Secondary Logo

Journal Logo

Special Topic: Caregiving

Decreasing Heart Failure Readmissions Among Older Patients With Cognitive Impairment by Engaging Caregivers

Agarwal, Kathryn S. MD; Bhimaraj, Arvind MD, MPH; Xu, Jiaqiong PhD; Bionat, Susan DNP, ACNP-BC; Pudlo, Michael APRN, ACNP-BC; Miranda, David MD; Campbell, Claire MD; Taffet, George E. MD

Author Information
The Journal of Cardiovascular Nursing: 5/6 2020 - Volume 35 - Issue 3 - p 253-261
doi: 10.1097/JCN.0000000000000670
  • Free


Heart failure (HF) is the most common reason for hospital admission for older adults.1,2 After hospital discharge, the frequency of readmission within 30 days ranges from 15% to 30% with annual estimated Medicare costs of 30 billion dollars.1,3 Heart failure readmissions are used as a quality measure, and extensive efforts are made to manage patients with HF and prevent their frequent, expensive readmissions.

Cognitive impairment (CI) is very common in older patients with HF and has a profound impact on HF outcomes.4,5 Heart failure patients with unrecognized CI have increased readmission and 6-month mortality rates.6 Older patients with CI were much more likely to be readmitted within 30 days of discharge than those with intact cognition. We found that patients older than 70 years with normal Mini-Cog scores of 4 or 5 had 50% lower readmission rates compared with those with abnormal Mini-Cog scores less than 4.5

Heart failure management is complex and requires high levels of executive cognitive function to comply with postdischarge directives.7,8 Executive dysfunction may impair a patient's ability to make decisions in complex situations, recognize symptoms, and implement action plans.6 Cognitive impairment, in both memory and executive function, may be a marker for more severe HF and is also associated with lesser adherence to HF medication regimens.9 Evidence that readmissions are a manifestation of inability to self-manage is supported by the sizable jump in hospitalization rate observed shortly after discharge from post–acute care facilities where medical care is managed by facility staff.4 Whereas self-care and patient engagement are considered as formulae for success in patients with HF, factoring in the impact of CI has been underappreciated in HF readmission reduction programs.

As part of a nurse-driven quality improvement initiative to supplement our Project BOOST (Better Outcomes for Older Adults through Safe Transitions) care transition program of enhanced discharge education, nurses performed Mini-Cog screens for CI before discharge. If there was evidence of CI on Mini-Cog screening, providing HF education including self-management strategies to family members or caregivers before the patients' discharge was encouraged. Caregiver inclusion was not always accomplished for those with and without CI, providing an opportunity to address 2 hypotheses: (1) CI is a marker of more severe HF, driving increased readmissions, and (2) including caregivers in nursing discharge education can reduce readmission by providing additional “cognitive” support to patients regardless of severity of HF. We looked for associations between standard measures of HF severity and Mini-Cog scores and those who had caregiver education at discharge. We provide preliminary evidence that identification of CI and involvement of family and caregivers provide an immense opportunity to improve HF care and reduce readmissions. We suggest that personalized care for those with impaired cognition as well as caregiver engagement is the “next frontier in hospital medicine.”10,11 These 2 steps seem critical to improve HF outcomes and should be studied in future trials.


The institutional review board for human subject research for Baylor College of Medicine reviewed and approved this research protocol by expedited procedures, and then it was approved by administrative review by the institutional review board for Houston Methodist Hospital. The protocol was approved with a waiver of consent and HIPAA (Health Insurance Portability and Accountability Act) authorization.

Setting and Study Population

The population analyzed included 2 cohorts of patients 70 years or older discharged from hospital HF units with a primary diagnosis of HF during calendar year 2014 (2014 cohort) or from December 2015 to May 2016 (2015 cohort). In 2014, a daily report permitted identification of those pending discharge in the next 48 hours to home, and potential participants were referred to collaborating nurses, who administered the Mini-Cog after verbal assent. During this time, 121 patients with HF older than 70 years discharging to home were evaluated. In 2015, a heart failure disease management (HFDM) service led by advanced practice nurses (APNs) was developed to standardize HF care across the hospital, including education, guideline-directed medical therapy optimization, and arrangement of a 7-day postdischarge appointment. The HF service is composed of HF trained APNs under the guidance of an advanced HF physician. As part of the HFDM service evaluation, a Mini-Cog test was performed before hospital discharge in 112 patients with HF older than 70 years, identifying patients with CI at risk for readmission, and this group also included 31 patients discharged to skilled nursing facilities. In both cohorts, Mini-Cog testing was only completed on English-speaking patients who were alert and physically able to write and to understand the instructions. Only patients 70 years or older were included because of the authors' involvement in another quality improvement program focused on this population in the hospital.

