Perioperative neurocognitive disorder, particularly postoperative cognitive impairment, is common and associated with multiple medical and social adversities, although data from China are lacking. To examine the incidence, trajectory, and risk factors for subjective cognitive and short-term memory impairment after surgery in the Chinese population. This cohort study used data from the China Surgery and Anesthesia Cohort to assess surgical patients aged 40 to 65 years from 2 medical centers between July 15, 2020, and March 31, 2023, with active follow-up within 1 year after the surgery. Of 11 158 patients who were successfully recruited (response rate, 94.4%), 10 149 participants were eligible and available for analysis. From this population, separate cohorts were constructed for analyzing subjective cognitive impairment (8105 noncardiac and 678 cardiac surgery patients) and short-term memory impairment (5246 noncardiac and 454 cardiac surgery patients). Twenty-four potential risk factors regarding comorbidities, preoperative psychological conditions, anesthesia- or surgery-related factors, and postsurgical events were included. Outcomes included subjective cognitive function measured by the 8-Item Informant Interview to Differentiate Aging and Dementia (AD8; scores range from 0 to 8, with higher scores indicating more severe cognitive impairment) and short-term memory measured by the 3-Word Recall Test (TRT; scores range from 0 to 3, with lower scores indicating more severe short-term memory impairment) at 1, 3, 6, and 12 months after noncardiac and cardiac surgery. Generalized linear mixed models were used to identify risk factors associated with the presence of AD8 (score ≥2) or TRT (score <3) abnormality as well as the aggressively deteriorative trajectories of those cognitive measurements. For noncardiac surgery patients, the AD8 analysis included 8105 patients (mean [SD] age, 52.3 [7.1] years; 3378 [41.7%] male), and the TRT analysis included 5246 patients (mean [SD] age, 51.4 [7.0] years; 1969 [37.5%] male). The AD8 abnormality incidence rates after noncardiac surgery increased from 2.2% (175 of 8105) at 7 days to 17.1% (1059 of 6191) at 6 months after surgery, before appearing to decrease. In contrast, the TRT abnormality incidence rates followed a U-shaped pattern, with the most pronounced incidence rates seen at 7 days (38.9% [2040 of 5246]) and 12 months (49.0% [1394 of 2845]). Similar patterns were seen among cardiac surgery patients for the AD8 analysis (678 patients; mean [SD] age, 53.2 [6.3] years; 393 [58.0%] male) and TRT analysis (454 patients; mean [SD] age, 52.4 [6.4] years; 248 [54.6%] male). Among noncardiac surgery patients, the top risk factors for aggressively deteriorative AD8 trajectory and for AD8 abnormality, respectively, after surgery were preoperative sleep disturbances (Pittsburgh Sleep Quality Index ≥16 vs 0-5: odds ratios [ORs], 4.04 [95% CI, 2.20-7.40] and 4.54 [95% CI, 2.40-8.59]), intensive care unit stay of 2 days or longer (ORs, 2.43 [95% CI, 1.26-4.67] and 3.07 [95% CI, 1.67-5.65]), and preoperative depressive symptoms (ORs, 1.76 [95% CI, 1.38-2.24] and 2.23 [95% CI, 1.79-2.77]). Analyses for TRT abnormality and trajectory, as well as the analyses conducted among cardiac surgery patients, found fewer associated factors. This cohort study of middle-aged Chinese surgery patients found subjective cognitive and short-term memory impairment within 12 months after both cardiac and noncardiac surgery, with multiple identified risk factors, underscoring the potential of preoperative psychological interventions and optimized perioperative management for postoperative cognitive impairment prevention.