Development and validation of a risk prediction model for cognitive impairment in breast cancer patients

Abstract

Background

Breast cancer patients often experience cognitive impairment as a complication during treatment, which seriously affects their quality of life. This study aimed to assess the risk factors associated with cognitive impairment in breast cancer patients and to construct and validate a nomogram model to predict cognitive impairment in this population.

Methods

In this study, we used a convenience sampling method to select 423 breast cancer patients who attended the Department of Breast Surgery at the First Hospital of Jinzhou Medical University from September 2023 to March 2024. We analyzed these patients’ cognitive impairment risk factors through LASSO regression and logistic regression analysis to develop a predictive model. The model was evaluated using the area under the curve (AUC) from the receiver operating characteristic (ROC) curve and the calibration curve and decision curve analysis.

Results

This study found a prevalence of cognitive impairment of 19.62% among breast cancer patients. A nomogram model was developed based on six influencing factors: age, educational level, pathological type, treatment program, emotional state, and fatigue. The area under the curve (AUC) for the model’s training and validation groups was 0.944 and 0.931, respectively. The model calibration curves showed a high degree of consistency, and the decision curve analysis (DCA) indicated good clinical applicability of the model.

Conclusions

This nomogram demonstrates good discrimination, calibration, and clinical applicability, making it a more intuitive predictor of the risk of cognitive impairment in breast cancer patients.

Introduction

Breast cancer has been reported as the most prevalent cancer worldwide [1]. According to data from 2022, the incidence of female breast cancer in China ranks among the top five malignant tumors [2]. Currently, breast cancer patients are treated with various methods, including surgery, chemotherapy, radiotherapy, endocrine therapy, and targeted therapy. While these treatments can achieve good results, they may also lead to various adverse reactions [3]. Among these, cognitive impairment has been recognized as a common complication during treatment for breast cancer patients [3].

Cognitive impairment in breast cancer patients is primarily characterized by changes in attention, memory, executive function, and processing speed [4]. A cross-sectional study of breast cancer patients noted that approximately 17% to 75% of those treated with chemotherapy experienced changes in cognitive domains [5]. Chemotherapy-induced cognitive impairment is associated with alterations in brain grey matter volume, grey matter density, and cortical thickness [6]. It is also linked to disruptions of the blood–brain barrier caused by chemotherapeutic drugs, systemic and chronic inflammation, and accelerated cellular senescence [7]. Additionally, the cancer itself, as well as other treatments such as radiotherapy and endocrine therapy, may lead to similar cognitive impairments [7, 8]. As a result of these altered cognitive functions, patients may experience a decline in work performance and quality of life, which could lead to adjustments in treatment decisions [3, 5].

In China, the prevalence of cognitive impairment in breast cancer patients ranges from 24% to 81.82% [9, 10]. Országhová et al. showed that factors such as age, anxiety, and depression affect cognitive functioning in cancer patients [11]. Meanwhile, Jansen’s study points out that there may be reversible changes in cognitive impairment among breast cancer patients [12]. Therefore, it is essential to identify early predictors of cognitive impairment in breast cancer patients and to recognize high-risk groups. However, there is still limited research on risk prediction tools for cognitive impairment in this population.

The Theory of Unpleasant Symptoms (TOUS) consists of three core concepts: symptoms, influencing factors, and performance outcomes [13]. Symptoms are subjective experiences of individuals that can be influenced by pathological, physiological, psychological, and environmental factors. Manifestation outcomes include functional activities and cognitive performance. These three core concepts are interconnected and influence each other.

Nomograms are based on multifactor regression analyses and can show quantitative relationships between multiple variables [14]. They visualize the regression equation’s results in graphical form for enhanced readability. The nomogram model can predict the probability of a particular event occurring relatively quickly, making it highly operational.