From 2012 to 2016, the HF units used an enhanced discharge education structure—Project BOOST (Better Outcomes for Older Adults through Safe Transitions)—to reduce 30-day readmissions by improving care transitions, and elements of Project BOOST are still ongoing.12 Key elements of BOOST are enhanced education including having the patient “teach back” information to nurses, individualized care transition tools to help patients recognize symptoms necessitating medical attention, and follow-up phone calls from nurses. In addition to Project BOOST, in the 2015 cohort, the HFDM APNs gave supplemental education to patients and caregivers. The HFDM service focused on education specific to HF guideline-directed medical therapy and self-care. The elements of Project BOOST were applied to all patients on the HF units during this study period; however, formal screening for CI and active efforts to engage caregivers in the discharge education were only part of the process for patients included in this quality improvement program.


If a patient scored less than 4 on the Mini-Cog screen, the nursing staff on the floor was directed to include caregivers in discharge education and planning. In both cohorts, neon green stickers were placed in the chart stating, “Family First. Please include caregivers in all discharge planning and education.” In the 2015 cohort, if the APNs identified patients with CI, they tried to meet with family/caregivers and patients to give supplemental education while staff nurses continued including caregivers with “Family First.” Staff nurses documented in the electronic medical record either “education given to patient” or “education given to patient and family/caregiver(s).” As a quality improvement program, hoping to include all caregivers in discharge education, the inclusion of caregivers in education was not randomized but based on nurses' ability to include them in the process.

A diagnosis of CI or dementia was not coded or added to the problem list of a patient based on the Mini-Cog result. The Mini-Cog result was transmitted to the primary care physician and to the HF clinic for follow-up at the time of discharge. The focus of this intervention was on the need for additional support for patients at discharge, not diagnostic workup, and later to repeat and follow up on the abnormal Mini-Cog results.

Mini-Cog Screening and Data Retrieval

The Mini-Cog13 combines a memory test (recall 3 unrelated words) with a clock-drawing test.14 The time chosen for the clock-drawing test (11:10) and the words (“river,” “nation,” “finger”) were selected based on previous studies.15 Trained nurses or the APNs graded the participants on a 5-point scale: 2 points for a normal clock-drawing test (no partial credit) and 1 point for each of the 3 words recalled after a 3-minute delay given to draw the clock. The clock-drawing test scores were reviewed by the investigators. We used the cutoff score of 415; scores less than 4 were designated as abnormal or consistent with CI. The initial validation of the Mini-Cog cites a sensitivity of 76% and a specificity of 89% for dementia using a cutoff score of 3 compared with conventional neuropsychiatric testing.16 However, in McCarten et al's15 study, 55% of those with scores of 3 had significant CI on further testing. Thus, in hoping to detect less obvious cases of CI that may experience difficulty in self-care, we chose scores less than 4 to be considered abnormal. A normal score of 4 or 5, not consistent with CI, required drawing a normal clock, its measure of executive function.

Patients were excluded from data analysis if they had a nonprimary HF diagnosis, did not have a Mini-Cog score, were younger than 70 years, had a cardiac transplant or ventricular assist device, or were assigned to hospice at discharge.

Investigators reviewed each patient's medical record for documentation of caregiver/family inclusion in discharge education, readmission within 30 days of discharge to any of 5 system hospital locations, and other data, including history of atrial fibrillation or hemodialysis and components of the Readmission Risk Score (RRS).17,18 The RRS was developed using a claims-based model for patients with HF for medical record review to estimate risk-standardized hospital readmission rates.18 The RRS includes comorbidities, laboratory results, and vital signs on admission to predict readmission for older patients with HF (mean age, 79.9 years).17,18 The components of the RRS were extracted from the charts as recommended and entered into the online RRS to obtain a risk score ( The elements extracted from the chart and entered into the RRS include age, sex, in-hospital cardiac arrest, vital signs on admission, laboratory results on admission, left ventricular ejection fraction (LVEF), or history in admission notes with evidence of diabetes mellitus, HF, coronary artery disease, previous percutaneous coronary intervention, aortic stenosis, previous stroke, chronic obstructive pulmonary disease, or dementia. Brain natriuretic peptide (BNP) level on admission was also recorded as another measure of HF severity. Only 20 patients (8.5%) had a history of dementia in their admission notes, and these were included in data analysis. The online RRS yields percentage risk for 30-day readmission based on HF severity, and this was recorded for data analysis.

