Contributed by Angela Chen, BA1, 2; Robert Whittington, MD1; Cecilia Canales, MD, MPH1.
1. Department of Anesthesiology and Perioperative Medicine, UCLA David Geffen School of Medicine, Los Angeles, CA, USA.
2. Chicago Medical School, Rosalind Franklin University of Medicine & Science, North Chicago, IL, USA
Background
Older adults are highly vulnerable to postoperative delirium, particularly those with preoperative mild cognitive impairment or dementia. However, cognitive impairment in older adults who speak a language other than English (LOE) is often underdiagnosed or misdiagnosed,1,2 increasing the risk for adverse postoperative cognitive outcomes. Guidelines recommend that older adults should undergo preoperative cognitive screening to identify those patients at greatest risk for postoperative delirium.3
The primary objective of this pragmatic study was to compare preoperative cognitive screening rates for LOE and English-speaking patients after the implementation of routine preoperative screening of older adults with the Mini-Cog. We also aimed to assess the association of preoperative impairment on Mini-Cog and postoperative delirium in both LOE and English-speaking patients and whether this association differed between these two groups.
Methods
At our institution, we implemented a routine preoperative cognitive screening program for older adults (age ≥ 65 years). During the subsequent three years, a total of 2446 older surgical adults were eligible for screening with the Mini-Cog. The Mini-Cog was selected over other cognitive screening tools given it takes only approximately three minutes to complete, has been validated in 21 languages, has high interrater reliability, has been shown to have minimal education or ethnicity bias, is readily available for clinical use, and has been shown to be predictive of delirium.4,5
Patients were categorized as either LOE or English-speaking; LOE was deemed to be present if at least 1 of the following criteria were met: (1) LOE self-identified by the patient; (2) a translator was used to obtain surgical consent; (3) a review of records indicated the patient needed language assistance during the perioperative care period.6
Delirium was identified during the first five days after surgery using both twice per day Confusion Assessment Method (CAM) assessment and chart review using published criteria.7 Given the fluctuating nature of delirium, this combined approach has been shown to better capture acute episodes of delirium than CAM assessments alone.4
Results
Among the 2446 patients included, 685 (28%) were deemed LOE and 72% (1765) English-speaking. Spanish (47.2%) was the most common LOE language. Baseline characteristics between the two groups were not different (mean age: LOE 75.1 ± 6.9 vs. English 74.7 ± 6.7 years; female sex: LOE 63.5% vs English 65.5%).
A total of 1956 (80%) of the 2446 patients were screened with the Mini-Cog. The proportion of patients not screened was significantly higher and nearly four times greater in the LOE (vs. the English-speaking) group (287/685 (41.9%) vs. 203/1761 (11.5%), P<0.001). Among the screened patients, the proportion who screened Mini-Cog positive (≤ 2) was not different between the LOE (vs English speaking) groups (29% vs. 28%, P=0.681).
Incident postoperative delirium occurred significantly more frequently in the LOE (vs. English) group (23.9% vs. 18.4%, P=0.002). In both the LOE and English groups, a positive (vs. negative) preoperative Mini-Cog was associated with a significantly greater odds of postoperative delirium (LOE: OR=3.9; 95% CI, 2.1–7.3; P< .001; English: OR=3.5; 95% CI, 2.6-4.8; P< .001).
However, the similarity in these odds ratios suggests this relationship was not influenced by language preference. Of the 490 patients across both groups who did not receive preoperative cognitive screening, 42% developed postoperative delirium. Among the cognitively unscreened, the delirium rate was not significant different between LOE: 38% vs English-speaking: 47%, P=0.061 groups. However, both of these delirium rates were higher than in cognitively screened groups regardless of language preference (LOE: 23.9%; English 18.4%).
Conclusions and Future Directions
In summary, language disparity influences preoperative cognitive screening rates. Despite the institutional implementation of a routine preoperative cognitive screening program for all patients ≥65 years of age, LOE patients were nearly four times more likely not to undergo such screening. Interim analysis suggests overall adherence to the protocolized cognitive screening was good. Yet when we stratified patients by language preference, disparities in cognitive screening adherence became apparent.
Discussions with healthcare providers implementing the Mini-Cog, revealed knowledge gaps on the part of the providers that the Mini-Cog could be used with an interpreter or that it has been validated in multiple languages. We suspect that since cognitive screening assessment time is usually longer when an interpreter is used, providers were more likely to truncate the overall preop assessment by skipping the Mini-Cog assessments in situations where the LOE patients/families did not have time to wait for the interpreter to arrive.
Abnormal cognitive screening with the Mini-Cog was similarly predictive of postoperative delirium in both English and non-English speaking patients, highlighting the importance of preoperative cognitive screening in all patients regardless of primary language preference. Moreover, the delirium rates among screened patients (whether English speaking or LOE) were lower than in the unscreened, suggesting that clinicians may be using these data to target interventions that reduce delirium. Also, delirium screening with the CAM may not always be able to be completed in LOE patients.
Ultimately, future studies are warranted to validate the findings of this study on a larger scale and determine if the implementation of language-equitable programs can diminish the disparities we observed and improve all patient outcomes. Since this evaluation of the program is based on language preference, we have increased protocol adherence in the LOE group by allotting longer appointment times for patients who require interpreters for their care.
References
- Manuel SP, Nguyen K, Karliner LS, Ward DT, Fernandez A. Association of English Language Proficiency With Hospitalization Cost, Length of Stay, Disposition Location, and Readmission Following Total Joint Arthroplasty. JAMA Netw Open. 2022;5(3):e221842. doi: 10.1001/jamanetworkopen.2022.1842
- Chin AL, Negash S, Hamilton R. Diversity and disparity in dementia: the impact of ethnoracial differences in Alzheimer disease. Alzheimer Dis Assoc Disord. 2011;25(3):187-195. doi: 10.1097/WAD.0b013e318211c6c9
- Chow WB, Rosenthal RA, Merkow RP et al. Optimal preoperative assessment of the geriatric surgical patient: a best practices guideline from the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society. J Am Coll Surg. 2012;215(4):453-466. doi: 10.1016/j.jamcollsurg.2012.06.017
- Culley DJ, Flaherty D, Fahey MC et al. Poor Performance on a Preoperative Cognitive Screening Test Predicts Postoperative Complications in Older Orthopedic Surgical Patients. Anesthesiology. 2017;127(5):765-774. doi: 10.1097/ALN.0000000000001859
- Decker J, Kaloostian CL, Gurvich T et al. Beyond Cognitive Screening: Establishing an Interprofessional Perioperative Brain Health Initiative. J Am Geriatr Soc. 2020;68(10):2359-2364. doi: 10.1111/jgs.16720
- Karliner LS, Napoles-Springer AM, Schillinger D, Bibbins-Domingo K, Perez-Stable EJ. Identification of limited English proficient patients in clinical care. J Gen Intern Med 2008;23(10):1555-1560. doi: 10.1007/s11606-008-0693-y
- Kuhn E, Du X, McGrath K et al. Validation of a consensus method for identifying delirium from hospital records. PLoS One. 2014;9(11):e111823. doi: 10.1371/journal.pone.0111823