Due to the current lack of accurate predictive tools for cognitive impairment in breast cancer patients, this study aimed to investigate the risk factors associated with cognitive impairment in these patients by combining a literature review with the theory of unpleasant symptoms and developing a visual nomogram model to identify patients at risk of cognitive impairment in advance. By considering the different predictive factors, we can implement timely, targeted, and effective interventions to minimize the likelihood of cognitive changes and reduce the negative impact of cognitive impairment on patients.

Methods

Study design and participants

This study used convenience sampling to investigate breast cancer patients who attended the Breast Surgery Department of the First Affiliated Hospital of Jinzhou Medical University from September 2023 to March 2024. The inclusion criteria were as follows: ① Age ≥ 18; ② Comply with the diagnostic criteria of the Chinese Anti-Cancer Society Breast Cancer Diagnostic Guidelines and Criteria, with the time of diagnosis being more than two weeks prior [15, 16]; ③ Basic language comprehension and response skills; ④ Voluntarily participated in this study and signed an informed consent form; ⑤ Complete medical records. Exclusion criteria included: ① History of neurological or psychiatric disorders; ② Those taking medications related to cognitive function; ③ Those with intracranial abnormalities or intracranial metastases on cranial MRI or CT; ④ Advanced malignant disease. To estimate the sample size for this study, we used a calculation based on the sample size formula for epidemiological cross-sectional studies. The formula is as follows: $n=\frac{z_{1-\alpha /2}^2*P\left(1,,-,,P\right)}{d^2}*\left(1,,+,,20,\%\right)$ In this formula, n represents the sample size, z is the statistic, 1-α/2 denotes a two-sided test, P is the prevalence, and d is the precision. Previous studies have reported a prevalence of 16.9% for cognitive impairment in breast cancer patients [17]. Assuming a confidence interval of 95% and a d of 5%, while considering the two-sided test and a 20% non-response rate, the sample size for this study was calculated to be at least 259 cases. Ultimately, 423 breast cancer patients were included in this study. They were randomly divided into a training set (297 cases) and a validation set (126 cases) in a ratio of 7:3.

Survey instruments

This study developed predictors of cognitive impairment in breast cancer patients by comprehensively searching, reading, collating, and discussing relevant literature. At the same time, we obtained general profile information about the patients from the hospital’s electronic medical record system.

General information questionnaire

The researcher designed a general information questionnaire to collect data on age, marital status, occupation, educational level, methods of medical payment, disease staging, pathological type, and treatment program.

Functional assessment of chronic illness therapy-fatigue

The scale was developed by Cella et al. [18] and translated into Chinese and validated in cancer patients by Hou et al. [19]. It consists of 13 items and is a unidimensional scale [19]. A 5-point Likert scale was used, with scores ranging from 0 to 4 for “not at all” to “very much.” Positive scores were assigned to items An5 and An7, while negative scores were assigned to the other items. Total scores range from 0 to 52, with higher total scores indicating less fatigue. The Cronbach´s α in this study was 0.903.

Positive and negative affect schedule

The scale was developed by Watson and revised in Chinese by Qiu et al. [20, 21]. The scale consists of 18 items and includes two dimensions: positive and negative emotions [21]. A 5-point Likert scale was used, with scores assigned from 1 to 5 for “very slight or not at all” to “very strong,” respectively. Higher scores for positive or negative emotions indicated higher levels of those emotions. The Cronbach´s α for this study’s positive and negative subscales were 0.887 and 0.872, respectively.

Perceived social support scale

The scale was developed by Zimet et al. [22], sinicized by Jiang et al. [23], and evaluated by Zhang and other scholars for application in Chinese inpatients [24]. The scale consists of 12 items, including three dimensions: “family support,” “friend support,” and “other support” [22]. A 7-point Likert scale was used, with scores ranging from 1 to 7 for “strongly disagree” to “strongly agree,” resulting in a total score ranging from 12 to 84. Higher scores indicate a higher level of social support for the individual’s understanding. The Cronbach´s α for this scale in this study was 0.871.