Statistical Analysis

Data were presented as mean ± SD for continuous variables and number (percentage) for categorical variables by normal and abnormal Mini-Cog screens or taught family and not taught family among those patients with abnormal Mini-Cog scores. P value was based on the Student t test or the Mann-Whitney test for continuous variables and χ2 test or Fisher exact test for categorical variables, whenever appropriate. Readmission risk score, BNP, and LVEF were presented as median (interquartile ranges) by Mini-Cog scores, and Kruskal-Wallis rank test was used to test the difference among Mini-Cog scores. The nonparametric test for trend across ordered groups of Mini-Cog scores19 was used to test the relationships between RRS, BNP, and LVEF and Mini-Cog scores. Univariate and multivariate logistic regression models were applied to assess the association of RRS, ejection fraction, abnormal Mini-Cog score, and nurse documentation of education with caregivers with 30-day readmission. The Hosmer-Lemeshow goodness-of-fit test was used to determine how well the multivariate logistic regression model was fitted. All analyses were performed with STATA version 15 (2017, StataCorp LLC; College Station, Texas). Statistical significance was 2-tailed and defined as P < .05 for all tests.


Altogether, 323 patient encounters were evaluated within the 2 combined cohorts. A total of 91 patients were excluded: 84 did not have documented Mini-Cog results, and 7 did not have documentation whether caregivers were or were not included in discharge education. The study includes the remaining 232 patients 70 years or older admitted for HF; within this group, 37% (n = 85) had a Mini-Cog score of 4 or 5, which we considered normal, and 63% (n = 147) had scores less than 4, which were considered abnormal or consistent with CI. Table 1 documents the baseline characteristics of the groups with and without CI as evidenced by their Mini-Cog score. There were no differences in baseline medical characteristics including LVEF, laboratory results on admission, and comorbidities except for average age being higher and more patients with a history of stroke in the abnormal Mini-Cog group. The calculated RRS score was not different between the 2 groups. The BNP at the time of index admission was slightly higher in those with abnormal Mini-Cog scores. The relationships between the Mini-Cog scores and the RRS, BNP, and LVEF are shown in Table 2. There was no significant difference between the RRS and LVEF for different Mini-Cog scores; however, there was a trend toward higher BNP levels on admission in patients who have lower Mini-Cog scores (P for trend = .02). This suggests that greater impairment in cognition was not associated with increased measures of readmission risk but potentially worse severity of HF on admission.

Baseline Characteristics Comparing Patients With Normal vs Abnormal Mini-Cog Screens
Median (Interquartile Range) of Readmission Risk Score, Brain Natriuretic Peptide (pg/mL), and Left Ventricular Ejection Fraction (%) by Mini-Cog Score

The 30-day readmission rate was 14.1% in the group with normal cognition compared with 23.8% for those with abnormal Mini-Cog (P = .09) (Table 1). These groups included 31 patients discharged to skilled nursing facilities where the HF management would be performed by the treating team. Of this small group discharged to facilities, 25 of the 31 patients had a Mini-Cog score less than 4 and had a 30-day readmission rate of 4% (1/25). If we excluded those discharged to facilities (n = 31), those with abnormal Mini-Cog scores had a 26.8% (33/123) readmission rate compared with 14.1% in those with normal Mini-Cog scores (11/78) (P = .04).

In the patients with normal Mini-Cog scores, family teaching in 35 of the 85 patients had no effect in readmissions compared with those without teaching (14.3% vs 14.0% 30-day readmission rate, P = 1.0). Thus, we kept normal Mini-Cog scores without evidence of CI as 1 group and analyzed those with CI as 2 groups: with and without caregivers included in education. Of those with abnormal Mini-Cog scores or CI, 59.2% (n = 87) had documentation of caregivers included in discharge education, and 40.8% (n = 60) had documentation that caregivers were not included (Table 3). The patients with caregiver inclusion were older, had lower Mini-Cog scores and a higher blood urea nitrogen level on admission, and more had a diagnosis of dementia. In regard to measures of HF severity such as the RRS and BNP levels, there were no differences between those with CI who had caregivers included versus those who had not, as shown in Table 3; however, the readmission rate was more than double (35.5%) for those in whom caregivers were not taught compared with those where the family was included (16.1%) (P = .01). The Figure illustrates the readmission rates for patients based on the presence of CI and whether caregivers were included in education for those with CI.

