associated with improvements in some, but not all, domains of quality of life. With each 1-point gain on the FIM™ expression item, patients were 15% less likely to report problems with mobility, 21% less likely to report problems with self-care, and 16% less likely to report problems participating in usual activities. Similar associations for gains on the FIM™ comprehension item were observed (). There were no statistically significant associations between communication gains in inpatient rehabilitation and the likelihood of self-reported hospital readmissions, or of reporting problems with pain, anxiety, or depression within 90–180 days post-stroke.
Table 3. Associations between 90–180-day health-related quality of life outcomes, hospital readmissions, mortality, and communication gains in rehabilitation for patients with post-stroke communication support needs.
Mortality
Adjusted for covariates, the presence of communication support needs on admission to inpatient rehabilitation was associated with greater rates of mortality within one-year post-stroke (). Compared with those without communication support needs, the risk of death within the first three months after stroke was three times greater (HR 3.14, 95% CI 1.94, 5.07) for those with communication support needs. At one year after stroke, 11% (n = 602) of patients with communication support needs had died, and the one-year cumulative risk of death was 4.5%.
Table 4. Associations between mortality within 1-year and post-stroke communication support needs.
Discussion
We found that two-thirds of patients with stroke had communication support needs. These findings support establishment of communication-accessible environments for stroke rehabilitation. Environmental changes may include communication boards, reductions in background noise, use of simplified language in spoken and written information, and communication partner training.Citation10 Such adaptations may facilitate more effective and independent patient communication, potentially overcoming some of the implications communication support needs present during stroke recovery.Citation6–11
The primary purpose of the FIM™ is to identify the need for assistance, however, FIM™ admission and change scores may be a predictor of patient length of stay and survival after stroke.Citation26–28 To date, no known research exists exploring the predictive nature of the comprehension and expression items in isolation, nor whether the FIM™ scores are accurate predictors for those with communication support needs. In this study, patients with communication support needs were more likely to be older, have a more severe stroke (defined as being unable to walk on admisson), have longer length of stay during their rehabiliation and have a worse outcome (e.g. die) than those without. These findings are congruent with the existing pattern of the FIM™’s predictive ability and suggest further exploration of the communication items as predictors of length of stay and survival is warranted.
The use of the FIM™ for identifying patient communication support needs emphasized the magnitude of those patients need of support. Mean admission scores were 3.5 (expression) and 3.8 (comprehension) (Supplementary Figures S1 and S2; median scores provided in Supplementary Table S2). The low mean admission scores indicated that two-thirds of stroke inpatients required support from another person to express, or to understand, basic ideas. Basic ideas are defined by the FIM™ scoring guide as daily physiological needs, such as hunger, thirst, self-care, sleep, and discomfort.Citation5 As stroke may often be the catalyst for complex decision-making around an individual’s healthcare, communication needs are essential to acknowledge when determining decision-making ability. Otherwise, communication impairment may be conflated with reduced capacity, resulting in inappropriate appointment of a guardian or power of attorney.Citation8 In such circumstances, the autonomy and preferences of those with communication support needs may be overlooked during stroke recovery.Citation7
Many patients with communication support needs made improvements on the selected FIM™ measures. However, mean scores for comprehension and expression remained within the range of requiring communication support for basic ideas at discharge. Additionally, interpreting the clinical relevance of these outcomes is challenged by the paucity of research in clinically meaningful changes for those with communication support needs. Interpretation of change scores can be guided by values of clinically important change, including the Minimal Clinically Important Difference (MCID). Such values signify the smallest difference in score which is considered important by patients.Citation29 A C-FIM™ change score of 3, and total FIM™ change score of 22, were established as the MCID for patients with stroke. However, those with communication support needs were not included in the estimation analysis.Citation30 Consequently, it can only be interpreted with caution that the mean C-FIM™ (+4.19) and total FIM™ (+24) gains observed in the present study may indicate clinically meaningful improvements.
Interpretation of long-term outcomes was also challenging. Less than 50% of respondents to the follow-up survey had communication support needs. Responders varied only significantly in age (median +1.2 years, Supplementary Table S3) from non-responders with communication support needs. Patients who opted out, or were within 6 months of a previous admission, were ineligible for follow-up.Citation24 Patients were contacted twice through postal mail, and once via telephone to complete follow-up surveys as per the AuSCR protocol. This process is a point of difference from speech and language-specific studies following stroke, which typically collect follow-up outcomes in-person, with study assessors applying supportive communication techniques.Citation31 Without additional multi-modal communication support, the language load of the EQ-5D-3 L measure or accompanying survey information likely presented barriers to participation for this cohort. Communication gains in inpatient rehabilitation were not associated with decreased likelihood of reporting problems with pain, anxiety, or depression. Accessible tools may not have been readily available in rehabilitation settings to support the reporting of pain or mood changes. Readmission rates, self-reported outcomes, and reduced likelihood of survival may reflect unmet communication needs within health services for this cohort. Understanding those needs are likely to achieve positive outcomes.
