Cognition in heart failure: An overview of the concepts and their measures

Abstract

Purpose

To review cognitive impairment and explore current measurement concerns faced by nurse practitioners caring for individuals with heart failure.

Data sources

Review of peer-reviewed research articles published on the topic.

Conclusions

Cognitive impairment is prevalent among individuals with heart failure. Impairment frequently involves one or more domains, including attention, memory, and executive function. No gold standard screening measure was identified from the reviewed literature.

Implications for practice

It is imperative that clinicians are aware of cognitive impairment and its implications for their patients with heart failure. Cognitive impairment likely contributes to multiple clinical implications, including a decreased ability to attend to and comprehend patient education materials and an inability to appropriately assess and self-manage symptoms.

Approximately 28%–58% of individuals with heart failure (HF) suffer from cognitive impairment, commonly identified as difficulty with concentration and/or memory (Bennett & Sauve, 2003; Pressler, 2008; Vogels, Scheltens, Schroeder-Tanka, & Weinstein, 2007). The clinical significance of cognitive impairment is probably best recognized when one examines the decreased quality of life, increased mortality, and utilization of healthcare resources associated with HF (Bosworth et al., 2004; Cameron et al., 2010; Pressler et al., 2010; Sauve, Blankenbiller, Lewis, & Bennett, 2002). While researchers have made great strides in understanding the etiology and treatment of HF, healthcare costs related to HF have increased more than 170% over the past decade, with costs for 2010 estimated at over $39 billion (American Heart Association [AHA], 2010). Hospital readmission accounts for a large portion of the cost (Wenger, 2008). Even with improved discharge planning and patient education, 25% of older persons discharged from the hospital with a diagnosis of HF are readmitted within 30 days (AHA2010). Inability to follow complex regimens and recognize worsening symptoms are identified as reasons for readmissions (Bennett, Pressler, Hays, Firestine, & Huster, 1997; Happ, Naylor, & Roe-Prior, 1997; Naylor, Stephens, Bowles, & Bixby, 2005; Wenger, 2008).

Cognitive impairment likely contributes to the inability to manage complex medical regimens, recognize worsening symptoms, and avoid frequent hospitalizations (Bauer et al., in press; Dickson, Deatrick, & Riegel, 2008; Dickson, Tkacs, & Riegel, 2007). For example, individuals with HF are taught by medical providers to monitor daily weights and breathing status. If a weight gain of greater than two pounds or an increase in shortness of breath occurs, they are instructed to increase diuretic dosage and/or contact their medical provider. Many individuals with HF are not able to accurately process information related to symptoms and do not increase medication dosage or contact medical providers. This situation is common in the HF population; however, the mechanisms behind information-processing difficulties are not clear.

Background

Cognition is a complex system involving multiple brain processes that allow an individual to perceive information (from both the internal and external environment), learn and remember specific information, and finally, use information previously processed to reason or problem solve in novel situations (i.e., behavior; Lezak, 2004; Rosenzweig, Breedlove, & Watson, 2005). The information-processing sequence is categorized into more specific tasks, referred to as cognitive domains (see Table 1; Lezak, 2004; Strauss, Sherman, & Spreen, 2006). The cognitive domains generally measured during a standard neuropsychological assessment are: attention, working memory, immediate and delayed memory, learning, speech/language, visual/spatial/constructional, and executive function (Lezak; Strauss et al., 2006). Table 2 illustrates representative neuropsychological assessment tests, as well as the task(s) involved for each cognitive domain.

Table 1.

Working definitions for commonly measured cognitive domains

Domain	Working definition
Attention	Ability to focus on certain stimuli while blocking out extraneous stimuli
Working memory	Ability to encode data and then perform some other mental operation with that data
Learning	Involves the repetition of stimuli that encourages further mental processing of the stimuli
Memory	Ability to recall stimuli
Speech/language	Symbols used to exchange information
Visual/spatial/constructional	Ability to perceive, conceptualize, and then respond with some motor output to visual stimuli
Executive function	Ability to decide, plan, implement, and evaluate goal-directed behavior; especially in novel situations

Table 2.