Montreal cognitive assessment

The scale was developed by Nasreddine et al. based on clinical evidence concerning the MMSE scale [25]. It was sinicized and applied by Wang et al. [26]. The scale includes eight cognitive domains: memory functioning, visuospatial functioning, executive functioning, attention, numeracy, language functioning, time orientation, and place orientation [26]. The total score ranges from 0 to 30, with a score of ≥ 26 indicating cognitive normality, a cut-off value of 25 for individuals with ≤ 12 years of education, and lower scores indicating worse cognitive functioning. The Cronbach´s α for this scale in this study was 0.745.

Data analysis

We used SPSS 26.0 and R 4.4.1 to perform statistical analyses on the data from this study. Continuous variables that did not follow a normal distribution were described using the median and interquartile range (M[P25, P75]), while categorical variables were presented as frequency (n) and percentage (%). We used the Chi-square (χ²) and Mann–Whitney U tests to compare groups.

We used LASSO regression to screen for predictors. The optimal parameters were determined using the tenfold cross-validation method, with the penalty coefficient lambda (lambda.1se) corresponding to one standard error from the minimum mean square error as the optimal adjustment parameter. The screened predictors were included as independent variables in binary logistic regression analyses, incorporating those with P values < 0.05 into the model construction.

We plotted the predictive model as a visual nomogram. The model was internally validated through Bootstrap repetitive sampling, performed 1,000 times in the training set and externally validated in the validation set. The area under the receiver operating characteristic (ROC) curve (AUC) assessed the model’s discriminatory power. Calibration curves evaluated the consistency of the model. The clinical applicability of the model was assessed using decision curve analysis (DCA).

Results

Participant characteristics

We included a total of 423 breast cancer patients in this study. In the training group of 297 patients, the median age was 51 (44.5, 62.5). Of these, 55 patients (18.52%) were diagnosed with cognitive impairment. In the validation group of 126 patients, the median age was 53.5 (44, 63). Of these, 28 cases (22.22%) were diagnosed with cognitive impairment. The difference between the training and validation sets regarding cognitive impairment-related variables, such as age, treatment regimen, cancer stage, and pathology type, was insignificant (P > 0.05). See Table 1.

Table 1.

Comparison of baseline characteristics between training and validation groups

Variable	Training (n = 297)	Validation (n = 126)	P value
Age	51(44.5,62.5)	53.5(44,63)	0.842
Age group(years)			0.809
< 60	192(64.6)	83(65.9)
≥ 60	105(35.4)	43(34.1)
Educational level			0.774
Primary and lower	54(18.2)	27(21.4)
Junior	109(36.7)	42(33.3)
High school or trade school	67(22.6)	28(22.2)
Junior college	39(13.1)	18(14.3)
Undergraduate and above	28(9.4)	11(8.7)
Marital status			0.768
Unmarried	4(1.3)	3(2.4)
Married	270(90.9)	111(88.1)
Divorcee	12(4)	4(3.2)
Widowhood	11(3.7)	8(6.3)
Occupation			0.976
Jobless	83(27.9)	34(27)
Labor	141(47.5)	62(49.2)
Mental labor	73(24.6)	30(23.8)
Monthly family income			0.564
< 1000	13(4.4)	7(5.6)
1000–3000	48(16.2)	23(18.3)
3001–5000	137(46.1)	47(37.3)
5001–10000	74(24.9)	32(25.4)
> 10,000	25(8.4)	17(13.5)
Disease staging			0.690
I	115(38.7)	48(38.1)
II	89(30.0)	41(32.5)
III	52(17.5)	27(21.4)
IV	41(13.8)	10(7.9)
Pathological type			0.468
Triple-negative breast cancer	25(8.4)	8(6.3)
Non-triple-negative breast cancer	272(91.6)	118(93.7)
Treatment program			0.135
Chemotherapy	175(58.9)	84(66.7)
Non-chemotherapy	122(41.1)	42(33.3)
Methods of medical payment			0.869
Self-financed	21(7.1)	8(6.3)
New agricultural cooperation	118(39.7)	49(38.9)
City and town health insurance	55(18.5)	26(20.6)
Workers’ medical insurance	103(34.7)	43(34.1)
Emotional state			0.941
Positive emotion	178(59.9)	76(60.3)
Negative emotion	119(40.1)	50(39.7)
Fatigue	38(33,42)	38(34,43)	0.401
Appreciating social support	54(50,60)	54(51,60)	0.877