Inpatient health services are resourced through an activity-based funding model in Australia.Citation20 Under this model, FIM™ outcomes are one measure used to evidence the complexity of care a patient requires, and thus, quantify the resources services receive. The FIM™ motor items are weighted in the funding model to reflect the impact of physical support needs on the cost of care.Citation20 Cognition scores also form part of the funding model, however, are not weighted despite the known additional complexity and resource use of communication treatments in stroke rehabilitation.Citation16 The accuracy of unweighted FIM™ communication items to inform rehabilitation costs is undetermined. Patients in the current study presented with mean C-FIM™ admission scores of 18.57. The average cost of rehabilitation care reported by the Independent Hospital Pricing Authority for patients with unweighted admission C-FIM™ scores within the range of 18–35 was AUD $22,258.Citation20 However, the length of stay for the reported cohort, at 19.93 days, is notably shorter than the 27 days of patients in the present study.Citation20 Differences such as these, and the variability in reporting communication support needs across hospitals (Supplementary Figure S3), present challenges in understanding if the reported average costs adequately reflect the resources required for meaningful communication outcomes.
Altogether, there are three important reasons identified in this study to understand the sensitivity of the Measure™ in measuring outcomes for those with communication support needs. First, to understand if the FIM™ can adequately measure communication support needs and gains in this population. Second, to ensure outcomes documented on the FIM™ can be interpreted sufficiently to inform resource allocation decisions. Finally, to allow existing comprehensive data sources, such as AROC, to be analyzed to further understand this population.
Strengths
Exploration of communication support needs within routine data collection is limited. However, the largest, and most recent, analysis included over 88,000 individuals, identifying communication needs in 64% of patients’ post-stroke, consistent with the present study.Citation3 Another key strength of this study was the use of two large patient-level linked datasets from hospitals and rehabilitation centers across metropolitan, regional, and rural areas of Australia. All individuals experiencing communication support needs who were admitted to inpatient rehabilitation were included in the study, in contrast to the long-standing exclusion of patients with communication impairments from stroke research.Citation32 Data were included across the continuum of care from acute stay, inpatient rehabilitation, and long-term follow-up. The large sample size and systematic collection of outcomes are further strengths of this study. Likewise, the nature of the data collection processes which included standardized data entry, data dictionaries, and monitoring of inter-rater reliability minimized reporting bias.
Limitations
While this study provides important insights into the role of communication support needs as a predictive factor in stroke, several limitations exist. We defined our study cohorts using the FIM™. The FIM™ is a concise scale with some pragmaticCitation5 and psychometricCitation21 limitations. This assessment lacks specificity concerning the nature of communication support or whether the reported communication support needs were a result of the stroke. Another limitation is that the FIM™ measures an individual’s ability to perform tasks independently in an inpatient setting. However, in cases where individuals have social support for communication, their actual participation in daily life may surpass what is predicted or indicated by their FIM™ scores. An additional limitation to consider is that the scoring of FIM™ items can exhibit variation among different assessors, potentially introducing subjectivity into the assessment process. There is evidence that the two FIM™ communication items have moderate inter-rater reliability.Citation21 Further research is needed to create more robust functional measures, allowing studies of this nature to generate more informative data for policy development. Our group are creating Minimal Important Change (MIC) values tailored to individuals with communication support needs within the stroke population which would enhance the interpretative value of future data. This would facilitate the recognition and documentation of clinically significant improvements in stroke recovery.Citation33
The present study design may present limitations in generalizing our results to the broader population of people with post-stroke communication support needs. For example, factors influencing referrals to inpatient rehabilitation are unknown. As such, this study does not include patients with communication support needs who did not receive inpatient rehabilitation, and results may not generalize to this group. Our analyses were restricted to the available variables provided to the AuSCR and AROC in the data collection period. Therefore, associations between treatment types, or dose, provided in rehabilitation and the gains on the FIM™ or in quality of life could not be made. Follow-up data were collected in the AuSCR via patient self-report between 90 and 180 days post-stroke and may be susceptible to recall bias. Finally, it is acknowledged that post-stroke functional outcomes are measured with different instruments across regions,Citation34 which may limit the relevance of these findings to international settings.
