Prevalence and Outcomes of Suspected Dysphagia in Elderly Inpatients Undergoing Cardiac Rehabilitation Authors: Naoto Miyawaki, Akira Takashima Published: 2026 ABSTRACT BackgroundSwallowing dysfunction may complicate recovery in inpatients aged ≥75 years undergoing cardiac rehabilitation (CR). This study examined the prevalence of suspected dysphagia and its associations with physical function, activities of daily living, and discharge outcomes.Methods and ResultsThis prospective single-center study enrolled 59 inpatients undergoing CR between May and December 2023. Patients were screened within 3 days of admission using the Eating Assessment Tool-10 (EAT-10; score ≥3) or the Repetitive Saliva Swallowing Test (RSST; <3 swallows/30 s). Twenty-nine patients (49.2%) screened positive. At admission, patients with suspected dysphagia had lower handgrip strength (mean [±standard deviation] 15.19±7.20 vs. 20.01±7.24 kg; P=0.013), smaller calf circumference (median [interquartile range] 28 [26–31] vs. 31 [29–33] cm; P=0.002), and lower Short Physical Performance Battery scores (median [interquartile range] 4 [3–9] vs. 8.5 [5–12]; P=0.004). The discharge Barthel Index was also lower (median [interquartile range] 80 [50–90] vs. 85 [80–100]; P=0.037). Among patients admitted from home (n=57), non-home discharge/in-hospital death was more frequent in those with suspected dysphagia (25.0% vs. 6.9%; odds ratio 4.50), as was long-term care insurance certification (67.9% vs. 20.7%; P<0.001).ConclusionsSuspected dysphagia was common and associated with poorer physical function, lower activities of daily living at discharge, and greater care needs. Early swallowing screening may support discharge planning. Central Figure Swallowing dysfunction is common among older adults and strongly associated with aspiration pneumonia, malnutrition, functional decline, and admission to long-term care facilities.1,2 The Eating Assessment Tool-10 (EAT-10) and the Repetitive Saliva Swallowing Test (RSST) are simple, non-invasive, and feasible screening tools for bedside use, enabling the identification of patients with suspected dysphagia, which is defined as screening positive for swallowing dysfunction on either test rather than having instrumentally confirmed dysphagia.3,4 Swallowing impairment may complicate recovery among inpatients with cardiovascular disease by limiting nutritional intake, reducing exercise tolerance, and increasing the risk of adverse clinical outcomes.5–8 In Japan, the national fee schedule specifies provider categories for cardiovascular disease-specific rehabilitation fee (cardiac rehabilitation; H000), but speech–language–hearing therapists are not listed.9 However, speech–language–hearing therapists are explicitly listed as providers for stroke-related rehabilitation fees (H001).9 Hence, direct involvement of speech–language pathologists (SLPs) in inpatient cardiac rehabilitation (CR) may not be routinely implemented across institutions, and early detection of suspected dysphagia frequently relies on non-SLP rehabilitation staff, including physical therapists and other CR team members. Despite the clinical relevance of swallowing problems in this setting, few studies have systematically evaluated the prevalence of suspected dysphagia identified by screening with the EAT-10 and/or RSST and its associations with functional recovery and discharge-related outcomes in elderly inpatients undergoing CR. Therefore, the aims of this study were to determine the prevalence of suspected dysphagia based on EAT-10 and RSST screening among inpatients aged ≥75 years undergoing CR and to examine its associations with physical function, activities of daily living (ADL), discharge destinations, and long-term care insurance (LTCI) certification. Methods Study Design and Participants This single-center prospective observational study was conducted at Kitajima Taoka Hospital (Izankai Medical Corporation, Tokushima, Japan) between May and December 2023. Of 151 eligible patients who participated in inpatient CR reimbursed under the Japanese National Health Insurance system, 45 who were aged <75 years were excluded. Among the remaining 106 patients aged ≥75 years, 47 were excluded because they declined to participate (n=4), had severe dementia or altered consciousness (n=24), had missing swallowing screening data (n=12), had a history of cerebrovascular or neuromuscular disease with clinically diagnosed dysphagia (n=4), had received SLP intervention at admission (n=2), or had experienced an acute cerebrovascular event during hospitalization (n=1). Finally, 59 patients were included in the analysis (Figure 1). Eligible patients were aged ≥75 years, were admitted to general or community-care wards with a diagnosis of cardiovascular disease, and received structured inpatient CR in accordance with the JCS/JACR guideline.10Figure 1.Flow diagram of patient selection and analysis. Suspected dysphagia was defined as a positive screening result at admission based on an Eating Assessment Tool-10 (EAT-10) score ≥3 and/or a Repetitive Saliva Swallowing Test (RSST) result of <3 swallows/30 s. Among patients aged ≥75 years, individuals were excluded if swallowing screening data were incomplete (EAT-10 and/or RSST), if they had severe dementia or altered consciousness, if they had pre-existing cerebrovascular or neuromuscular disease with clinically diagnosed dysphagia, if they had prior speech–language pathologist (SLP) intervention at admission, or if they developed an acute cerebrovascular event during hospitalization. Flow