Continuous variables that do not fit a normal distribution are denoted as M (P25, P75), while categorical variables are denoted by frequency (n) and percentage (%)

Risk factors for cognitive impairment

In this study, we performed LASSO regression analysis on the training set data and identified seven variables with non-zero regression coefficients, as shown in Fig. 1A and B. We then conducted binary logistic regression analysis based on the LASSO results to determine the risk factors for cognitive impairment in breast cancer patients. We screened six predictors: age, educational level, pathological type, treatment program, emotional state, and fatigue. Refer to Table 2 for details, and see Table 3 for important variable assignments.

Fig. 1.

A LASSO regression coefficient path diagram. B LASSO regression cross-validation curves. The optimal lambda is determined using the ten-fold cross-validation method. The left vertical dashed line indicates the minimum value of the cross-validation error, while the right vertical dashed line indicates the minimum value of the cross-validation error within one standard deviation

Table 2.

Predictors of cognitive impairment

Predictors		B	P	OR	95%CI
Age(years)		1.023	0.049	2.782	1.004–7.710
Educational level	Primary and lower	Reference
	Junior	-1.181	0.025	0.307	0.110–0.860
	High school or trade school	-2.136	0.025	0.118	0.018–0.766
	Junior college	-2.044	0.045	0.130	0.018–0.957
	Undergraduate and above	-2.722	0.038	0.066	0.005–0.864
Pathological type		1.937	0.005	6.940	1.810–26.612
Treatment program		1.792	0.003	5.999	1.808–19.907
Emotional state		1.388	0.005	4.008	1.517–10.588
Fatigue		-0.121	0.001	0.886	0.825–0.951
Constant		0.436	0.838	1.547

Table 3.

Variable assignment table

Variable	Assignment method
Age(years)	< 60 = 1; ≥ 60 = 2
Educational level	Primary and lower = 1; Junior = 2; High school or trade school = 3; Junior college = 4; Undergraduate and above = 5
Pathological type	Non-triple-negative breast cancer = 0; Triple-negative breast cancer = 1
Treatment program	Non-chemotherapy = 0; Chemotherapy = 1
Emotional state	Positive emotion = 1; Negative emotion = 2
Fatigue	Original value inputs

Development of predictive models

In this study, we included six independent factors to establish a nomogram for predicting cognitive impairment in breast cancer patients. We summed the scores obtained from each variable in the nomogram, and the total score was used to determine the probability of cognitive impairment in these patients (see Fig. 2).

Fig. 2.

Nomogram for detecting cognitive impairment in breast cancer patients. Each factor corresponds to a single point at the top of the graph for an individual patient. We sum the scores from all individual points to obtain the total score for that patient. The probability of cognitive impairment in a breast cancer patient is derived by projecting the total score onto the incidence risk axis

Validation of the prediction model

In this study, we validated the model using data from both the training and validation sets. The results showed an AUC value of 0.944 (95% CI: 0.919–0.970) in the training set, with an optimal cutoff of 0.762, a sensitivity of 0.927, and a specificity of 0.835 (Fig. 3A). In the validation set, the AUC value was 0.931 (95% CI: 0.878–0.984), with an optimal cutoff of 0.755, a sensitivity of 0.857, and a specificity of 0.898 (Fig. 3B). These findings suggest that the nomogram model effectively discriminates between cognitive and non-cognitive impairments in early screening.

Fig. 3.

A Training group ROC curve; B Validation group ROC curve. The X-axis indicates specificity, and the Y-axis indicates sensitivity

Calibration curves were used to assess the degree of model fit and prediction accuracy. The calibration curve represents a scatter plot of the actual probability of occurrence versus the predicted probability [27]. The closer the calibration curve is to a fitted straight line, the better the expected value matches the measured value, indicating greater accuracy of the model [27]. The results show that the calibration curves fit well with the actual curves, suggesting that the predicted probability of cognitive impairment aligns with the exact probability for both data sets (see Fig. 4A, B).