Baseline Characteristics of Patients With Abnormal Mini-Cog Screens With and Without Caregiver Education
This figure is the central illustration, because it illustrates the readmission rates for heart failure patients with cognitive impairment (CI) and no caregiver involvement in education (35.5%), those with CI with caregiver inclusion in education (16.1%), and those without CI (14.1%). The inclusion of caregivers in discharge education was associated with readmission rates similar to those without CI.

If we excluded all patients who had family teaching, patients with CI or an abnormal Mini-Cog score less than 4 had a 33% readmission rate (20/60) versus a 14% readmission rate for those with a normal Mini-Cog score (7/50) (P = .03). The readmission rate for those with abnormal Mini-Cog scores where the family member was taught was similar to those with normal cognition (Figure).

The association of risk factors with 30-day readmission is presented in Table 4. By multivariate analysis, the factors associated with hospital readmissions were abnormal Mini-Cog screen (odds ratio, 2.23; 95% confidence interval, 1.06–4.68; P = .03) and nurse-documented education with family (odds ratio, 0.46; 95% confidence interval, 0.24–0.90; P = .02). The RRS and ejection fraction (LVEF) were not associated with readmission in this older population. The model was fitted well (Hosmer-Lemeshow P = .44).

Odds Ratio and 95% Confidence Interval for 30-Day Readmission

Because the 2 cohorts had education provided by different levels of nurses (registered nurse [RN] vs APN and RN combined), we studied outcomes of the patients with CI with education between the 2 cohorts. In the first cohort, there were 81 patients with abnormal Mini-Cog scores, 35 had teaching (43% of the patients with CI had caregiver teaching), and of those, 6 were readmitted—a readmission rate of 17.1%. In the second cohort, there were 66 patients with abnormal Mini-Cog scores, 52 received teaching (78.8% of the patients with CI had caregiver teaching), and of those, 8 were readmitted in less than 30 days, for a 15.4% readmission rate. The readmission rates between the 2 cohorts who had abnormal Mini-Cog scores and had discharge education by RN versus a combination of RN and APN had no statistical difference. The success rate of having teaching performed was significantly better with the combined efforts of the RN and APN (78.8% vs 43.2%, P < .001).


Heart failure is a chronic, systemic condition contributing significant morbidity to older adults. The complexity of HF management produces a tremendous burden for medical self-management compiled with the multiple comorbidities these patients have. Although 30-day all-cause readmission rates after hospitalization for an acute HF exacerbation have become a hospital metric associated with fiscal accountability, no single strategy to reduce the rate seems to be effective.20 Previous studies have established that the prevalence of CI ranges from 40% to 75% depending upon criteria and HF population studied.21,22 In this study, CI was present in 63% of individuals with HF at hospital discharge (Mini-Cog score < 4). Thus, in hospitalized patients older than 70 years, CI is a factor in self-management for most patients with HF.

Patients admitted for HF with cognitive dysfunction have been reported to have 5-fold higher in-hospital mortality and increased 1-year mortality.23 In our sample of older patients, lower Mini-Cog scores were associated with a trend toward higher BNP levels but were not associated with lower ejection fraction or greater RRS. In Tables 1 and 2, there was no relationship between these parameters and the Mini-Cog scores analyzed in comparison of normal versus abnormal, nor by multivariate analysis (Table 4) were these measures associated with increased odds of readmission.

The main finding of our program is that patients with HF and CI had significantly lower 30-day readmission rates when their families or caregivers were involved in the discharge education process, providing an auxiliary support to aid in HF management. Patients with caregiver inclusion in education had readmission rates not different from those with normal Mini-Cog scores. Even in a healthcare enviroment with enhanced transitions programs, CI can affect postdischarge outcomes for patients with HF. Cognitive impairment has been identified as a marker for readmission risk in HF4 and may be identified in a standardized fashion before discharge prompting transition support, including engagement of key family and caregivers as care partners.

