This study evaluated important and understudied factors in hospice care: cognition and decision-making capacity in patients receiving hospice care and who had no chart diagnosis of a cognitive disorder and no clinically obvious cognitive impairment (e.g., disorientation, incoherence, unresponsiveness). In general, it appears that patients receiving inpatient care are more cognitively impaired than their homecare counterparts, possibly because they are nearer to death or very ill with symptoms that may have interfered with cognition. As hypothesized, however, these results indicate that regardless of inpatient/outpatient status, more than half (54%) of hospice patients with no known or obvious cognitive impairment nonetheless manifest significant impairments in a variety of cognitive domains. The results also suggest that this cognitive impairment may not be attributable to mere medication effects, as there was no significant relationship between opiate/steroid medication dosage and neuropsychological performance, at least in patients receiving inpatient hospice care.
With regard to the relationship between neuropsychological performance and decisional capacity, participants with significant cognitive impairment performed significantly worse on the decision-making measure than those without such impairment. In addition, these results suggest that performance on measures of general cognitive function, verbal learning, verbal memory, verbal reasoning, and verbal (semantic) fluency predict decision-making capacity.
These findings are consistent with previous research indicating high rates of cognitive impairment in the palliative care population (1
). It appears that, consistent with the published literature (5
), the sources of cognitive impairment in hospice patients are numerous and could be cumulative as the illness advances. In addition, the results from this larger sample (which includes participants from our previous published results) indicate a greater frequency of cognitive impairment than those reported previously (4
). Our results also extend previous findings that verbal abilities predict decisional capacity (14
), as verbal learning, verbal memory, verbal reasoning, and verbal fluency were among the significant predictors of decisional capacity. It should be noted that only a small percentage of those screened met inclusion criteria for the study, indicating that a large proportion of hospice patients have a clinical presentation that interferes with research participation and testing. Thus, the overwhelming majority of patients receiving hospice care have clinically obvious and/or documented cognitive impairments. These clinical features likely translate to lack of decision-making capacity, such that this study seriously underestimates the prevalence of diminished cognitive ability and decisional capacity in all patients receiving hospice care.
Limitations of the study include lack of longitudinal information regarding cognitive decline and its impact on psychosocial functioning, which precluded formal diagnoses of dementia or other cognitive disorders for participants who demonstrated significant cognitive impairment. Further investigations are needed to yield more precise diagnoses. In addition, our criteria for significant cognitive impairment may have excluded those with more subtle, subclinical cognitive dysfunction. However, the correlational analysis of cognitive performance and decisional capacity incorporated all levels of performance and would have captured this subclinical cognitive impairment that could affect decision-making ability. Although we explored a number of factors hypothesized to contribute to impaired cognition (e.g., medication, psychiatric symptoms), the negative findings suggest that there are other contributing factors that were not measured, including sequelae of the advanced illness itself. Further, the measure for decisional capacity, the UBACC-T, is a newly developed instrument and its psychometric properties are not yet established; our results should therefore be interpreted with caution as the UBACC-T is not a validated instrument. Follow-up studies should include examination of the reliability and validity of the UBACC-T, and perhaps include a known valid instrument for decision-making to assess concurrent validity (e.g., the MacArthur Competence Assessment Tool – Treatment; MacCAT-T) (34
). Also, the scenario presented in the UBACC-T was hypothetical (albeit relevant), so it may not fully generalize to real-world decisions in which patients are invested and likely to pay close attention. Despite these limitations, this study augments the existing literature on cognitive ability and decision-making capacity in patients receiving hospice care. Participants were selected through rigid adherence to exclusion criteria in an effort to create the most cognitively “normal” appearing sample possible based on clinical judgment via usual hospice care. Although the vast majority of screened patients were excluded, these results indicate that of those who would likely be considered cognitively intact based on clinical impression, over half exhibit significant cognitive impairment.
Because a large proportion of this sample demonstrated cognitive impairment and these deficits were found to be related to poorer decisional capacity, there may be implications for treatment. Greater attention could be paid to cognition in hospice patients, including impairment in verbal abilities (learning, memory, reasoning, fluency) that could contribute to poor communication among patients, providers, and family members, as well as impaired understanding or reasoning related to treatment decisions, including medication or other treatment adherence. Clinicians should be cognizant of the strong potential for impaired cognition and decision-making capacity, particularly in patients who otherwise appear intact. Assessing capacity can be done at the bedside with or without structured instruments, but needs to be considered especially when high stakes decisions are at hand. Clinicians can identify patients at risk for impaired decision-making and ameliorate impaired capacity when possible. Ensuring that family or other interested parties are enlisted to help in decision-making is a good option, especially early in the illness trajectory. Previous research has demonstrated that more than 75% and perhaps greater than 90% of patients near the end of life want a family member to help them make decisions with the doctor about their care (35
). Many clinicians favor advance care planning because it facilitates discussion of end of life issues between patients, physicians, and caregivers, and provides an opportunity to respect a patient’s predetermined wishes even if the patient lacks decision-making capacity at some point in the future (37
). Advance care planning may also decrease stress in the surrogate decision maker (38
), however, many individuals with cognitive impairment still do not complete advance care planning, raising the possibility that patients and families need more information or counseling related to understanding the goal of advance directives and the options they provide (39
Patients who have difficulty understanding or appreciating information may benefit from brief interventions like multiple learning trials and summaries of information, receiving information through multiple methods (e.g. hearing, seeing), and being provided ample time and opportunity to paraphrase what was explained and to review information again (40
). Further research on methods to improve the identification of cognitive and decisional impairments at the end of life are needed, as the stakes of advanced, life-threatening illnesses for an individual, their family, and society in general can be immense. Armed with the knowledge that hospice patients often exhibit cognitive deficits and that such deficits could contribute to impaired decisional capacity, palliative care clinicians can initiate interventions to improve cognitive function and/or decisional capacity to enhance end of life communication and outcomes for both patients and families.