Fig. 4.

A Training group calibration curve; B Validation group calibration curve. The horizontal axis represents the predicted probability of cognitive impairment, while the vertical axis represents the actual probability of cognitive impairment. The diagonal dashed line represents the perfect state of the ideal model. The solid line represents the nomogram’s performance; when it is closer to the dashed line, it indicates better prediction accuracy

DCA was used to assess the model’s value for clinical applications. DCA evaluates the risks of clinical interventions while considering the balance of benefits that decision-makers require [28]. The horizontal axis represents the threshold probability, the vertical axis represents the net benefit, and the lines between the horizontal and vertical axes illustrate the benefits of the different predictor variables [28]. The results indicate that the net clinical benefit of the prediction model for both the modeling and validation sets is higher than that of the two extreme cases, suggesting that the model is adequate for predicting cognitive impairment and has good clinical applicability. See Fig. 5A and B.

Fig. 5.

A Training group decision curve; B Validation group decision curve. The horizontal axis represents the threshold probability, while the vertical axis represents the net benefit. The black horizontal line indicates the prediction that all patients have no cognitive impairment, whereas the grey line indicates that all patients are predicted to have cognitive impairment. The red curve represents the net benefit of the nomogram

Discussion

Although the survival and mortality rates of breast cancer patients have improved significantly through preventive interventions, cognitive impairment remains one of the common complications affecting these patients, seriously impacting their daily activities and quality of life [29]. Previous studies have focused on the risk factors for cognitive impairment in breast cancer patients, providing reliable support for the construction of the risk prediction model presented in this paper [9, 10]. Therefore, this study identified the risk factors associated with cognitive impairment in breast cancer patients and constructed a risk prediction model based on these factors. Specifically, it was found that age, educational level, pathological type, treatment program, emotional state, and fatigue independently influenced cognitive impairment in this population. This information is crucial for the early detection and prevention of cognitive impairment.

The prevalence of cognitive impairment in breast cancer patients in this study was 19.62%, which is slightly higher than that reported by Vearncombe in chemotherapy breast cancer patients [17]. These results are reasonable, as the patients in this study were generally older and included treatment modalities beyond chemotherapy.

In this study, age was identified as a significant risk factor for cognitive impairment in breast cancer patients, with increasing age correlating with a higher risk. This finding is supported by research conducted by Mandelblatt et al., which indicates that older age is associated with lower scores across all cognitive domains [30]. As we age, various mechanisms can affect body cells through oxidative and inflammatory stress responses, resulting in neurological damage [30, 31]. Additionally, aging leads to changes in the white and gray matter of the frontal lobe, significantly impairing cognitive functions such as attention and memory [32]. Therefore, paying particular attention to older patients in clinical practice is essential. Studies indicate that physical activity and multidisciplinary interventions can help protect cognitive function in older adults [33]. Consequently, older breast cancer patients are encouraged to engage in regular exercise to reduce the risk of cognitive impairment and alleviate its adverse effects.

In addition, the results of this study show that education level is also a risk factor for the development of cognitive impairment in breast cancer patients. Educational level positively impacts cognitive function, with more educated patients able to delay subjective cognitive decline more effectively [34]. Education is an important proxy for cognitive reserve, and cognitive impairment may be related to glucose metabolism in the temporoparietal lobe. The cognitive reserve of patients with higher education backgrounds can partially compensate for cognitive impairment caused by glucose hypometabolism in the temporoparietal lobe [35]. A longitudinal study by Perrier et al. also showed that cognitive reserve plays a compensatory role in chemotherapy-induced hippocampal atrophy and cognitive impairment in breast cancer patients [36]. Therefore, increasing the knowledge base and learning ability can be effective in preventing cognitive impairment in breast cancer patients. A randomized controlled trial by Masika et al. showed that patients in the intervention group who received a visual arts program had better cognitive functioning than those in the control group [37]. Because visual arts programs use simple elements such as dots, lines, and curves to create beautiful shapes, the literacy requirements for participants are broader [37]. During clinical care, patients can be encouraged to participate in education-related activities such as reading and film-watching, and simple creative activities can be organized to stimulate cognitive function and increase the patients’ cognitive reserve, thus mitigating the negative impact of disease and treatment on cognitive function.