Other groups have recognized the importance of caregiver involvement in the care of patients with HF.24 Some groups target the dyad of patient with HF and caregiver as the mode of analysis. In these studies, there has been little attention to the patients' cognition. We found that involving the caregiver had little impact on readmission rate if the patient had a Mini-Cog score of 4 or 5. Thus, depending upon the prevalence of CI, dyad-focused interventions may or may not have been efficacious.25,26 By targeting only those with CI at discharge, efforts could be focused on those more likely to benefit from caregiver inclusion and precious resources conserved. In fact, the HFDM service found that the additional time required to schedule and educate caregivers was challenging. When new clinical demands and a new electronic medical record came that increased the challenge further, the Mini-Cog screening and “Family First” program were discontinued.

Many readmission reduction programs are education based,12,27,28 typically focused on patient knowledge and skills. These programs are less likely to be efficacious in patients with CI. In HF patients with CI, their deficit in executive function may be manifested in difficulty with recognizing changes in status, that is, weight gain, or actualization of an appropriate response to these changes.29 Several studies have highlighted the role of executive dysfunction in patients with vascular disease and HF, yielding difficulty with planning, implementation of plans, and risk-benefit decisions.7,30,31 The clock-drawing test in the Mini-Cog screen assesses executive function, and the inability to draw the clock is associated with risk of poor self-care.32 Clearly, intact memory is also required for medical self-management, and memory function may be tested with “teach-back education.” However, we think the underappreciation of the role of executive function has resulted in educational efforts that improve repetition of information but not outcomes such as readmission rates in older adults with CI.33

As a quality improvement initiative, there are many limitations to the study, and thus, this study is mainly hypothesis generating. These limitations include the lack of random assignment of patients to family/no family discharge education, because we hoped that all patients with CI would have caregiver inclusion. Differences in family availability and less caregiver support may explain lack of family inclusion at discharge and readmissions, confounding interpretation of results. However, more patients with CI had caregiver inclusion than those without CI. Fifty percent (43/85) of patients without CI did not have caregiver education, whereas only 41% (60/147) of patients with CI did not have caregiver education. Most patients were “Medicare fee for service” patients with physicians in multiple venues of care; thus, information about proportions who had successful postdischarge calls and early postdischarge follow-up visits was not available.

As a quality improvement program, within the confines of normal hospital care focused on the patient, a Min-Cog screen for CI was chosen to be practically applied by busy practitioners; however, the Mini-Cog is designed to be a screening tool and not a diagnostic tool. In addition, no additional information was obtained about the caregivers during discharge education, and review of medical records did not facilitate any characterization of the caregivers. Others have characterized the caregivers and found factors that suggested barriers impeding support to the patient with HF. These included significant health problems and incorrect understanding of HF management including medications.34 Almost half of caregivers in those studies felt that their HF knowledge was inadequate.35

Additional limitations include inclusion of only English-speaking patients and the lack of documentation of the exact education given. There were differences in the educational interventions between the 2 cohorts because the second cohort had APNs in addition to the RNs providing education; however, this seemed to only increase the likelihood of receiving the education but was not associated with an improved readmission rate for the cohort with APNs and RNs providing education. In the busy hospital setting, the specific discharge education content provided by the nurse or APN was not documented; the content was guided by BOOST and the team's HF expertise. Other limitations of this study include the retrospective collection of HF severity data and that we could only assess for death in the 30 days after admission as per our electronic medical record or readmission to 1 of our 5 hospitals; however, these readmissions to other hospitals are likely to be evenly distributed across study groups.


In summary, as part of a nurse-driven quality improvement initiative, 63% of 232 hospitalized patient with HF 70 years or older had CI as evidenced by a Mini-Cog score of less than 4. There was no difference in readmission risk between those with CI and those with normal cognition as evidenced by RRS or ejection fraction. However, patients with CI had higher BNP levels on admission than those with normal cognition. When those with CI were divided between those who had nursing discharge instruction provided only to the patient and those who had a family member or caregiver involved, the rates of readmission were dramatically lower in those where the family was included. Heart failure hospital readmissions reduction strategies have been implemented at great effort, but little attention has been given to the cognitive abilities of patients, especially those older than 70 years, to actualize these strategies. Creating systems that recognize CI in patients with HF and implement solutions is potentially very valuable to the provision of patient-centered care and to decrease the readmission burden. We hope that this information may spark future randomized trials to address how inclusion of caregivers may support improved outcomes for patients with CI. In this quality improvement program, we have shown that, by engaging caregivers in education, nurses helped counter the contribution of CI to HF readmission risk and reduced the readmission rate to that of patients without CI.