diagram of patient selection and analysis. Suspected dysphagia was defined as a positive screening result at admission based on an Eating Assessment Tool-10 (EAT-10) score ≥3 and/or a Repetitive Saliva Swallowing Test (RSST) result of <3 swallows/30 s. Among patients aged ≥75 years, individuals were excluded if swallowing screening data were incomplete (EAT-10 and/or RSST), if they had severe dementia or altered consciousness, if they had pre-existing cerebrovascular or neuromuscular disease with clinically diagnosed dysphagia, if they had prior speech–language pathologist (SLP) intervention at admission, or if they developed an acute cerebrovascular event during hospitalization. Swallowing Screening Swallowing screening was performed within 3 days of admission and, whenever feasible, before the initiation of structured rehabilitation sessions. Suspected dysphagia was defined as a positive screening result on either the EAT-10 or the RSST. The EAT-10 is a 10-item symptom questionnaire. A total score of ≥3 was used as the cut-off for suspected dysphagia, based on the original validation study3 and a meta-analytic diagnostic accuracy study.4 The EAT-10 was completed with staff assistance for all participants. The RSST assesses the number of saliva swallows within 30 s. In this study, <3 swallows within 30 s was used as the cut-off, consistent with validation and feasibility data.11,12 All screening tests were administered by trained rehabilitation staff using standardized procedures. Clinical Assessments and Outcomes Baseline characteristics included age, sex, primary cardiovascular diagnosis, and major cardiovascular procedures or treatments during hospitalization (e.g., percutaneous coronary intervention, endovascular therapy, or thoracic surgery). Laboratory parameters, including serum albumin, hemoglobin, and B-type natriuretic peptide, were measured at admission in the hospital’s central laboratory using standardized automated assays. Physical function was evaluated during hospitalization by trained rehabilitation staff. Handgrip strength was measured as the maximum value of 2 trials. Calf circumference was measured at the maximum diameter. The Short Physical Performance Battery (SPPB) and the 6-minute walk distance (6MWD) were assessed according to standard procedures when feasible. ADL were assessed using the Barthel Index (BI) at admission and within 48 h before discharge. The change in BI from admission to discharge (∆BI) was calculated as discharge BI minus admission BI. Discharge destination was categorized as home discharge, discharge to a long-term care facility, or in-hospital death. For analyses of the binary discharge outcome, non-home discharge was defined as discharge to a long-term care facility or in-hospital death. LTCI certification status was assessed at discharge. The “not certified” category included patients without LTCI certification, as well as those with a pending LTCI application at discharge. The primary outcome was the prevalence of suspected dysphagia based on EAT-10 or RSST screening. Secondary outcomes included physical function (handgrip strength, calf circumference, SPPB, and 6MWD), ADL (BI and ∆BI), discharge destination (including non-home discharge or in-hospital death), and LTCI certification status at discharge. Statistical Analysis The distribution of continuous variables was assessed for normality using the Shapiro–Wilk test. Normally distributed continuous variables are presented as the mean±SD and were compared between groups using Student’s t-test or Welch’s t-test, as appropriate. Continuous variables with a skewed distribution are presented as the median with interquartile range (IQR) and were compared using the Mann–Whitney U test. Categorical variables are presented as counts and percentages and were compared using Pearson’s Chi-squared test or Fisher’s exact test, as appropriate: Pearson’s Chi-squared test was used for 2×4 contingency tables, and Fisher’s exact test was used when expected cell counts were small for other categorical comparisons. For r×c (row-by-column) categorical comparisons with sparse cells, Fisher’s exact test (2-sided) was used. Effect sizes were calculated as Cohen’s d for normally distributed continuous variables and as the rank-biserial correlation (r) for non-parametric variables. For 2×2 categorical comparisons, the effect size was calculated using the phi coefficient (φ). For binary outcomes, odds ratios (ORs) with 95% confidence intervals (CIs) were calculated from 2×2 contingency tables using the Wald method. Analyses of discharge destination and LTCI certification status were restricted to patients admitted from home (n=57); hence, 2 patients admitted from long-term care facilities were excluded from these analyses. All analyses were performed using R version 4.3.1 (R Foundation for Statistical Computing, Vienna, Austria) with R Commander version 4.5.1. All statistical tests were 2-sided, with significance defined as P<0.05. Ethical Considerations This study was approved by the Institutional Review Board of Kitajima Taoka Hospital (Approval no. 2025-13) before data analysis. Written informed consent was obtained from all participants. Results Of the 151 patients initially screened, 59 were included in the final analysis (Figure 1). Of these 59 patients, 29 (49.2%) screened positive for suspected dysphagia based on either the EAT-10 (score ≥3) or RSST (<3 swallows/30 s). Table 1 presents the baseline characteristics of patients with and without suspected dysphagia. Table 2 presents the number of patients