The cognitive domains evaluated and tasks involved with common neuropsychological instruments

Cognitive domain	Neuropsychological test	Task involved
Attention	Trail Making Test Part A	Connecting a series of numbers (similar to a dot-to-dot puzzle)
	Attentional Matrices	Identifying the missing pattern piece in a series; difficulty increases over trials
Working memory	Digit Span Backwards	Repeating a series of numbers in the reverse order they were stated by the examiner (progressive trials increase the numbers to be repeated by one number per trial)
	Corsi’s block tapping test	Tapping on a series of blocks in the order they were tapped by the examiner (progressive trials increase number of blocks that are tapped by one per trial)
	Short Category Test	Determining the underlying organizational principle that relates all the figures/lines, etc. in a given sequence
Memory/learning	Verbal learning test	Listening to a sequence of unrelated words and then repeating back as many words as one can remember over a series of multiple trials
Executive function	Trail Making Test Part B	Connecting a series of numbers alternating with letters (similar to a dot-to-dot puzzle except that the connections run in order of 1 to A, A to 2, 2 to B, B to 3, 3 to C and so on)
	Letter fluency	Listing as many words that begin with a specific letter in 1 min
	Raven’s Progressive Matrices	Similar to Attentional Matrices: Identify the missing piece of a segment that will complete a larger pattern; difficulty increases as the trials progress
	Intra–extra dimensional set shift	Determining which stimulus is correct given specific examiner feedback; after six correct responses, the stimulus rules change
	Stockings of Cambridge	Copy patterns of colored balls; determining number of moves needed to copy a pattern of balls
Speech/language	Controlled oral word association	Listing as many words that begin with the letters “F,” “A,” and “S” over three 1-min trials
Visual/spatial/constructional	Complex figure drawing/copying	Copying a complex figure and then drawing the complex figure from memory after 10 to 15 min of “distraction” tests

Cognitive impairment

Definitions and assessment

Researchers have documented that cognition and behavior are tightly linked and that even mild cognitive impairment can affect an individual’s ability to function effectively in daily life (Wadley, Okonkwo, Crowe, & Ross-Meadows, 2008). Assessment of mild cognitive impairment is completed by interpreting neuropsychological test scores based on published normative data as well as validating test scores against functional measures that are meaningful within a specific population. For instance, measures of executive function may be correlated with ability to successfully return to work in a brain-injured population (Lezak, 2004; Strauss et al., 2006). It is this combination of information that allows neuropsychologists to determine the clinical significance of a test score. In general, within the neuropsychological literature, mild cognitive impairment is operationally defined as cognitive test performance that falls one to two standard deviations below expected normative performance. It is at that level of performance where functional status (e.g., ability to self-manage HF treatment regimens) declines, even though this subtle degree of impairment is difficult to identify during routine clinical visits.

HF literature has used inconsistent operational definitions of cognitive impairment. Furthermore, clinical significance of neuropsychological instruments has been rarely reported (Bauer et al., in press). Overall test scores from individuals with HF were compared to small samples of age-matched “healthy” controls and statistically significant differences in the mean group scores (e.g., p ≤ .05 or .01) were defined as cognitive impairment (Almeida & Tamai, 2001; Schmidt, Fazekas, Offenbacher, Dusleag, & Lechner, 1991; Serber et al., 2008; Trojano et al., 2003; Vogels et al., 2007). However, in some studies scores achieved by both groups were in the normal range (Vogels et al., 2007). The inconsistent definitions of cognitive impairment used in the current research literature and lack of sensitive screening instruments make it difficult for clinicians to identify HF patients with or at risk for cognitive impairment during routine visits. Further complicating the clinician’s ability to identify cognitive impairment within the HF population are issues related to the etiology of cognitive impairment.

