This retrospective study included 209 elderly patients with gynecological cancer who received paclitaxel and carboplatin chemotherapy at The Affiliated Hospital, Southwest Medical University (Luzhou, China) between May 2019 and July 2023. Inclusion criteria were as follows: Patients diagnosed with cervical, endometrial or ovarian cancer confirmed by pathology; patients who underwent a chemotherapy regimen of paclitaxel and carboplatin; and patients aged ≥60 years without communication barriers. Exclusion criteria included patients with contraindications to chemotherapy or those missing clinical data. Additionally, potential confounding factors, such as polypharmacy, performance status and lifestyle variables, were considered when evaluating risk factors for CINV.

The 209 patients had a mean age of 72.35±4.91 years, with an age range from 60 to 85 years. Detailed baseline characteristics, including body mass index (BMI), Eastern Cooperative Oncology Group (ECOG) performance status (15), comorbidities (e.g., hypertension, diabetes and coronary heart disease) and chemotherapy details, are summarized in Table I. Since this study focused exclusively on patients with gynecological cancer, only female participants were included in the analysis.

Patients' clinical data were collected through the electronic medical record system. Data on demographic characteristics, comorbidities, chemotherapy details and psychological factors (e.g., anxiety levels) were included. For psychological distress, anxiety levels were assessed using validated scales (16,17), and data on sleep duration and history of CINV were also recorded.

Psychological factors, including anxiety, depression and coping mechanisms, were evaluated using validated self-reported tools to ensure reliable and objective measurements. Anxiety levels were assessed using the Self-Rating Anxiety Scale (16), a 20-item tool scored on a 4-point Likert scale, with standardized scores ranging from 25 to 100. Cutoff levels were defined as follows: Normal (<50), mild anxiety (50–59), moderate anxiety (60–69) and severe anxiety (≥70). Similarly, depression was assessed using the Self-Rating Depression Scale (18), which follows the same scoring structure, with cutoff levels categorized as normal (<53), mild depression (53–62), moderate depression (63–72) and severe depression (≥73). Coping mechanisms were assessed using a structured questionnaire that categorized patients into adaptive or maladaptive groups based on their responses. Maladaptive coping strategies were associated with a higher risk of CINV. These tools provided a standardized approach for stratifying patients into risk groups based on psychological factors, as reflected in the analysis.

CINV was assessed using the Multinational Association for Supportive Care in Cancer (MASCC) Antiemesis Tool (MAT), which has been validated for use in various patient populations, including elderly patients with cancer (19). Acute and delayed CINV were evaluated on the first and seventh days of the chemotherapy cycle, respectively. Based on the MASCC scores, patients were classified into the CINV group (n=76) and the non-CINV group (n=133) using a cutoff value of ≥2 to define the presence of clinically significant CINV. This threshold aligns with prior studies validating the MAT in cancer populations and ensures reliable classification. The limitations of using the MAT, especially in elderly populations, were also considered in the analysis, including potential biases related to cognitive impairments or comorbidities.

All statistical analyses were performed using the R programming language (version 4.3.0; R Core Team), with specific packages employed to facilitate various aspects of the analysis. The stats package was used for t-tests, χ² tests, and logistic regression analyses to identify predictors of CINV. The tableone package was applied to generate descriptive statistics, including patient baseline characteristics (Table I) and subgroup comparisons (Table II). Additionally, the car package was utilized to check multicollinearity during logistic regression analysis presented in Table III. Descriptive statistics summarized patient characteristics. Differences between the CINV and non-CINV groups were evaluated using independent t-tests for continuous variables. For categorical variables, χ² tests were primarily used, but Fisher's exact test was applied when expected frequencies in contingency tables were low, to ensure compliance with statistical assumptions. Percentages for categorical variables were calculated based on the total sample size (n=209) rather than group-specific totals. Univariate and multivariate logistic regression analyses were conducted to identify risk factors for CINV. Variables with a P<0.1 in the univariate analysis were included in the multivariate analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated. P<0.05 was considered to indicate a statistically significant difference. Sensitivity analyses were also performed to assess the impact of potential confounders on the primary outcomes.

In line with the study by Yoshikawa et al (7), potential interventions targeting psychological distress, such as anxiety reduction techniques, were considered as part of the discussion. While not directly incorporated into the study's design, this is an area of ongoing investigation for mitigating CINV risk.