meeting the cut-off values for each screening tool and for both tools combined. There were no significant differences between patients with and without suspected dysphagia in age (85.5±6.7 vs. 83.5±4.6 years, respectively; Table 1; P=0.19) or serum albumin concentrations (3.41±0.60 vs. 3.66±0.62 g/dL, respectively; Table 1; P=0.128). In-hospital death was recorded for 3 patients in the suspected dysphagia group; no patients in the group without suspected dysphagia died in hospital (Table 3). Table 1.Baseline and Admission Characteristics According to Suspected Dysphagia Status Without suspected dysphagia (n=30)With suspected dysphagia (n=29)P valueAge (years)83.5±4.685.5±6.70.189Male sex21 (70.0)14 (48.3)0.115Primary diagnosis or index treatment category 0.915 Heart failure17 (56.7)16 (55.2) Endovascular therapy4 (13.3)5 (17.2) Percutaneous coronary intervention5 (16.7)3 (10.3) Post-thoracic surgery4 (13.3)5 (17.2) At admission Handgrip strength (kg)20.01±7.2415.19±7.200.013 Calf circumference (cm)31 [29–33]28 [26–31]0.002 SPPB8.5 [5–12]4 [3–9]0.004 6MWD (m)187.0±120.5140.9±106.80.126 Barthel Index80 [70–85]75 [50–80]0.148HDS-R25 [24–26]24 [19–27]0.377Serum albumin (g/dL)3.66±0.623.41±0.600.128Hemoglobin (g/dL)12.04±2.6510.90±1.930.064BNP (pg/mL)331.9 [85.6–487.7]113.7 [55.0–338.2]0.125Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). P values were calculated using Student’s t-test or Welch’s t-test for normally distributed continuous variables, the Mann–Whitney U test for continuous variables with a skewed distribution, and Fisher’s exact test for categorical variables. For 2×4 contingency tables, Pearson’s Chi-squared test was used when appropriate, and Fisher’s exact test was used when expected cell counts were small. Totals may not equal 100% because percentages were rounded to one decimal place. 6MWD, 6-minute walk distance; BNP, B-type natriuretic peptide; HDS-R, Hasegawa Dementia Scale–Revised; SPPB, Short Physical Performance Battery. Table 2.Detailed Results of Swallowing Screening Using the EAT-10 or the RSST (n=59)Screening resultNo. patients (%)Both EAT-10 and RSST positive7 (11.9)EAT-10 only positive8 (13.6)RSST only positive14 (23.7)Both EAT-10 and RSST negative30 (50.8)Suspected dysphagia (either EAT-10 or RSST positive)29 (49.2)Eating Assessment Tool-10 (EAT-10) positivity was defined as a score of ≥3, and Repetitive Saliva Swallowing Test (RSST) positivity was defined as <3 swallows/30 s. Suspected dysphagia was defined as positivity on either screening test. The EAT-10 was administered with staff assistance to all participants. EAT-10 and RSST are screening tools; therefore, positive results indicate suspected dysphagia (screening-positive swallowing dysfunction) rather than instrumentally confirmed dysphagia. Totals may not equal 100% because percentages were rounded to one decimal place. Table 3.In-Hospital and Discharge Outcomes According to Suspected Dysphagia Status (Available Case Analysis) Without suspected dysphagia (n=30)With suspected dysphagia (n=29)P valueSLP intervention0 (0.0)4 (13.8)0.052In-hospital death0 (0.0)3 (10.3)0.112Length of hospital stay (days)22.5 [10.25–35.0]38.0 [22.0–60.0]0.086SPPB at discharge11 [7–12]9.5 [4–11.25] (n=24)0.1626MWD at discharge (m)231.8±111.3220.5±121.6 (n=22)0.729Barthel Index at discharge85 [80–100]80 [50–90]0.037Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). P values were calculated using Student’s t-test or Welch’s t-test for normally distributed continuous variables, the Mann–Whitney U test for continuous variables with a skewed distribution, and Fisher’s exact test for categorical variables. The Short Physical Performance Battery (SPPB) and 6-minute walk distance (6MWD) at discharge were analyzed using available case data. In the suspected dysphagia group (n=29), discharge SPPB was available for 24 patients and unavailable for 5 (3 in-hospital deaths, 2 patients unable to complete the assessment); discharge 6MWD was available for 22 patients and unavailable for 7 (3 in-hospital deaths, 4 patients unable to complete the assessment). Suspected dysphagia was defined as a positive result on either the Eating Assessment Tool-10 (score ≥3) or the Repetitive Saliva Swallowing Test (<3 swallows/30 s). SLP, speech–language pathologist. Table 1.Baseline and Admission Characteristics According to Suspected Dysphagia Status Without suspected dysphagia (n=30)With suspected dysphagia (n=29)P valueAge (years)83.5±4.685.5±6.70.189Male sex21 (70.0)14 (48.3)0.115Primary diagnosis or index treatment category 0.915 Heart failure17 (56.7)16 (55.2) Endovascular therapy4 (13.3)5 (17.2) Percutaneous coronary intervention5 (16.7)3 (10.3) Post-thoracic surgery4 (13.3)5 (17.2) At admission Handgrip strength (kg)20.01±7.2415.19±7.200.013 Calf circumference (cm)31 [29–33]28 [26–31]0.002 SPPB8.5 [5–12]4 [3–9]0.004 6MWD (m)187.0±120.5140.9±106.80.126 Barthel Index80 [70–85]75 [50–80]0.148HDS-R25 [24–26]24 [19–27]0.377Serum albumin (g/dL)3.66±0.623.41±0.600.128Hemoglobin (g/dL)12.04±2.6510.90±1.930.064BNP (pg/mL)331.9 [85.6–487.7]113.7 [55.0–338.2]0.125Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). P values were calculated using Student’s t-test or Welch’s t-test for normally distributed continuous variables, the Mann–Whitney U test for continuous variables with a skewed distribution, and Fisher’s exact test for categorical variables. For 2×4 contingency tables, Pearson’s Chi-squared test was used when appropriate, and Fisher’s exact test was used when expected cell counts were small. Totals may not equal 100% because percentages were rounded to one decimal place. 