Etiology

The etiology of cognitive impairment in the HF population is unknown. Current literature describes the contributions of hypoperfusion, secondary to the HF disease process as a likely basis for cognitive impairment. The disease process of HF is characterized by autonomic nervous system dysregulation, specifically increased neuro-hormonal and sympathetic activation coupled with diminished parasympathetic activation (Francis et al., 1990; Mark, 1998; Middlekauff & Mark, 1998; Zucker, Wang, Brandle, Schultz, & Patel, 1995). The end result of this abnormal balance is a condition that damages cardiac muscle and leaves the heart unable to supply enough blood and oxygen to the body’s organs and tissues. At some point, which has not been clearly defined in the literature, as HF progresses, cerebral tissue perfusion becomes impaired (Woo, Macey, Fonarow, Hamilton, & Harper, 2003; Woo et al., 2005).

Several authors have documented neuroimaging alterations in the areas of the brain related to cognition in individuals with HF, such as the frontal and temporal lobes (AHA, 2010; Serber et al., 2008; Vogels et al., 2007). Unfortunately, no direct cause and effect relationships have been documented to date. Further, little is known about the affect of risk factors such as hypertension, hypercholesterolemia, sleep apnea, and diabetes on cognitive impairment. Other psychological and physiological conditions (including aging, psychiatric illnesses, such as schizophrenia, and acute medical injuries/illnesses) can also affect the ability to process information and result in cognitive impairment (Beatty et al., 2003; Evans & Bienias, 2005; Harvey, Wingo, Burdick, & Baldessarini, 2010; Keefe, Eesley, & Poe, 2005; Mooney et al., 2007; Waldenstein, Giggy, Thayer, & Zonderman, 2005). Understanding cognitive impairment in HF, as well as teasing out the contribution of other cardiac risk factors and psychological/physiological conditions is an essential goal for researchers and will provide clinicians with crucial information related to the identification (screening) and possible management strategies (interventions) for cognitive impairment.

The Measurement of cognitive impairment in heart failure

Global cognitive measurement

Differences in cognition were first examined pre- and postcardiac transplantation (Schall, Petrucci, Brozena, Cavarocchi, & Jessup, 1989). In the mid 1990s, researchers began to investigate cognition in the HF population (Cacciatore et al., 1998; Rengo et al., 1995; Zuccala et al., 1997). These early studies utilized instruments that measured global cognition or overall cognitive ability. The most common measure of global cognition used was the Mini-Mental Status Exam (MMSE; Folstein, Folstein, & McHugh, 1975). Orientation, memory and registration, attention, comprehension, and visual and writing skills are assessed during the MMSE and a single, global score is assigned. Scores for the MMSE range from 0 to 30 with higher scores indicating better global cognition. Use of traditional impairment cut-points (i.e., total score ≤ 24 out of a possible 30 points) has yielded conflicting results in the HF literature. Reasons for the inconsistencies are unclear; however, Folstein noted that the MMSE has decreased sensitivity to frontal lobe and subcortical changes, limiting the test’s ability to reliably identify executive-function impairment (Folstein, 1998). Frontal lobe changes and impairment of the executive function domain have been reported in the current HF literature, calling into question the appropriateness of the use of the MMSE as a screening instrument in the HF population.

Domain-specific cognitive measurement

The science surrounding cognition in HF has progressed and currently researchers are able to more accurately examine cognition with domain-specific measures. However, researchers do not utilize consistent assessment measures of the cognitive domains; therefore, evidence supporting impairment of individual cognitive domains varies. Table 3 provides a summary of studies that have utilized neuropsychological testing and the cognitive domains that were impaired. The most frequently measured cognitive domains in HF studies are attention, working memory, delayed memory, learning, executive function, and psychomotor speed, with approximately 28%–58% of individuals with HF demonstrating impairment of one or more of those cognitive domains (Bauer et al., in press; Bennett & Sauve, 2003; Pressler, 2008; Vogels et al., 2007). The following section details some of the common neuropsychological instruments used to assess each domain and examines the evidence for impairment.

Table 3.

