The United States is experiencing a demographic shift, with the number of adults aged 65 years or older projected to grow from ~57 million in 2022 to 75 million by 2035, exceeding 20% of the total population.¹ Older adults undergo 30% of all surgeries (Mattingly et al., 2021), and this is expected to increase as the population ages. Such increase is also found in a number of other countries. In elderly patients, postoperative neurocognitive disorders constitute one of the most common complications of surgery in older adults (Suwanabol et al., 2022; Evered et al., 2018). Meanwhile, preoperative cognitive impairment is a risk factor for not only postoperative delirium and major neurocognitive disorder but also for prolonged length of stay, readmission, and discharge to assisted care (Culley et al., 2017; Chen et al., 2022).

Preoperative cognitive screening is recommended by practice guidelines and consensus statements from the American College of Surgeons, the American Geriatrics Society, and the American Society of Anesthesiologists (Peden et al., 2021; Mohanty et al., 2016). A 1995 consensus statement on assessment of cognitive outcomes after cardiac surgery provided guidance on selection of neuropsychological tests, including recommendation of a core battery of tests, timing of assessments, and considerations during performance of the assessments (Murkin et al., 1995). However, a recent review of 274 studies published between 1995 and 2019 on postoperative cognitive dysfunction found considerable heterogeneity in testing batteries, with more than 250 unique tests used, making comparisons across studies challenging and raising the need for standardized tests that target multiple cognitive domains (Borchers et al., 2021).

The Uniform Data Set (UDS) Neuropsychological Battery has been used since 2005 by the National Institute on Aging's Alzheimer's Disease Research Centers Program (Morris et al., 2006). These standardized neuropsychological tests have been performed on over 52,000 participants, including those with normal cognition, mild cognitive impairment, and dementia.² The UDS has been used to study cognitive decline associated with normal aging, Alzheimer's disease (Kiselica et al., 2024), Parkinson's disease (Lea et al., 2021), and chronic traumatic encephalopathy (Alosco et al., 2021). The UDS was revised in 2018 to the current UDS version 3. A modified UDS for remote administration, known as the tele-adapted neuropsychological battery (T-cog), has been adopted since the COVID-19 pandemic and demonstrates good reliability (Gierzynski et al., 2024; Howard et al., 2023; Loizos et al., 2021; Smith et al., 2023). Key differences are the removal of visuospatial tests, such as the Benson complex figure test, the multilingual naming test (MINT), and the drawing portions of the Montreal Cognitive Assessment (MoCA), as well as the augmentation of the trail making tests into oral versions requiring use of the same cognitive domains but without use of pen and paper. These modifications reflect a growing emphasis on flexibility and accessibility in cognitive assessment methodologies.

Use of telehealth accelerated during the COVID-19 pandemic as a safe and effective alternative to in-person visits. Advantages of telehealth visits include convenience, access, and reduced travel costs. Accordingly, the use of remote cognitive assessments has been shown to be feasible in studying perioperative neurocognitive disorders. Studies thus far have used screening tools such as the MoCA Blind/Telephone, the Ascertain Dementia Eight-item Questionnaire, the Centers for Disease Control cognitive question, and the Modified Telephone Interview for Cognitive Status in the preoperative setting (Yan et al., 2024; Cooper et al., 2022). A prospective study of patients undergoing major gynecologic oncologic surgery used the Mini-Mental State Exam (MMSE) in person and virtually, preoperatively and up to 3 months postoperatively (Makkar et al., 2024). However, more extensive remote neuropsychological testing has not been used in the perioperative population. Remote comprehensive cognitive assessment is important, however, given the high prevalence of cognitive decline associated with surgery in older adults and the expanded role of telehealth in medicine and research.

Here, we examined the feasibility of using the remotely administered T-cog for perioperative cognitive assessment in older adults scheduled for total joint lower extremity arthroplasty. We found that completion rates for the UDS v3.0 T-cog achieves near 100%. Furthermore, patients reported overall high satisfaction with the testing battery and preferred it to in-person testing. In our study, we found a rate of cognitive impairment that is comparable to prior studies of patients in this population. These results indicate that the UDS v3.0 T-cog is both a feasible and a practical tool for assessing cognitive function in the older adult perioperative population.

The study protocol was approved by the New York University Grossman School of Medicine Institutional Review Board (6/27/2023, #i23-00664) and conducted in accordance with the latest revision of the Declaration of Helsinki. Written informed consent was obtained from all participants.