6MWD, 6-minute walk distance; BNP, B-type natriuretic peptide; HDS-R, Hasegawa Dementia Scale–Revised; SPPB, Short Physical Performance Battery. Baseline and Admission Characteristics According to Suspected Dysphagia Status Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). P values were calculated using Student’s t-test or Welch’s t-test for normally distributed continuous variables, the Mann–Whitney U test for continuous variables with a skewed distribution, and Fisher’s exact test for categorical variables. For 2×4 contingency tables, Pearson’s Chi-squared test was used when appropriate, and Fisher’s exact test was used when expected cell counts were small. Totals may not equal 100% because percentages were rounded to one decimal place. 6MWD, 6-minute walk distance; BNP, B-type natriuretic peptide; HDS-R, Hasegawa Dementia Scale–Revised; SPPB, Short Physical Performance Battery. Table 2.Detailed Results of Swallowing Screening Using the EAT-10 or the RSST (n=59)Screening resultNo. patients (%)Both EAT-10 and RSST positive7 (11.9)EAT-10 only positive8 (13.6)RSST only positive14 (23.7)Both EAT-10 and RSST negative30 (50.8)Suspected dysphagia (either EAT-10 or RSST positive)29 (49.2)Eating Assessment Tool-10 (EAT-10) positivity was defined as a score of ≥3, and Repetitive Saliva Swallowing Test (RSST) positivity was defined as <3 swallows/30 s. Suspected dysphagia was defined as positivity on either screening test. The EAT-10 was administered with staff assistance to all participants. EAT-10 and RSST are screening tools; therefore, positive results indicate suspected dysphagia (screening-positive swallowing dysfunction) rather than instrumentally confirmed dysphagia. Totals may not equal 100% because percentages were rounded to one decimal place. Detailed Results of Swallowing Screening Using the EAT-10 or the RSST (n=59) Eating Assessment Tool-10 (EAT-10) positivity was defined as a score of ≥3, and Repetitive Saliva Swallowing Test (RSST) positivity was defined as <3 swallows/30 s. Suspected dysphagia was defined as positivity on either screening test. The EAT-10 was administered with staff assistance to all participants. EAT-10 and RSST are screening tools; therefore, positive results indicate suspected dysphagia (screening-positive swallowing dysfunction) rather than instrumentally confirmed dysphagia. Totals may not equal 100% because percentages were rounded to one decimal place. Table 3.In-Hospital and Discharge Outcomes According to Suspected Dysphagia Status (Available Case Analysis) Without suspected dysphagia (n=30)With suspected dysphagia (n=29)P valueSLP intervention0 (0.0)4 (13.8)0.052In-hospital death0 (0.0)3 (10.3)0.112Length of hospital stay (days)22.5 [10.25–35.0]38.0 [22.0–60.0]0.086SPPB at discharge11 [7–12]9.5 [4–11.25] (n=24)0.1626MWD at discharge (m)231.8±111.3220.5±121.6 (n=22)0.729Barthel Index at discharge85 [80–100]80 [50–90]0.037Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). P values were calculated using Student’s t-test or Welch’s t-test for normally distributed continuous variables, the Mann–Whitney U test for continuous variables with a skewed distribution, and Fisher’s exact test for categorical variables. The Short Physical Performance Battery (SPPB) and 6-minute walk distance (6MWD) at discharge were analyzed using available case data. In the suspected dysphagia group (n=29), discharge SPPB was available for 24 patients and unavailable for 5 (3 in-hospital deaths, 2 patients unable to complete the assessment); discharge 6MWD was available for 22 patients and unavailable for 7 (3 in-hospital deaths, 4 patients unable to complete the assessment). Suspected dysphagia was defined as a positive result on either the Eating Assessment Tool-10 (score ≥3) or the Repetitive Saliva Swallowing Test (<3 swallows/30 s). SLP, speech–language pathologist. In-Hospital and Discharge Outcomes According to Suspected Dysphagia Status (Available Case Analysis) Unless indicated otherwise, data are given as the mean±SD, median [interquartile range], or n (%). P values were calculated using Student’s t-test or Welch’s t-test for normally distributed continuous variables, the Mann–Whitney U test for continuous variables with a skewed distribution, and Fisher’s exact test for categorical variables. The Short Physical Performance Battery (SPPB) and 6-minute walk distance (6MWD) at discharge were analyzed using available case data. In the suspected dysphagia group (n=29), discharge SPPB was available for 24 patients and unavailable for 5 (3 in-hospital deaths, 2 patients unable to complete the assessment); discharge 6MWD was available for 22 patients and unavailable for 7 (3 in-hospital deaths, 4 patients unable to complete the assessment). Suspected dysphagia was defined as a positive result on either the Eating Assessment Tool-10 (score ≥3) or the Repetitive Saliva Swallowing Test (<3 swallows/30 s). SLP, speech–language pathologist. Patients with suspected dysphagia had significantly lower handgrip strength (15.19±7.20 vs. 20.01±7.24 kg; P=0.013), smaller calf circumference (median 28 [IQR 26–31] vs. 31 [IQR 29–33] cm; P=0.002), and lower SPPB scores at admission (median 4 [IQR 3–9] vs. 8.5 [IQR 5–12]; P=0.004) than those without suspected dysphagia (Table 1; Supplementary Table 1). Physical performance outcomes at discharge were analyzed using available case data; in the suspected dysphagia group (n=29), discharge SPPB and 6MWD were available for 24 and 22 patients, respectively (data were unavailable because of in-hospital