Sample characteristics, impaired cognitive domains, and neuropsychological instruments used in current HF literature

Author and year (sample)	Domain(s) with significant impairment	Instrument
Almeida and Tamai (2001; 50 decompensated, hospitalized NYHA class III–IV HF and 30 healthy clinic patient controls)	Attention	Trail Making Test Part A
	Working memory	Digit Span Backwards
	Executive function	Trail Making Test Part B
	Psychomotor speed	Trail Making Test Part A
Bauer et al. (2011; 80 community-dwelling, NYHA class I–IV patients)	Attention	RBANS attention index Trail Making Test Part A
	Immediate memory	RBANS immediate memory index
	Delayed memory	RBANS delayed memory index
	Executive function	Letter fluency Trail Making Test Part B
	Psychomotor speed	Trail Making Test Part A
Bornstein et al. (1995; 62 preheart transplant patients and seven posttransplant and four nontransplanted heart transplant candidates)	Attention	Trail Making Test Part A
	Long-term memory	Visual reproduction subset
	Psychomotor speed	Finger tapping test
Gorkin et al. (1993; 318 community-dwelling, NYHA class I–II HF patients)	Attention	Digit Span Forward
	Psychomotor speed	Trail Making Test Part A
Hoth et al. (2008; 31 community-dwelling, NYHA class II–IV HF patients and 31 community-dwelling patients with CV disease, but no HF)	Attention	RBANS and Letter–Number sequencing
	Long-term memory	RBANS attention index
	Executive function	Trail Making Test Part B and letter fluency
	Psychomotor speed	Trail Making Test Part A and Stroop color word test
Incalzi et al. (2003; 369 hospitalized NYHA class II–IV HF patients)	Working memory	Corsi’s block tapping test
	Long-term memory	Rey auditory test
	Learning	Rey auditory test
	Executive function	Raven matrices
Schmidt et al. (1991; 20 community-dwelling HF patients and 20 healthy controls)	Attention	D2
	Long-term memory	Baeumler’s Learning and Gedachtnistest
	Learning	Baeumler’s Learning and Gedachtnistest
Serber et al. (2008; 12 community-dwelling NYHA class II–III HF patients and 54 healthy, age- and education-matched controls)	Attention	Trail Making Test Part A
	Executive function	Trail Making Test Part B and clock drawing test
Tanne et al. (2005; 20 NYHA class III patients enrolled in cardiac rehab program and five non-HF controls enrolled in same rehab program)	Attention	Trail Making Test Part A and Stroop color and word test
	Executive function	Trail Making Test Part B
	Psychomotor speed	Trail Making Test Part A
Trojano et al. (2003; 308 hospitalized NYHA class II–IV patients and 207 hospitalized non-HF patients)	Attention	Attentional Matrices
	Long-term memory	Rey auditory test
	Learning	Rey auditory test
Vogels et al. (2008; 43 community-dwelling NYHA class II–III patients, 33 community-dwelling patients with cardiac disease, but no HF, and 22 community-dwelling healthy controls)	Attention	Trail Making Test Part A and Stroop color word test
	Memory	Rey auditory test and Digit Span Forward and pattern recognition
	Executive function	Intra-extra dimensional set shift and Stockings of Cambridge and Trail Making Test Part B
	Psychomotor speed	Stroop color word test and Trail Making Test Part A
Wolfe et al. (2006; 38 community-dwelling HF patients, no NYHA class specified)	Attention	RBANS attention index
	Long-term memory	RBANS delayed memory index
	Learning	RBANS immediate memory index

NYHA = New York Heart Association; RBANS = Repeatable Battery for the Assessment of Neuropsychological Status; CV = cardiovascular disease.