What’s New and Important

  • Sixty-three percent of 232 patients with HF 70 years or older had evidence of CI at hospital discharge.
  • In patients with CI, nurses involving caregivers in education decreased 30-day readmissions from 35% to 16%.
  • In older patients with HF, identifying CI and including caregivers may reduce readmissions.


The authors thank the Care Navigator Program and the nursing staff on the units for their active participation. They appreciate the mentorship of Nancy Wilson and Senior Leaders in the John A. Hartford Foundation Practice Change Leaders program. They thank Soo Borson for sharing the Mini-Cog and her advice. They also thank Jason Taffet, Reina Styskel, and Mohammed Al-Jumayli for data collection.


1. Heidenreich PA, Albert NM, Allen LA, et al. Forecasting the impact of heart failure in the United States: a policy statement from the American Heart Association. Circ Heart Fail. 2013;6(3):606–619.
2. Russo CA, Elixhauser A. Hospitalizations in the Elderly Population, 2003: Statistical Brief #6. Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Rockville, MD: Agency for Healthcare Research and Quality (US); 2006.
3. McHugh MD, Chenjuan M. Hospital nursing and 30-day readmissions among Medicare patients with heart failure, acute myocardial infarction, and pneumonia. Med Care. 2013;51(1):52–59.
4. Patel A, Parikh R, Howell EH, Hsich E, Landers SH, Gorodeski EZ. Mini-cog performance: novel marker of post discharge risk among patients hospitalized for heart failure. Circ Heart Fail. 2015;8(1):8–16.
5. Agarwal KS, Kazim R, Xu J, Borson S, Taffet GE. Unrecognized cognitive impairment and its effect on heart failure readmissions of elderly adults. J Am Geriatr Soc. 2016;64(11):2296–2301.
6. Dodson JA, Truong TT, Towle VR, Kerins G, Chaudhry SI. Cognitive impairment in older adults with heart failure: prevalence, documentation, and impact on outcomes. Am J Med. 2013;126(2):120–126.
7. Cameron J, Worrall-Carter L, Page K, Riegel B, Lo SK, Stewart S. Does cognitive impairment predict poor self-care in patients with heart failure? Eur J Heart Fail. 2010;12(5):508–515.
8. Lee CS, Moser DK, Lennie TA, Riegel B. Event-free survival in adults with heart failure who engage in self-care management. Heart Lung. 2011;40(1):12–20.
9. Dolansky MA, Hawkins MA, Schaefer JT, et al. Association between poorer cognitive function and reduced objectively monitored medication adherence in patients with heart failure. Circ Heart Fail. 2016;9(12):e002475.
10. Coleman EA. Family caregivers as partners in care transitions: the caregiver advise record and enable act. J Hosp Med. 2016;11(12):883–885.
11. Horstman MJ, Salas E. Web exclusive. Annals for hospitalists inpatient notes—caregiver engagement in hospital medicine—the next frontier. Ann Intern Med. 2019;170(2):HO2–HO3.
12. Society of Hospital Medicine. Project BOOST Mentored Implementation Program. Philadelphia, PA: Society of Hospital Medicine; 2015.
13. Borson S, Scanlan J, Brush M, Vitaliano P, Dokmak A. The mini-cog: a cognitive 'vital signs' measure for dementia screening in multi-lingual elderly. Int J Geriatr Psychiatry. 2000;15(11):1021–1027.
14. Scanlan J, Borson S. The Mini-Cog: receiver operating characteristics with expert and naive raters. Int J Geriatr Psychiatry. 2001;16(2):216–222.
15. McCarten JR, Anderson P, Kuskowski MA, McPherson SE, Borson S. Screening for cognitive impairment in an elderly veteran population: acceptability and results using different versions of the Mini-Cog. J Am Geriatr Soc. 2011;59(2):309–313.
16. Borson S, Scanlan JM, Chen P, Ganguli M. The Mini-Cog as a screen for dementia: validation in a population-based sample. J Am Geriatr Soc. 2003;51(10):1451–1454.