This study found that patients with triple-negative breast cancer are 6.94 times more likely to develop cognitive impairment compared to those with non-triple-negative breast cancer. Triple-negative breast cancer is characterized by the non-expression of estrogen receptors (ER), progesterone receptors (PR), and human epidermal growth factor receptor 2 (Her-2) [38]. Estrogen and progesterone are steroid hormones that exert their effects primarily by binding to their respective receptors, with the central nervous system being a critical target for these hormones [39]. Research has shown that significant levels of ER and PR are present in the prefrontal cortex and hippocampus, where these hormones regulate synaptic plasticity and neurodegenerative changes through receptor activation. Additionally, Her-2 enhances neurotransmitter receptor expression by interacting with neuromodulatory proteins [40–42]. Consequently, patients with triple-negative breast cancer are particularly vulnerable to cognitive impairment due to disrupted endocrine hormone regulation resulting from the lack of ER, PR, and Her-2 expression. Therefore, it is crucial to promptly diagnose the pathological type of breast cancer and implement appropriate therapeutic measures to mitigate the negative impact on the patient’s cognitive function.

Additionally, this study identified chemotherapy as a risk factor for the development of cognitive impairment in breast cancer patients, aligning with findings from most existing research [42–44]. The pathophysiological basis of cognitive impairment in these patients lies in altered cerebral grey matter volume, grey matter density, and cortical thickness in specific brain regions [6]. Chemotherapy leads to bilateral reductions in cerebral grey matter volume and decreased grey matter density in the frontal cortex and cerebellum, which mediate the relationship between chemotherapy, brain aging, and cognitive impairment [45, 46]. Furthermore, chemotherapy may result in metabolic disorders. Alterations in cytokines and receptors are associated with multiple domains of cognitive function. Cytokines crossing the blood–brain barrier can lead to neurotoxicity and inflammation, impairing cognitive function [47]. Chemotherapy may affect patients with grey matter changes, metabolic disorders, vascular damage, and the accelerated aging process [44]. As a result, breast cancer patients receiving chemotherapy are more likely to experience altered cognitive function. A systematic review and network meta-analysis report that non-pharmacological treatments such as psychotherapy, music therapy, exercise, and qigong can reduce chemotherapy-induced cognitive impairment [48]. Therefore, healthcare professionals should develop individualized treatment plans based on the patient’s condition and closely monitor their response to chemotherapy drugs. At the same time, caregivers should select appropriate intervention programs, such as music therapy and exercise, tailored to each patient’s individual needs to mitigate the possibility of cognitive impairment.

Breast cancer patients frequently experience negative emotions such as anxiety and depression due to the challenges of illness and life stressors, which can significantly impact their cognitive function. This study indicates that patients with higher levels of negative affect are more likely to exhibit altered cognitive functioning, a finding that aligns with previous research [44, 49]. Self-reported anxiety is associated with verbal working memory and objective cognitive performance in daily life. Anxiety symptoms depress activity in the precuneus/posterior cingulate region, resulting in impaired connectivity between this region and cognitive tasks, consequently manifesting as altered cognitive functioning [50]. In addition, depressive symptoms may increase cortical amyloid burden and contribute to a rise in inflammatory cytokines, which decreases synaptic plasticity [51]. Studies have shown that cognitive behavioral therapy can alleviate anxiety and depression while improving cognitive functioning, and reminiscence therapy also positively impacts cognition and mood [52]. Therefore, healthcare professionals should pay attention to patients’ mental health, encourage simple cognitive training, recommend appropriate types of training, and identify and intervene in cases of anxiety and depression promptly to minimize their impact on cognitive function.