death [n=3] or an inability to complete the assessment). SPPB at discharge did not differ significantly between the groups with and without suspected dysphagia (median 9.5 [IQR 4–11.25] vs. 11 [IQR 7–12], respectively; P=0.162; Table 3). However, the BI at discharge was lower among those with than without suspected dysphagia (median 80 [IQR 50–90] vs. 85 [IQR 80–100]; P=0.037; Table 3). Other clinical characteristics did not differ significantly between the 2 groups (Table 3). Discharge destination and LTCI certification status were analyzed among patients admitted from home (n=57), excluding 2 patients admitted from long-term care facilities. Figures 2 and 3 show the associations between suspected dysphagia and non-home discharge (discharge to a long-term care facility or in-hospital death) and LTCI certification status at discharge, respectively. Figure 2.Discharge destinations (home or non-home/in-hospital death) among patients admitted from home (n=57), with and without suspected dysphagia. There was no significant difference in discharge destinations between the 2 groups (Fisher’s exact test, 2-sided P=0.079). Figure 3.Long-term care insurance (LTCI) certification status at discharge among patients admitted from home (n=57), with and without suspected dysphagia. LTCI certification status at discharge differed significantly between the 2 groups (Fisher’s exact test, 2-sided P=0.00048). Discharge destinations (home or non-home/in-hospital death) among patients admitted from home (n=57), with and without suspected dysphagia. There was no significant difference in discharge destinations between the 2 groups (Fisher’s exact test, 2-sided P=0.079). Long-term care insurance (LTCI) certification status at discharge among patients admitted from home (n=57), with and without suspected dysphagia. LTCI certification status at discharge differed significantly between the 2 groups (Fisher’s exact test, 2-sided P=0.00048). Item-level distributions of the BI (feeding, transfer, and mobility) at admission and discharge are presented in Supplementary Table 2, Sections A–C. The distribution of the transfer item differed significantly between the 2 groups at both admission and discharge (both P=0.026), whereas there were no significant intergroup differences in the feeding and mobility items. ADL recovery was evaluated using the ∆BI. EAT-10 and RSST scores at admission were not significantly correlated with ∆BI, and the results remained consistent in sensitivity analyses excluding in-hospital deaths (Figure 4; Supplementary Table 2, Sections D and E). Item-level analysis of the BI showed that between-group differences were most evident in transfer-related functions. Figure 4.Scatter plot of Eating Assessment Tool-10 (EAT-10) scores at admission and changes in the Barthel Index (∆BI) for 59 patients. The ∆BI was calculated as discharge BI minus admission BI. The association between the EAT-10 score at admission and ∆BI was evaluated using Spearman’s rank correlation. The EAT-10 was administered with staff assistance for all participants. To improve visibility, points have been slightly jittered along the x-axis for visualization purposes. Scatter plot of Eating Assessment Tool-10 (EAT-10) scores at admission and changes in the Barthel Index (∆BI) for 59 patients. The ∆BI was calculated as discharge BI minus admission BI. The association between the EAT-10 score at admission and ∆BI was evaluated using Spearman’s rank correlation. The EAT-10 was administered with staff assistance for all participants. To improve visibility, points have been slightly jittered along the x-axis for visualization purposes. Discharge Outcomes Among patients admitted from home (n=57), the composite outcome of non-home discharge (discharge to a long-term care facility) or in-hospital death occurred in 25.0% (7/28) of patients with suspected dysphagia, compared with 6.9% (2/29) of patients without suspected dysphagia (Figure 2). In the group with suspected dysphagia, 4 patients were discharged to long-term care facilities and 3 died during hospitalization (Table 4). To provide additional information, we summarized the primary reasons for non-home discharge and commonly documented rehabilitation-related needs among the 9 patients admitted from home who were not discharged home (Table 4). These descriptive data are presented to illustrate real-world care needs and post-acute coordination requirements. Table 4.Reasons for Non-Home Discharge and Commonly Documented Rehabilitation-Related Needs Among 9 Patients Admitted From Home But Not Discharged Home(A) Primary reason for non-home dischargeWithout suspected dysphagia (n=2)With suspected dysphagia (n=7)Primary reason for transfer In-hospital death0 (0.0)3 (42.9) Functional decline/ADL assistance0 (0.0)3 (42.9) Medical management required1 (50.0)1 (14.3) Social reasons (caregiver/home environment)1 (50.0)0 (0.0)(B) Commonly documented rehabilitation-related needs (multiple responses allowed)Without suspected dysphagia (n=2)With suspected dysphagia (n=7)Required rehabilitation content Gait training2 (100.0)2 (28.6) Diet modification0 (0.0)4 (57.1) Swallowing screening0 (0.0)3 (42.9) Caregiver education0 (0.0)2 (28.6) Feeding assistance0 (0.0)2 (28.6) Transfer training0 (0.0)2 (28.6) Posture adjustment0 (0.0)1 (14.3) Not applicable (in-hospital death)A0 (0.0)1 (14.3)Values are n (%). In (B), multiple responses were allowed; therefore, within each group, the sum of counts may exceed the number of patients, and the sum of percentages may exceed 100%. Rehabilitation-related needs