Future directions
Further exploration of the FIM™ as a valid outcome measure for people with communication needs is required. Addition of a brief communication screening assessment to existing data collection protocols may assist understanding of the nature of communication support needs in stroke rehabilitation settings, aiding future data interpretation. Similarly, the addition of communication-specific variable may assist understanding of the associations observed between communication support needs and long-term outcomes. For example, aiding understanding of whether the greater risk of mortality seen in patients with communication support needs is a consequence of an overall frailer medical presentation, or the result of the negative impacts of communication on engagement with inpatient care.
Conclusion
This project represented an initial exploration of large-scale linked data using the FIM™ as the primary measure, to better understand post-stroke communication support needs. Critically, this indicates a requirement to prioritize communicatively accessible environments in stroke care settings. Further exploration of the FIM™ as a valid outcome measure for those with communication needs is required.
Acknowledgments
The authors of this study thank members of the Australian Stroke Clinical Registry (AuSCR) Steering and Management Committee and the Australasian Rehabilitation Outcomes Centre (AROC) Management Advisory Group. We would also like to acknowledge site investigators, clinicians and patients who contributed data to the AuSCR and the AROC.
Disclosure statement
Prof Cadilhac is the current Data Custodian for the AuSCR. Professors Cadilhac, Faux and A/Prof Kilkenny are members of the AuSCR Steering or Management Committees. Dr Ross Clifton is the Director of the Australasian Rehabilitation Outcomes Centre (AROC), and Ms Alexander is the AROC Database Manager and Statistician.
Additional information
Funding
References
- Wray F, Clarke D. Longer-term needs of stroke survivors with communication difficulties living in the community: a systematic review and thematic synthesis of qualitative studies. BMJ Open. 2017;7(10):e017944. doi:10.1136/bmjopen-2017-017944.
- Hilari K. The impact of stroke: are people with aphasia different to those without? Disabil Rehabil. 2011;33(3):211–218. doi:10.3109/09638288.2010.508829.
- Mitchell C, Gittins M, Tyson S, Vail A, Conroy P, Paley L, et al. Prevalence of aphasia and dysarthria among inpatient stroke survivors: describing the population, therapy provision and outcomes on discharge. Aphasiology. 2021;35(7):950–960. doi:10.1080/02687038.2020.1759772.
- O’Halloran R, Worrall LE, Hickson L. The number of patients with communication related impairments in acute hospital stroke units. Int J Speech Lang Pathol. 2009;11(6):438–449. doi:10.3109/17549500902741363.
- Dodds TA, Martin DP, Stolov WC, Deyo RA. A validation of the functional independence measurement and its performance among rehabilitation inpatients. Arch Phys Med Rehabil. 1993;74(5):531–536. doi:10.1016/0003-9993(93)90119-U.
- Brady MC, Fredrick A, Williams B. People with aphasia: capacity to consent, research participation and intervention inequalities. Int J Stroke. 2013;8(3):193–196. doi:10.1111/j.1747-4949.2012.00900.x.
- Borrett S, Gould LJ. Mental capacity assessment with people with aphasia: understanding the role of the speech and language therapist. Aphasiology. 2021;35(11):1463–1481. doi:10.1080/02687038.2020.1819954.
- McCormick M, Bose A, Marinis T. Decision-making capacity in aphasia: SLT’s contribution in England. Aphasiology. 2017;31(11):1344–1358. doi:10.1080/02687038.2017.1355441.
- Hemsley B, Werninck M, Worrall L. “That really shouldn’t have happened”: people with aphasia and their spouses narrate adverse events in hospital. Aphasiology. 2013;27:706–722. doi:10.1080/02687038.2012.748181.
- O’Halloran RP, Grohn BBSP, Worrall LP. Environmental factors that influence communication for patients with a communication disability in acute hospital stroke units: a qualitative metasynthesis. Arch Phys Med Rehabil. 2012;93(1):S77–S85. doi:10.1016/j.apmr.2011.06.039.
- Clancy L, Povey R, Rodham K. “Living in a foreign country”: experiences of staff–patient communication in inpatient stroke settings for people with post-stroke aphasia and those supporting them. Disabil Rehabil. 2020;42(3):324–334. doi:10.1080/09638288.2018.1497716.
- Australian Stroke Clinical Registry. About the AuSCR; 2021. https://auscr.com.au/about/.
- University of Wollongong. About AROC. Australasian Rehabilitation Outcomes Centre; 2021. https://www.uow.edu.au/ahsri/aroc/about/.
- Australian Institute of Health and Welfare. Stroke snapshot; 2020. https://www.aihw.gov.au/reports/australias-health/stroke.
- Kilkenny MF, Kim J, Andrew NE, et al. Maximising data value and avoiding data waste: a validation study in stroke research. Med J Australia. 2019;210(1):27–31. doi:10.5694/mja2.12029.