Attention

Attention is essential in order to further process the stimuli. Attention involves many cortical and subcortical structures, including the frontal and parietal association cortices, anterior cingulate cortex, basal ganglia, and cerebellum (Blumenfeld, 2002; Rosenzweig et al., 2005). In the HF literature, attention was most commonly measured using the Trail Making Test Part A (Almeida & Tamai, 2001; Bauer et al., in press; Bornstein, Starling, Myerowitz, & Haas, 1995; Gorkin et al., 1993; Gunstad et al., 2005; Putzke et al., 2000; Serber et al., 2008; Tanne et al., 2005). Attention was impaired in almost every reviewed HF study (Almeida & Tamai, 2001; Alves et al., 2005; Bauer et al., in press; Bornstein et al., 1995; Gorkin et al., 1993; Gunstad et al., 2005; Hoth, Poppas, Moser, Paul, & Cohen, 2008; Incalzi et al., 2003; Putzke et al., 2000; Schmidt et al., 1991; Tanne et al., 2005; Trojano et al., 2003; Vogels et al., 2007; Wolfe, Worrall-Carter, Foister, Keks, & Howe, 2006). However, in three studies, attention was reported to be preserved. Incalzi et al. (2003) utilized Attentional Matrices and reported there was no difference in scores of attention between community-dwelling individuals with HF and healthy controls. Deshields, McDonough, Mannen, and Miller (1996) and Roman et al. (1997) used the Trail Making Test Part A in samples of individuals needing cardiac transplantation and found no significant differences in scores of attention between individuals awaiting transplantation and healthy controls (Deshields, McDonough, Mannen, & Miller, 1996) and pre/post transplantation (Roman et al., 1997).

Sensitivity is an important factor for clinicians to consider and may contribute to the contradictory results reported above. In regards to attention, an overwhelming number of studies document impairment of attention in HF. Therefore, Attentional Matrices may not be sensitive to the type of impairment found in the HF population. Further, although there is evidence supporting the sensitivity of Trail Making Test Part A in the HF population, it appears that Trail Making Test Part A may not be a sensitive tool for the cardiac transplantation population.

Working memory

Working memory is often evaluated under the domain of attention because functional imaging demonstrates that areas of the frontal lobe are involved with working memory tasks. Working memory was tested using various instruments, including digit span backward (Almeida & Tamai, 2001; Grubb, Simpson, & Fox, 2000; Vogels et al., 2007), Corsi’s block tapping test (Incalzi et al., 2003; Trojano et al., 2003), and short category test (Putzke et al., 2000). Results from studies examining working memory are conflicting, even with use of the same instrument. For example, studies by Grubb Simpson, and Fox (2000) and Vogels et al. (2007) used digits backwards to examine working memory and controlled for common confounders such as age and education, but reported contradictory results in similar samples of community-dwelling individuals with HF.

Learning

Learning involves many different areas of the brain. Functional imaging studies have shown activation in the prefrontal cortex and medial temporal lobe regions with learning of new material (Blumenfeld, 2002). This is not unexpected as learning involves attention and memory. In all the reviewed studies that examined learning, it was noted to be impaired despite use of multiple neuropsychological instruments and use of case–control sampling or comparisons to healthy controls (Alves et al., 2005; Incalzi et al., 2003; Roman et al., 1997; Schmidt et al., 1991; Trojano et al., 2003).

Memory

Memory is divided into multiple categories; this review will focus on the most common memory categories pertinent to HF: immediate and delayed. Immediate memory generally refers to information that has been stored for only a matter of minutes while delayed memory refers to information that has been stored for at least 15 min. The medial temporal lobe area (in particular, the hippocampus and parahippocampal gyrus) and the medial dencephalic area (including the thalamic mediodorsal nucleus, anterior nucleus of the thalamus, and mammillary bodies) are involved in the formation, consolidation, and retrieval of memory (Blumenfeld, 2002; Rosenzweig et al., 2005). The communication between these two areas is also very important to memory function and is carried out by various white matter tracts.

Multiple studies in the HF literature reported that delayed memory was impaired (Bauer et al., in press; Bornstein et al., 1995; Hoth et al., 2008; Incalzi et al., 2003; Schmidt et al., 1991; Trojano et al., 2003; Vogels et al., 2007; Wolfe et al., 2006). Of interest is that the majority of assessments utilized list-learning tasks (e.g., California Verbal Learning Test), which include not only a delayed memory score but also an immediate memory score (Alves et al., 2005; Bauer et al., in press; Bornstein et al., 1995; Hoth et al., 2008; Incalzi et al., 2003; Trojano et al., 2003; Vogels et al., 2007; Wolfe et al., 2006). In multiple studies, immediate memory scores were not reported to be different from the comparison group, suggesting that immediate memory may be preserved in HF (Deshields et al., 1996; Grubb et al., 2000; Putzke et al., 2000). Therefore, it appears that although data are encoded and able to be retrieved at short intervals (i.e., immediate memory), storage and/or subsequent retrieval of data is at issue (i.e., delayed memory) in individuals with HF.