were abstracted from chart documentation during hospitalization; therefore, patients who died in hospital could also contribute to these counts. AThis indicates a patient who died after rapid clinical deterioration for whom no specific rehabilitation-related needs (e.g., diet modification or gait training) were documented in the chart during hospitalization. ADL, activities of daily living. Table 4.Reasons for Non-Home Discharge and Commonly Documented Rehabilitation-Related Needs Among 9 Patients Admitted From Home But Not Discharged Home(A) Primary reason for non-home dischargeWithout suspected dysphagia (n=2)With suspected dysphagia (n=7)Primary reason for transfer In-hospital death0 (0.0)3 (42.9) Functional decline/ADL assistance0 (0.0)3 (42.9) Medical management required1 (50.0)1 (14.3) Social reasons (caregiver/home environment)1 (50.0)0 (0.0)(B) Commonly documented rehabilitation-related needs (multiple responses allowed)Without suspected dysphagia (n=2)With suspected dysphagia (n=7)Required rehabilitation content Gait training2 (100.0)2 (28.6) Diet modification0 (0.0)4 (57.1) Swallowing screening0 (0.0)3 (42.9) Caregiver education0 (0.0)2 (28.6) Feeding assistance0 (0.0)2 (28.6) Transfer training0 (0.0)2 (28.6) Posture adjustment0 (0.0)1 (14.3) Not applicable (in-hospital death)A0 (0.0)1 (14.3)Values are n (%). In (B), multiple responses were allowed; therefore, within each group, the sum of counts may exceed the number of patients, and the sum of percentages may exceed 100%. Rehabilitation-related needs were abstracted from chart documentation during hospitalization; therefore, patients who died in hospital could also contribute to these counts. AThis indicates a patient who died after rapid clinical deterioration for whom no specific rehabilitation-related needs (e.g., diet modification or gait training) were documented in the chart during hospitalization. ADL, activities of daily living. Reasons for Non-Home Discharge and Commonly Documented Rehabilitation-Related Needs Among 9 Patients Admitted From Home But Not Discharged Home Values are n (%). In (B), multiple responses were allowed; therefore, within each group, the sum of counts may exceed the number of patients, and the sum of percentages may exceed 100%. Rehabilitation-related needs were abstracted from chart documentation during hospitalization; therefore, patients who died in hospital could also contribute to these counts. AThis indicates a patient who died after rapid clinical deterioration for whom no specific rehabilitation-related needs (e.g., diet modification or gait training) were documented in the chart during hospitalization. ADL, activities of daily living. LTCI Certification Among the patients admitted from home (n=57), LTCI certification at discharge was more frequent among those with than without suspected dysphagia (67.9% [19/28] vs. 20.7% [6/29], respectively; Figure 3). SLP Intervention Overall, 4 patients received SLP intervention during hospitalization; patient-level details are provided in Supplementary Table 3. All 4 patients were male and screened positive for suspected dysphagia. Their ages were 79, 79, 88, and 97 years, and Hasegawa Dementia Scale–Revised scores ranged from 15 to 25 (of a possible total of 30, with lower scores indicating a greater likelihood of cognitive impairment). Triggers for SLP referral included physician-diagnosed pneumonia (including suspected aspiration pneumonia) in 3 patients and coughing/choking during meals observed by nurses in 1 patient. The time from admission to the first SLP session varied considerably (2–39 days), and the frequency of SLP sessions ranged from approximately 3–4 to 4–5 sessions/week. Changes in diet form during hospitalization were documented in patient medical records. Two patients transitioned from a regular diet to a texture-modified diet, one transitioned from a modified diet to a regular diet, and one remained on a modified diet. The admission BI ranged from 20 to 75 (on a 0–100 scale, with higher scores indicating greater independence in activities of daily living), and the ΔBI ranged from −45 to +5. Two patients were discharged to long-term care facilities, one was discharged home, and one died during hospitalization. Postintervention reassessment of EAT-10 or RSST was not systematically performed. Discussion Using 2 validated screening tools (EAT-10 and RSST), we identified suspected dysphagia in nearly half (49.2%) of inpatients aged ≥75 years undergoing CR. Suspected dysphagia in patients was associated with poorer physical function at admission, lower BI at discharge, and less frequent completion of discharge physical performance assessments. Collectively, these findings suggest that suspected dysphagia in elderly CR inpatients is associated with greater functional vulnerability and higher subsequent care needs. Association With Physical Function The observed relationship between reduced muscle strength and suspected dysphagia is related to the concept of “sarcopenic dysphagia.” Lower handgrip strength and smaller calf circumference may reflect generalized sarcopenia, and lower SPPB at admission suggests reduced global physical performance in patients with suspected dysphagia. As expected, handgrip strength was higher in men than in women, reflecting known sex differences in muscle strength; however, the direction of the association between suspected dysphagia and lower handgrip strength was similar in both sexes. Sex-stratified comparisons of handgrip strength by suspected dysphagia status did not reach statistical