Chemotherapy drugs may damage mitochondria, leading to mitochondrial dysfunction, a mechanism by which patients develop fatigue. This mechanism is similar to that underlying cognitive impairment caused by mitochondrial dysfunction [53]. Thus, there is a strong association between fatigue and cognitive impairment. In addition, cancer and treatment-related factors can cause patients to experience severe fatigue, which can negatively affect cognitive abilities such as executive function and recognition memory [54]. Studies have shown that positive thought-based therapies, such as positive stress reduction and positive cognitive therapy, can positively affect fatigue and cognitive status in breast cancer patients [55]. Therefore, care programs should be tailored to the patient’s specific needs and encourage the development of skills within their capabilities, such as yoga and mindfulness meditation. These interventions can help improve cognitive function by enhancing thinking, reactivity, and self-regulation, ultimately reducing symptoms of fatigue and lowering the risk of cognitive impairment.

Based on a multicenter cross-sectional study, Liu et al. constructed a risk prediction model for cognitive impairment in breast cancer patients undergoing chemotherapy [56]. The model was developed using four influencing factors: general information and fatigue [56]. However, cognitive impairment may also occur in patients receiving different treatment regimens, not just those undergoing chemotherapy. Therefore, it is essential to expand the study to include patient populations with varying treatment options. Our study explored the factors influencing cognitive functioning in breast cancer patients, focusing on pathological, physiological, psychological, and environmental factors. We constructed a nomogram model by combining patients’ general information with relevant scales. We validated this nomogram model internally and externally, demonstrating that it is highly discriminatory, well-calibrated, and clinically applicable. As a simple clinical tool for personalized assessment, it can help identify individuals at high risk of developing cognitive impairment, allowing healthcare professionals to tailor interventions to each patient’s situation.

Limitations

While the risk prediction model developed in this study shows promising performance, several areas for improvement remain. Firstly, since this is a single-center study, the data may be subject to bias. Additionally, the validation group also derived its data from the same research center, which could undermine the credibility of the findings. Therefore, expanding the sample size and conducting multicenter studies is essential to enhance the model’s accuracy. Secondly, as this study employs a cross-sectional design, it does not establish causal relationships between variables and outcomes. Future research could benefit from a prospective longitudinal study design to better understand the dynamic interactions among these variables over time. Finally, while this study included several relevant indicators, future investigations should consider incorporating additional demographic and socioeconomic variables to enhance the findings’ applicability in clinical practice.

Conclusion

This study developed and validated a visual nomogram model for predicting the risk of cognitive impairment in breast cancer patients. The model incorporates six easily accessible variables, demonstrating intense discrimination, calibration, and clinical applicability. It serves as a simple, convenient, and clinically useful tool for the early identification and screening of breast cancer patients at high risk for cognitive impairment.

Abbreviations

ROC

Receiver operating characteristic

AUC

The area under the receiver operating characteristic curve

DCA

Decision curve analysis

TOUS

The Theory of Unpleasant Symptoms

Authors’ contributions

Xm Z was involved in the study design, data collation and analysis, and writing of the original manuscript; Jy L was involved in data interpretation and project guidance; Zy D, Y Q, Xc L, and Gx Z were involved in data collection and collation; and Hx C guided and supervised the research process and revised the paper. All authors read and approved the final manuscript.

Funding

This work was supported by the doctoral research Fund of Wannan Medical College (Approval No.WYRCQD2023043).

Data availability

Due to the privacy of the participants involved in the study data, the datasets generated and/or analyzed in this study are currently not publicly available but can be obtained from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The study was conducted according to the Declaration of Helsinki and approved by the Ethics Committee of Jinzhou Medical University. It followed the principles of voluntariness and risk minimization. The questionnaire was anonymous to protect the privacy of the participants. All participants provided informed consent. All methods were performed according to relevant guidelines and regulations.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.