17. Sudhakar S, Zhang W, Kuo YF, Alghrouz M, Barbajelata A, Sharma G. Validation of the readmission risk score in heart failure patients at a tertiary hospital. J Card Fail. 2015;21(11):885–891.
18. Keenan PS, Normand SL, Lin Z, et al. An administrative claims measure suitable for profiling hospital performance on the basis of 30-day all-cause readmission rates among patients with heart failure. Circ Cardiovasc Qual Outcomes. 2008;1(1):29–37.
19. Cuzick J. A Wilcoxon-type test for trend. Stat Med. 1985;4:87–90.
20. Kociol RD, Peterson ED, Hammill BG, et al. National survey of hospital strategies to reduce heart failure readmissions: findings from the Get With the Guidelines-Heart Failure registry. Circ Heart Fail. 2012;5(6):680–687.
21. Hanon O, Vidal JS, de Groote P, et al. Prevalence of memory disorders in ambulatory patients aged >/=70 years with chronic heart failure (from the EFICARE study). Am J Cardiol. 2014;113(7):1205–1210.
22. Albabtain M, Brenner MJ, Nicklas JM, et al. Hyponatremia, cognitive function, and mobility in an outpatient heart failure population. Med Sci Monit. 2016;22:4978–4985.
23. Zuccala G, Pedone C, Cesari M, et al. The effects of cognitive impairment on mortality among hospitalized patients with heart failure. Am J Med. 2003;115(2):97–103.
24. Toukhsati SR, Hare DL. Towards optimal heart failure care: couples-oriented strategies to improve patient adherence and health outcomes. Curr Cardiol Rev. 2016;12(3):243–248.
25. Lyons KS, Gelow JM, Hiatt SO, et al. The role of dyadic confidence on engagement in heart failure care behaviors. Gerontologist. 2018;58(4):635–643.
26. Deek H, Chang S, Newton PJ, et al. An evaluation of involving family caregivers in the self-care of heart failure patients on hospital readmission: randomised controlled trial (the FAMILY study). Int J Nurs Stud. 2017;75:101–111.
27. Patel PH, Dickerson KW. Impact of the implementation of project re-engineered discharge for heart failure patients at a Veterans Affairs Hospital at the Central Arkansas Veterans Healthcare System. Hosp Pharm. 2018;53(4):266–271.
28. Coleman EA, Parry C, Chalmers S, Min SJ. The care transitions intervention: results of a randomized controlled trial. Arch Intern Med. 2006;166(17):1822–1828.
29. Dolansky MA, Hawkins MAW, Schaefer JT, et al. Cognitive function predicts risk for clinically significant weight gain in adults with heart failure. J Cardiovasc Nurs. 2017;32(6):568–575.
30. O'Sullivan M, Morris RG, Markus HS. Brief cognitive assessment for patients with cerebral small vessel disease. J Neurol Neurosurg Psychiatry. 2005;76(8):1140–1145.
31. Pugh KG, Kiely DK, Milberg WP, Lipsitz LA. Selective impairment of frontal-executive cognitive function in African Americans with cardiovascular risk factors. J Am Geriatr Soc. 2003;51(10):1439–1444.
32. Reynolds NA, Ski CF, McEvedy SM, Thompson DR, Cameron J. Construct validity of the Heart Failure Screening Tool (Heart-FaST) to identify heart failure patients at risk of poor self-care: Rasch analysis. J Adv Nurs. 2018;74(6):1412–1422.
33. Awoke MS, Baptiste DL, Davidson P, Roberts A, Dennison-Himmelfarb C. A quasi-experimental study examining a nurse-led education program to improve knowledge, self-care, and reduce readmission for individuals with heart failure. Contemp Nurse. 2019;55(1):15–16.
34. Durante A, Paturzo M, Mottola A, Alvaro R, Vaughan Dickson V, Vellone E. Caregiver contribution to self-care in patients with heart failure: a qualitative descriptive study. J Cardiovasc Nurs. 2019;34(2):E28–E35.
35. Lee AA, Piette JD, Heisler M, Janevic MR, Langa KM, Rosland AM. Family members' experiences supporting adults with chronic illness: a national survey. Fam Syst Health. 2017;35(4):463–473.

cognitive disorders; elderly; heart failure; hospital readmissions

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.