Speech/language and visual/spatial/constructional

Very few studies have examined the speech/language domain and the visual/spatial/construction domain. Alves et al. (2005) reported that impaired performance on measures of language correlated to findings of increased brain abnormalities as measured by single-photon emission computed tomography in a sample of older individuals with HF. Vogels et al. (2007) used fragmented line drawings in an ascending order of completeness and the MMSE subscore for visual figure copy to assess the visual/spatial/constructional domain and the MMSE subscore for language to assess the speech/language domain. The authors reported no performance differences between the groups in either the speech/language or visual/spatial/constructional domains. Given the lack of data related to these two domains, further study is recommended in order to determine whether speech/language and visual/spatial/constructional skills are impaired in patients with HF.

Executive function

Executive function encompasses a human’s most complex behavior. It allows an individual to respond appropriately to novel scenarios. Executive function involves primarily the prefrontal and frontal cortices. It is a critical domain but difficult to assess. Lezak (2004) suggested that executive functions have four common components: (a) volition, (b) planning, (c) purposive action, and (d) effective performance (Lezak, 2004, p. 611). The best measures of executive function are obviously “real world” scenarios. Unfortunately, “real world” scenarios are very difficult to duplicate on standardized neuropsychological tests.

Executive function was evaluated in several of the reviewed studies, using multiple neuropsychological instruments, including Trail Making Test Part B, letter fluency tasks, and Raven’s Progressive Matrices. In the majority of studies, executive function was reported to be impaired when compared to controls (Almeida & Tamai, 2001; Hoth et al., 2008; Incalzi et al., 2003; Tanne et al., 2005; Vogels et al., 2007). However, two of the published studies (Alves et al., 2005; Trojano et al., 2003) found no difference between HF and comparison groups. The complexity of executive function calls into question the sensitivity of instruments in the HF population. The inconsistent results may be related to instruments being differentially sensitive in the HF population and illustrate the need for studies to compare multiple instruments.

Implications for the advanced practice nurse

In summary, studies suggest that individuals with HF commonly suffer from cognitive impairment in one or more cognitive domains; most frequently attention, memory, learning, and executive function. Impairment of these domains has multiple clinical implications including a decreased ability to attend to and comprehend patient education materials and an inability to appropriately assess and self-manage symptoms. It is imperative that clinicians are aware of these implications when working with their HF patients.

Unfortunately, the examination of cognitive impairment is still in its infancy and available scientific information regarding this problem is limited to describing the phenomena. Many gaps remain for both scientists and clinicians; however, there are two important limitations that need to be addressed. First, regardless of the etiological mechanism, cognitive impairment is prevalent among HF patients and clinicians lack sensitive screening instruments. Second, once cognitive impairment is identified, clinicians need effective interventions to prevent and/or compensate for the cognitive impairment present in the HF population.

The neuropsychological batteries used in most HF research studies are long and impractical for clinic use. Some studies reported that as many as a third of the participants were unable to complete the batteries because of fatigue (Incalzi et al., 2003; Vogels et al., 2007). As science achieves a better understanding of the relationship between individual domain decline and function status decline, domain-specific screening instruments can be identified. Unfortunately, to date there is a dearth of information related to screening instruments and no consensus regarding screening instruments exists among the key professional organizations (Cameron, Ski, & Thompson, 2011).

As reported earlier, screening with global measures of cognitive function such as the MMSE have resulted in disparate findings and multiple authors have questioned the sensitivity of such measures (Bauer et al., in press; Harkness, Demers, Heckman, & McKelvie, 2011). Executive-function impairment is well documented in the HF population and measures of executive function have previously been correlated to functional status in a stable cardiovascular sample (Jefferson, Paul, Ozonoff, & Cohen, 2006), and may affect an individual’s ability to process and react to symptoms. Therefore, executive-function measures may serve as the best screening instruments.