- Brogan EL, Kim J, Grimley RS, et al. The excess costs of hospitalization for acute stroke in people with communication impairment: a Stroke123 data linkage substudy. Arch Phys Med Rehabil. 2023;104(6):942–949. doi:10.1016/j.apmr.2023.01.015.
- Thayabaranathan T, Andrew NE, Stolwyk R, Lannin NA, Cadilhac DA. Comparing the EQ-5D-3L anxiety or depression domain to the hospital anxiety and depression scale to identify anxiety or depression after stroke. Top Stroke Rehabil. 2022;29(2):146–155. doi:10.1080/10749357.2021.1895494.
- Mosalski S, Shiner CT, Lannin NA, et al. Increased relative functional gain and improved stroke outcomes: a linked registry study of the impact of rehabilitation. J Stroke Cerebrovascular Dis. 2021;30(10):106015. doi:10.1016/j.jstrokecerebrovasdis.2021.106015.
- Von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. PLoS Med. 2007;4(10):1623–1627. doi:10.1371/journal.pmed.0040296.
- Independent Hospital Pricing Authority. Development of the Australian National subacute and non-acute patient classification version 5.0; 2021. https://www.ihacpa.gov.au/.
- Ottenbacher KJ, Hsu Y, Granger CV, Fiedler RC. The reliability of the functional independence measure: a quantitative review. Arch Phys Med Rehabil. 1996;77(12):1226–1232. doi:10.1016/S0003-9993(96)90184-7.
- Kidd D, Stewart G, Baldry J, et al. The functional independence measure: a comparative validity and reliability study. Disabil Rehabil. 1995;17(1):10–14. doi:10.3109/09638289509166622.
- van der Putten JJMF, Hobart JC, Freeman JA, Thompson AJ. Measuring change in disability after inpatient rehabilitation: comparison of the responsiveness of the Barthel Index and the functional independence measure. J Neurol, Neurosurg & Psychiatry. 1999;66(4):480–484. doi:10.1136/jnnp.66.4.480.
- Cadilhac DA, Dalli LL, Morrison J, et al. The Australian stroke Clinical registry annual report 2020. The Florey Institute of Neuroscience and Mental Health; 2021.
- Counsell C, Dennis M, McDowall M, Warlow C. Predicting outcome after acute and subacute stroke. Stroke. 2002;33(4):1041–1047. doi:10.1161/hs0402.105909.
- Atalay A, Turhan N. Determinants of length of stay in stroke patients: a geriatric rehabilitation unit experience. Int J Rehabil Res. 2009;32(1):48–52. doi:10.1097/MRR.0b013e32830d3689.
- García-Rudolph A, Cegarra B, Opisso E, Tormos JM, Bernabeu M, Saurí J. Predicting length of stay in patients admitted to stroke rehabilitation with severe and moderate levels of functional impairments. Medicine. 2020;99(43):e22423–e22423. doi:10.1097/MD.0000000000022423.
- Scrutinio D, Monitillo V, Guida P, et al. Functional gain after inpatient stroke rehabilitation: correlates and impact on long-term survival. Stroke. 2015;46(10):2976–2980. doi:10.1161/STROKEAHA.115.010440.
- Jaeschke R, Singer J, Guyatt GH. Measurement of health status: ascertaining the minimal clinically important difference. Control Clin Trials. 1989;10(4):407–415. doi:10.1016/0197-2456(89)90005-6.
- Beninato M, Gill-Body KM, Salles S, Stark PC, Black-Schaffer RM, Stein J. Determination of the Minimal clinically important difference in the FIM instrument in patients with stroke. Arch Phys Med Rehabil. 2006;87(1):32–39. doi:10.1016/j.apmr.2005.08.130.
- Brady MC, Kelly H, Godwin J, Enderby P. Speech and language therapy for aphasia after stroke. Cochrane Database. 2016;6. doi:10.1002/14651858.CD000425.pub4.
- Shiggins C, Ryan B, O’Halloran R, et al. Towards the consistent inclusion of people with aphasia in stroke research irrespective of discipline. Arch Phys Med Rehabil. 2022;103(11):2256–2263. doi:10.1016/j.apmr.2022.07.004.
- Harvey S, Stone M, Zingelman S, et al. Evaluation of the acceptability and feasibility of implementing a minimum dataset to monitor stroke care for people with aphasia. Stroke 2023 – the combined stroke society of Australasia and smart strokes nursing and allied health scientific meeting. Int J Stroke. 2023;18. doi:10.1177/17474930231188838.
- Centers for Medicare & Medicaid Services. Inpatient rehabilitation facility (IRF) quality reporting program (QRP). https://www.cms.gov/medicare/quality/inpatient-rehabilitation-facility.