significance, possibly due to the limited sample size (Supplementary Table 1). At discharge, SPPB scores among patients who were able to complete the assessment did not differ significantly between groups; however, discharge physical performance assessments were not available for all patients, particularly among those with suspected dysphagia, which may indicate limited physiologic reserve and greater complexity of care needs in this population. Importantly, our study assessed suspected dysphagia based on screening rather than a diagnostic work-up; therefore, these findings should be interpreted as an association between suspected swallowing dysfunction and frailty-related measures. Nevertheless, the results align with previous reports indicating that diminished physical performance is linked to poorer swallowing-related outcomes in older populations and patients with cardiovascular disease.5–7,14 ADL Recovery and Discharge Outcomes Recent reports from Japan have further emphasized the importance of structured CR in older patients with cardiovascular disease, including its relevance to functional outcomes and ADL.15,16 Although admission BI was broadly comparable between the 2 groups, discharge BI was lower among patients with suspected dysphagia, suggesting an association between suspected swallowing dysfunction and limited functional recovery during hospitalization. Exploratory item-level BI analyses indicated that between-group differences in ADL were most apparent in transfer-related function (Supplementary Table 2, Sections A–C), which may be clinically relevant for discharge planning because transfer independence often influences the level of post-discharge support required. To more directly evaluate ADL recovery, we examined ∆BI and its associations with admission EAT-10 and RSST; however, no significant associations were found, and the results were consistent in sensitivity analyses excluding in-hospital deaths (Supplementary Table 2, Sections D and E). Among patients admitted from home (n=57), the composite outcome of non-home discharge or in-hospital death tended to be more frequent in those with suspected dysphagia (Figure 2). Although statistical significance was not reached and the 95% CIs were wide, the point estimate suggested a potentially higher risk of non-home discharge in patients with suspected dysphagia, warranting attention and further study. Previous evidence indicates that functional status is an important determinant of discharge destination in older inpatients.13 Moreover, more patients with suspected dysphagia could not complete physical performance assessments at discharge than those without, which may reflect limited reserve and greater complexity of care needs. Long-Term Care Burden LTCI certification at discharge was markedly more frequent in the group with than without suspected dysphagia (67.9% [19/28] vs. 20.7% [6/29] of patients admitted from home; Figure 3). Although LTCI certification is influenced by multiple factors beyond swallowing function alone, this finding suggests that suspected dysphagia was associated with higher care needs at discharge and may help identify patients who could benefit from early multidisciplinary planning for post-discharge support. Required Collaborative Partners and Reasons for Non-Home Discharge Beyond inpatient management, our findings underscore the need for coordinated post-acute collaboration for patients who screen positive for suspected dysphagia. Relevant partners may include SLPs and dietitians for swallowing and nutrition management; nurses and physicians for medical monitoring (e.g., aspiration risk and respiratory status); care managers for LTCI-related coordination; home-visit nursing services; outpatient or home-based rehabilitation providers; and long-term care facilities when home discharge is not feasible. Clinically, non-home discharge may reflect functional limitations (e.g., lower BI scores and an inability to complete mobility assessments) and the need for ongoing risk management and support related to suspected swallowing dysfunction, nutritional concerns, cognitive impairment, and caregiver capacity. Therefore, early identification of suspected dysphagia may facilitate earlier multidisciplinary discussions and more realistic planning for discharge destination and necessary services. To further contextualize these needs, we provide descriptive data on reasons for non-home discharge and commonly documented rehabilitation-related needs among the small subset of patients admitted from home who were not discharged home (Table 4). Clinical Implications in the Japanese Inpatient CR Setting In Japan, multidisciplinary dysphagia management within cardiac care programs may not be routinely integrated in all settings. In such contexts, simple bedside screening performed by non-SLP rehabilitation staff may facilitate timely multidisciplinary assessment and discharge planning when swallowing dysfunction is suspected, including nutrition support when indicated. Importantly, because these are screening tools, a positive result should be interpreted as suspected dysphagia, prompting further clinical evaluation rather than serving as a definitive diagnosis. SLP Intervention and Clinical Decision-Making After Screening In our cohort, SLP intervention was provided to a small subset of patients with suspected dysphagia (Supplementary Table 3). Referral triggers were pneumonia (including suspected aspiration pneumonia) or observed coughing/choking during meals, suggesting that clinical signs of aspiration risk prompted SLP