Table I provides a detailed overview of the baseline characteristics of the 209 elderly patients with gynecological cancer included in this study. The mean age was 72.35±4.91 years, with a BMI of 22.68±2.96. Most patients (73.2%) had an ECOG performance status of 0-1, indicating relatively good functional status, while 26.8% had an ECOG score of ≥2, reflecting significant impairment. In terms of comorbidities, 26.3% of patients had hypertension, 20.6% had diabetes and 14.4% had coronary heart disease. Lifestyle factors such as smoking (0.5%) and alcohol consumption (4.8%) were rare among the cohort. A history of motion sickness and morning sickness was reported by 11.5 and 23.0% of the patients, respectively. Regarding chemotherapy sessions, 50.2% of patients underwent one session, 29.7% had two sessions and 20.1% had three sessions. Tumor location was predominantly the uterus (57.4%), followed by the ovaries (41.1%) and fallopian tubes (1.4%). For TNM staging (20), 67.5% were in stages III–IV and 32.5% were in stages I–II. Pre-chemotherapy sleep duration was <7 h in 45.5% of patients and severe pre-chemotherapy anxiety was reported by 51.7%. A history of CINV was reported in 35.9% of patients and antiemetics were used by 22.5%. These baseline characteristics provided a comprehensive understanding of the cohort and highlighted potential risk factors for CINV, such as age, psychological state and sleep duration.

Table II compares the characteristics of patients in the CINV group (n=76) and the non-CINV group (n=133). The mean age in the CINV group was significantly younger (70.26 years) compared with that in the non-CINV group (73.54 years) (P<0.001). While BMI did not differ significantly between the two groups, hypertension was more prevalent in the CINV group (13/76; 17.1%) than that in the non-CINV group (42/133; 31.6%) (P=0.022). Pre-chemotherapy sleep duration and pre-chemotherapy anxiety levels showed significant differences between the groups. A larger proportion of patients in the non-CINV group reported ≥7 h of sleep [67.7% (43/133) vs. 31.6% (24/76); P<0.001]. Similarly, no or mild anxiety was more common in the non-CINV group (57.1%; 76/133) than in the CINV group (32.9%; 25/76) (P<0.001). Other factors, such as ECOG scores, diabetes, coronary heart disease, smoking, drinking, motion sickness history and morning sickness history, did not differ significantly between the groups. These findings emphasize the potential role of modifiable psychological factors, such as sleep duration and anxiety, in influencing CINV risk, as well as demographic and clinical factors such as age and hypertension.

Table III presents the univariate and multivariate logistic regression analyses for risk factors associated with CINV. In the univariate analysis, age was significantly associated with CINV, with each additional year increasing the odds of occurrence by 16.4% (OR, 1.164; 95% CI, 1.088-1.244; P<0.001), and hypertension was also a significant factor, with hypertensive patients having more than twice the odds of developing CINV compared with non-hypertensive patients (OR, 2.237; 95% CI, 1.111-4.505; P=0.024). Conversely, shorter pre-chemotherapy sleep duration (<7 h) was associated with a reduced likelihood of CINV (OR, 0.221; 95% CI, 0.120-0.404; P<0.001), and severe anxiety before chemotherapy significantly increased the risk of CINV (OR, 2.720; 95% CI, 1.509-4.902; P<0.001). In the multivariate analysis, after adjusting for potential confounders, age remained a significant predictor of CINV, with a slightly lower adjusted OR (OR, 1.134; 95% CI, 1.059-1.215; P<0.001). Pre-chemotherapy sleep duration retained its protective association (OR, 0.285; 95% CI, 0.149-0.545; P<0.001) and severe pre-chemotherapy anxiety continued to significantly increase the risk of CINV (OR, 2.368; 95% CI, 1.237-4.534; P=0.009). However, hypertension, which was significant in the univariate analysis, lost its statistical significance in the multivariate model (OR, 1.568; 95% CI, 0.713-3.444; P=0.263). Other factors, including BMI, ECOG performance status, diabetes, coronary heart disease, motion sickness history, morning sickness history, tumor location, TNM staging, number of chemotherapy sessions, history of CINV and use of antiemetics, were not significantly associated with CINV in either univariate or multivariate analyses. These results highlight the strong predictive roles of age, sleep duration and anxiety level in the development of CINV.