The Clock Draw Test (CDT) has been labeled as an executive-function measure (Riegel et al., 2002). The task requires an individual to draw the face of clock and a specific time assigned by the evaluator. Although the CDT has been used successfully in dementia populations, it has had limited success identifying mild cognitive impairment (Ehreke, Luppa, Konig, & Riedel-Heller, 2010; Ismail, Rajji, & Shulman, 2010). Another common measure of executive function, letter fluency, was explored in a small study by Bauer et al. (2011). Letter fluency (see Table 2) identified 53% of a sample of stable, elderly HF patients as cognitively impaired and scores significantly correlated with both instrumental activities of daily living and functional status (Bauer et al., in press). However, given the limitations of this study (small, homogenous sample), further validation of the letter fluency test is underway.

The Montreal Cognitive Assessment (MoCA) assesses multiple cognitive domains, including attention, memory, language, and executive function (Nasreddine et al., 2005). The instrument can be completed in approximately 10 min and scoring ranges from 0 to 30. In the initial evaluation of the tool, scores of ≤26 had a sensitivity of 90% and a specificity of 78% for mild cognitive impairment in a geriatric sample containing healthy controls, patients with mild cognitive impairment, and patients with Alzheimer’s dementia (Nasreddine et al., 2005). Harkness et al. (2011) evaluated the MoCA in a small sample of stable, elderly patients with HF. The authors reported that more than 70% of their sample met the criteria for mild cognitive impairment. While there were limitations to this study, it provides initial support for use of the MoCA in the stable HF population, but further validation is warranted.

While researchers are exploring possible screening instruments, another issue lingers for clinicians. Once cognitive impairment is identified, interventions should be implemented. Unfortunately, the research to date has focused on describing cognitive impairment and exploring its prevalence in the HF population. While there are no intervention strategies documented in the current literature, it seems likely that interventions related to cognitive impairment will take two forms: preventive and compensatory. Before preventive interventions can be designed and tested, more knowledge of the etiology and trajectory of cognitive impairment must be documented.

Because attention, memory, and executive function are most frequently impaired, design of compensatory interventions should be explored. Educational materials and interventions designed for patients with low literacy may be useful for patients with memory impairment. Low-literacy materials often utilize pictures to remind patients of medication, diet, and other management therapies. These types of materials may be particularly helpful if the visual/spatial areas of the brain are preserved in HF.

No study was located that explored compensatory interventions for cognitive impairment in the HF population. However, multiple studies were reviewed that explored advanced practice interventions that may, at least partially, compensate for the impairment documented in executive function (keep in mind that executive function is the domain that helps an individual “put it all together,” i.e., the ability to assess symptoms and respond accordingly). In a recent review, Delgado-Passler and McCaffrey (2006) emphasized the positive influence Advanced Practice Nurse (APN)-led interventions had on hospital readmission rates and emergency room visits. Decrease in rehospitalization and emergency room visits was most apparent in studies that utilized frequent contact by the APN following hospital discharge (Benatar, Bondmass, Ghitelman, & Avitall, 2003; Brandon, Schuessler, Ellison, & Lazenby, 2009; Komosa & Hayes, 2007; Naylor et al., 1999).

Frequent contact (via telephone or home visit) provided more rapid recognition of increased symptom severity and issues with medication adherence. In turn, the APNs were able to adjust therapies more quickly or schedule provider visits in a timelier manner. Komosa and Hayes (2007) reported that the length of frequent contact varied and was discontinued when patients were considered stable and demonstrated effective self-management skills related to diet, medications, and symptom recognition. Interestingly, rehospitalization rates continued to decline after the study protocol was completed, this is contradictory to many previous study findings (Komosa & Hayes, 2007). This leads one to believe that there may be key time periods within the HF disease trajectory where cognitive impairment is more severe and APN follow-up needs to be more concentrated.

In summary, cognitive impairment is a common problem among individuals with HF. While sensitive screening instruments are not yet available, clinicians need to be aware of this phenomenon. Further, while interventions have not been designed and tested, patient education materials and disease management programs that help compensate for potential memory and executive function losses may be useful.