Potential Effect of Screening for Subtle Cognitive Deficits on Hospital Readmission

To the Editor: Several conditions that significantly affect functionality and independence may be subtle and go unrecognized, potentially leading to nonadherence to medical recommendations and readmission. Existing risk-prediction models for hospital readmission have been shown to perform poorly.1 Studies suggest that unrecognized cognitive deficits may exist after the illness that necessitated the admission was successfully treated, resulting in an unappreciated risk for readmission. The risk of cognitive impairment increases with age and is amplified with hospitalization resulting in significant morbidity.2 The frequency of cognitive impairment ranges from 15% to 35% in hospitalized elderly adults on general medicine services but may be even higher.3 Identifying vulnerable individuals with cognitive deficits at the time of hospital admission is critical to prevent, establish a diagnosis of, and treat delirium.4 Cognitive impairment is also associated with depression in late life and correlates with poorer quality of life and greater healthcare use.5 Identification of subtle cognitive deficits can prove to be challenging, because many cognitively impaired individuals with intact language and memory can be perceived to be functionally independent. Executive cognitive functions are cognitive processes that orchestrate complex, goal-directed actions.6 Impairment of the former undermines an individual's independence by interfering with the direction, planning, execution, and supervision of complex behavior. Screening individuals for obscured cognitive impairment at the time of hospital discharge could be the first step in early identification of mild to moderate cognitive impairment and allow for interventions to reduce related disability and avoidable readmissions. This study examined subtle cognitive deficits that often go undetected in association with delirium, depression, and executive dysfunction. Individuals aged 65 and older admitted with diagnoses of congestive heart failure, exacerbation of chronic obstructive pulmonary disease, pneumonia, or myocardial infarction regardless of motor deficits were included. Exclusion criteria were admission from a skilled nursing or assisted living facility, medical history of dementia or cognitive impairment, English as a secondary language, and an education level less than high school. A trained nurse screened older patients on Day 2 or 3 of admission. All individuals were screened for delirium using the Confusion Assessment Method,7 instrumental activities of daily living using Lawton's scale, executive dysfunction using the Controlled Oral Word Association Test, and the oral version of the Trail-Making Test Part B. Depression screening was performed using the Patient Health Questionnaire. The comparison (control) group consisted of age-matched elective surgical postoperative patients without a diagnosis of dementia or the aforementioned four diagnoses and not admitted from a nursing or assisted living facility. The study sample consisted of 43 cases and 27 controls. Rates of delirium, depression, and executive dysfunction were 15.2%, 19.6%, and 83.7%, respectively, in the study group and 7.7%, 0%, and 50%, respectively, in the control group. Rate of readmission within 1 calendar year was evaluated; 21 of the 23 (91.3%) readmitted cases and three of the five (60%) readmitted controls tested positive for executive dysfunction (P < .05). It was possible to identify a large prevalence of executive dysfunction in a population at high risk for readmission. Screening older patients, in particular those with underlying diagnoses known to have a high risk for readmission for subtle cognitive deficits, may help to direct interventions and the allocation of limited resources to improve healthcare outcomes, including prevention of readmission. Institutional review board approval for this quality improvement study was obtained from Greenwich Hospital, Yale New Haven Health. The authors are grateful for the participation of Charles Seelig, MD, for support with data analysis and data integrity validation; Patricia Babcock, RN, and Spike Lipschutz, MD, for administrative support; and Alva Bound, RN, and William Graham for data collection. Conflict of Interest: Dr. Steven Buslovich is employed as a physician by Greenwich Hospital, the setting of the study, but he was not compensated for the work involved in this study. Author Contributions: Buslovich: study concept and design, acquisition of participants and data, analysis and interpretation of data, preparation of manuscript. Kennedy: study design, interpretation of data, preparation of manuscript. Sponsor's Role: None.