In this study, we tested the feasibility of using a remotely conducted UDS v3.0 T-cog to assess cognitive function in a unique population of older adults who are about to undergo lower extremity surgery. We found that the UDS v3.0 T-cog is user-friendly, allows near total completion rates, and is feasible to administer before surgery to older adults. Further, we found that ~28% of the patients in our cohort showed signs of cognitive impairment.

To the best of our knowledge, this is the first study to utilize a comprehensive UDS v3.0 cognitive testing battery to study cognitive function in older patients in a perioperative setting. In the last 5 years, remote assessment of cognitive function has gained considerable traction (Katz et al., 2021; Chappelle et al., 2023; Lai et al., 2022), and UDS v3.0 T-cog has been advocated by the National Institute on Aging's Alzheimer's Disease Research Centers Program, been widely adopted (Sachs et al., 2024; Hackett et al., 2021), and shown validity (Gierzynski et al., 2024; Howard et al., 2023; Smith et al., 2023). In our study, we are approaching a patient population who are more difficult to enroll for in-person assessment, due to their impending surgery. There is increased interest, however, to characterize cognitive function in this patient population. Screening tests have been used for remote preoperative cognitive assessments (Yan et al., 2024; Cooper et al., 2022; Yu et al., 2021), however, few studies have focused on using comprehensive assessment tools in this patient population. This is important, as older adults may manifest cognitive deficits in different domains (Cullum et al., 2000). Surgery and pain may affect different dimensions of cognition, and a comprehensive battery allows assessment of which cognitive dimension(s) may be affected. Likewise, certain treatments may affect one or more dimensions of cognition selectively. Further, more detailed assessment allows more detailed studies on mechanisms of cognitive decline, especially as it allows associations of behavioral outcomes with genetic/genomic and neurophysiological data. Our study shows that overall completion rate for remote assessment is very high. Interestingly, not only do patients in our cohort demonstrate overall satisfaction with the remote UDS v3.0 T-cog, but they also endorse preference for the remote assessment to in-person assessment. In fact, our data completion rate compares favorably to in-person studies, which typically demonstrate 78–95% data completion rate (Ballard et al., 2012; Moller et al., 1998; Silbert et al., 2015; Berger et al., 2022).

In our study, we found that 28% of patients showed cognitive impairment (CI), a finding consistent with previously reported data in the same age group with other remotely conducted or in-person conducted studies (Culley et al., 2016; Kapoor et al., 2022; Susano et al., 2020). Additionally, participants with CI demonstrated significantly lower performance across multiple cognitive domains, particularly those related to learning and memory (Figure 2). These results further support the validity and feasibility of the UDS v3.0 T-cog as a reliable screening tool for cognitive impairment.

The feasibility of and patient satisfaction with the UDS v3.0 T-cog assessment is promising for longitudinal testing. Indications for longitudinal assessment are multiple. First, longitudinal assessment can be performed for patients at risk for developing dementia (Cruz-Almeida et al., 2019). Second, it allows us to correlate behavioral findings with blood or cerebrospinal fluid biomarkers. Third, longitudinal use of UDS v3.0 T-cog can be performed to assess potential adverse effects on cognition for a number of drugs. Finally, it can be used to study progression or emergence of cognitive decline after trauma or surgery, such as for studies of postoperative cognitive function.

While our study has demonstrated overall feasibility for the remote UDS, there are some limitations. We have used two different options for remote assessment—an audio only option and a video option. A video option allows observation of patients and potentially improved ability of researchers to assess perceived validity with the T-cog Validity Checklist. However, audio allows patient participation in cases in which full video-conferencing capability is lacking. In our study, 73% utilized video assessments over the telephone option. Furthermore, this is a single center study, and thus future multi-center studies are needed to further demonstrate feasibility, especially for populations where access to remote testing is more limited. Another limitation of remote assessment is that it may not fully capture all facets of executive function such as visuospatial ability. Lastly, Verbal Naming Test Z-scoring can be confounded by patients who learned English as a second language, as these patients may have scored poorly on this task due to translation challenges rather than cognitive impairment.

In this study we have shown the feasibility of a remotely conducted UDS neuropsychological testing battery in a perioperative patient population. We found that ~28% of patients over the age of 65 in this clinical setting showed cognitive impairment. Importantly, the UDS v3.0 T-cog is well-tolerated and feasible to administer in the real-world setting. These results support the use of the UDS v3.0 T-cog to assess cognitive function of patients in the perioperative setting.