referral in practice. After a positive screening result, a stepwise institutional clinical pathway was used to guide subsequent evaluation and management. Specifically, if either the EAT-10 or RSST was positive, or if coughing or choking during meals was observed by rehabilitation or ward staff, the finding was reported to the attending physician for further clinical review. Patients with reduced arousal or inability to follow simple commands were deferred and reassessed when clinically stable. When clinically indicated, a secondary swallowing assessment by an SLP was requested. Based on this assessment, patients were generally managed according to the following categories: (1) no clinically relevant swallowing dysfunction, in which case oral intake and routine monitoring were continued; (2) mild suspected swallowing dysfunction, in which case postural and mealtime guidance, oral hygiene support, and follow-up reassessment were provided; and (3) swallowing dysfunction requiring active management, for which diet or liquid modification and direct and/or indirect swallowing therapy by the SLP were considered, together with nutritional assessment or nutrition support team consultation as needed. However, because postintervention reassessment using the EAT-10 or RSST was not systematically performed, we could not evaluate changes in screening status attributable to SLP involvement. Policy and Clinical Implications Our findings support the clinical value of routine swallowing screening to identify suspected dysphagia among elderly inpatients undergoing CR. Implementing simple bedside screening early during hospitalization may facilitate timely multidisciplinary assessment and discharge planning, including coordination of nutrition support and post-acute services when indicated. Further multicenter studies are warranted to refine risk stratification and to determine whether structured, screening-triggered pathways are associated with better functional recovery and discharge-related outcomes. Study Limitations This study has several strengths, including prospective screening within a defined early hospitalization window and the use of 2 feasible screening tools that can be implemented by non-SLP rehabilitation staff. However, a few limitations warrant consideration. First, this was a single-center observational study with a limited sample size; therefore, the 95% CIs were wide for some outcomes. Given the limited sample size, multivariable regression modeling for clinical outcomes was not performed to avoid overfitting; hence, residual confounding cannot be excluded. Although serum albumin was included in the baseline characteristics, nutritional status, comorbidity burden, and overall clinical condition may have influenced the observed associations. Second, patients with severe dementia were excluded from the study. Because severe cognitive impairment may influence swallowing-related assessment and clinical presentation, this exclusion may have introduced selection bias and may have led to an underestimation of the prevalence of suspected dysphagia. Therefore, the present findings are most applicable to older inpatients undergoing CR who are able to participate in bedside swallowing screening, and may not be generalizable to patients with severe dementia or marked communication difficulty. Third, the EAT-10 and RSST are screening tools and do not provide a definitive diagnosis of swallowing dysfunction; hence, misclassification is possible. Fourth, postintervention reassessment of swallowing screening was not systematically conducted after SLP involvement, limiting evaluation of longitudinal changes in swallowing function. Fifth, LTCI certification reflects multiple clinical and social factors and should not be interpreted as solely attributable to swallowing dysfunction. Finally, multiple comparisons were conducted without formal adjustment; therefore, the findings should be interpreted cautiously and considered hypothesis-generating. Future studies with larger, multicenter cohorts and standardized diagnostic confirmation and follow-up assessments are warranted to refine risk stratification and evaluate whether structured screening-triggered pathways are associated with better functional recovery and discharge-related outcomes in this population. Conclusions A considerable proportion of inpatients aged ≥75 years undergoing CR screened positive for suspected dysphagia using the EAT-10 or RSST. Suspected dysphagia in patients was associated with poorer physical function and lower ADL at discharge. It was also related to greater care needs, including a tendency towards a higher frequency of non-home discharge or in-hospital death among those admitted from home and a markedly higher rate of LTCI certification at discharge. These findings suggest that suspected dysphagia identified by routine screening in this population is associated with clinical vulnerability and increased subsequent care needs. Hence, early screening by rehabilitation staff, coupled with timely multidisciplinary evaluation and coordinated discharge planning, may help address these needs in real-world inpatient CR settings. Disclosures The authors declare that there are no conflicts of interest. IRB Information The study protocol was approved by the Institutional Review Board of Kitajima Taoka Hospital, Medical Corporation Kitajima Izankai (Approval no. 2025-13). Written informed consent was obtained from all participants before enrollment. Supplementary Files Supplementary Table 1. Supplementary Table 2. Supplementary Table 3.