Three cognitive performance profiles emerge in older adults on day seven after elective surgery

Older adults recovering from elective surgery do not all follow the same cognitive path in the first week after an operation. A prospective observational study from Jingzhou, China, has found that day-seven scores on the Mini-Mental State Examination (MMSE) cluster into three distinct, severity-ordered profiles among patients aged 65 and over, and that demographic, behavioral, clinical, and social factors differ across those profiles. The work, led by researchers at Jingzhou Central Hospital and Yangtze University, applies a data-driven statistical technique to a question that is usually answered with a single cutoff score.

The study enrolled 308 consecutively recruited patients aged 65 years or older who underwent elective surgery under general anesthesia. Cognition was assessed on postoperative day seven with the MMSE, and the researchers organized five study-defined postoperative domain scores into a latent profile analysis, a method that searches for hidden subgroups within a population rather than assuming everyone belongs to the same distribution. Candidate correlates were organized using the health ecological model, which treats health outcomes as the product of layered influences ranging from individual biology to social circumstances.

The analysis identified three profiles arranged along a severity gradient. The high-performance group was the largest, containing 153 patients, or 49.7 percent of the sample. The low-performance profile included 83 patients, or 26.9 percent, and the intermediate profile included 72 patients, or 23.4 percent. Rather than a binary distinction between patients who pass and fail a screening threshold, the results suggest that early postoperative cognition in this population is better described as a graded spectrum with two lower-performing subgroups that differ in size and in their associated characteristics.

The statistical approach also allowed the team to examine which factors travel with membership in each profile. Univariate tests were followed by an exploratory multinomial regression weighted by posterior probabilities, meaning that each patient’s probability of belonging to each profile informed the estimates rather than forcing a hard assignment. Two demographic variables stood out as associated with both lower-performance profiles: older age and lower education. This pattern is consistent with the well-documented role of cognitive reserve, in which educational attainment is thought to buffer the expression of cognitive difficulty even when underlying brain changes are similar.

The low-performance profile carried a broader set of associations. Greater dependence in activities of daily living, lower income, and living alone were each linked to membership in this lowest-scoring group. These are social and functional markers rather than surgical variables, and their convergence in one profile fits the ecological framing of the study: the circumstances in which an older person lives and the degree of independence they maintain appear alongside age and education in distinguishing who falls into the lowest-performing subgroup on day seven.

The intermediate profile showed a different signature. Poorer sleep was the factor associated with membership in this middle group, distinguishing it from both the high and low profiles. That a behavioral variable aligned with the intermediate rather than the lowest profile is one of the more intriguing features of the findings, because it suggests the two lower-performing subgroups are not simply more severe versions of the same phenomenon but may reflect different constellations of contributing circumstances.

The authors are explicit about what the design cannot show. The profiles are cross-sectional snapshots of day-seven performance, and the study did not retain baseline cognitive scores for comparison. Without a preoperative measurement, the profiles cannot demonstrate postoperative decline; a low score on day seven may reflect longstanding cognitive status as much as anything that happened around surgery. The study also cannot establish a clinical diagnosis, and because all participants underwent general anesthesia, no comparison group exists to attribute any pattern specifically to the anesthetic technique itself.

Those limitations shape the study’s conclusions. The authors state that longitudinal studies with retained baseline scores, appropriate comparison groups, and broader neuropsychological assessment are required before any profile-specific interventions can be recommended. The MMSE, while widely used, is a brief screening instrument, and a fuller neuropsychological battery could capture domains that a single composite score misses. The observational design, meanwhile, means the reported associations are not evidence of causal effects, and the exploratory regression should be read as hypothesis-generating rather than confirmatory.

What the study does offer is a structured map of early postoperative cognitive performance in a defined population, together with a set of measurable characteristics that distinguish the subgroups. If future longitudinal work confirms these patterns with baseline data and broader testing, the profile approach could help identify which older surgical patients might benefit from closer cognitive follow-up, and which social and functional circumstances deserve attention in perioperative care planning. For now, the findings stand as an observational description of a severity gradient, not a verdict on anesthesia, surgery, or decline.

Subject of Research: Latent profiles of day-seven cognitive performance and associated ecological factors in older adults after elective surgery

Article Title: Latent profiles of day-7 cognitive performance and associated ecological factors in older adults after elective surgery: an observational study

Article References: Liao, Y., Zhang, X., Zhang, Y., Shi, L., & Liu, Y. (2026). Latent profiles of day-7 cognitive performance and associated ecological factors in older adults after elective surgery: an observational study. Frontiers in Psychiatry, 17, Article 1975739. https://doi.org/10.3389/fpsyt.2026.1975739

Image Credits: AI Generated

DOI: 10.3389/fpsyt.2026.1975739

Keywords: cognitive performance, elective surgery, older adults, latent profile analysis, MMSE, postoperative cognition, health ecological model, perioperative nursing, observational study, general anesthesia, Latent, profiles