Latent profile analysis of potential health-promoting behaviors in patients after esophageal cancer surgery: a multicenter cross-sectional survey Authors: Juan Wang, Yuan Yu, Li Liu, Yan Miao, Yue Fu, Yanting Ning, Junli Liang Published: 2025 ABSTRACT ObjectivesThe objectives of this study were to identify potential health-promoting behaviors in patients after esophageal cancer surgery and explore their influencing factors by performing latent profile analysis.MethodsA convenience sampling method was used to select 438 postoperative patients with esophageal cancer from 14 cancer centers across the country as survey participants. A sociodemographic and disease-related information collection form, the revised version of the Health-Promoting Lifestyle Profile-II, the Social Support Rating Scale, and the Chinese version of the cancer self-efficacy scale were used. A latent profile analysis of potential health-promoting behaviors was conducted, and the factors influencing each category were identified through single-factor analysis and logistic regression analysis.ResultsThe health-promoting behaviors of patients after esophageal cancer surgery could be divided into three categories: low-level health-promoting behaviors (27.6%, n = 121), medium-level health-promoting behaviors (56.2%, n = 246), and high-level health-promoting behaviors (16.2%, n = 71). The results of logistic regression analysis revealed that age, education level, place of residence, the receipt of neoadjuvant treatment before surgery, the level of social support, and the level of self-efficacy after esophageal cancer surgery were factors influencing health-promoting behaviors (all P < 0.05).ConclusionsThe health-promoting behaviors of patients after esophageal cancer surgery are heterogeneous. Medical staff can perform individualized interventions for patients after esophageal cancer surgery according to the different types of influencing factors to improve their health-promoting behaviors. Introduction According to data from the World Health Organization in 2020, there were 324,000 new cases of esophageal carcinoma (EC) and 301,000 related deaths in China, accounting for 53.70% and 55.35%, respectively, of the global incidence of and deaths related to esophageal cancer [1]. Radical esophagectomy is a common way to treat esophageal cancer [2], but patients often experience many physical symptoms and psychological imbalances after surgery, such as persistent adverse reactions; worry and anxiety about the future and disease progression; and emotional disturbances such as anxiety, depression, and fear of recurrence, which may arise from postoperative physical discomfort, functional limitations, or concerns about prognosis, and a lack of social support [3]. At this time, in addition to achieving prolonged survival, patients should receive comprehensive physical, psychological, and social support [4]. Health-promoting behaviors refer to behaviors that lead individuals to realize their health potential, that is, they perform activities to achieve a higher level of health [5]. A higher level of health means that individuals achieve optimal physical, mental, and social well-being, including but not limited to following nutritional guidelines, engaging in regular physical activity, managing stress, and actively participating in follow-up care. The World Cancer Research Fund reported that the adoption of health-promoting behaviors by cancer patients can increase their treatment compliance, enhance their confidence in fighting cancer, and improve their prognosis and quality of life. Social support refers to the interactions between individuals embedded in the social environment and social network members, contributing to healthy living habits and patterns [6]; this is also a factor influencing the health-promoting behavior of patients after esophageal cancer surgery [7, 8]. Further evidence-based exploration of this process is needed. Previous studies have shown [9, 10] that self-efficacy is a leading factor in the health promotion model and is an important predictor of individuals’ health-promoting behaviors [11]. It is necessary to pay attention to the level of social support and self-efficacy of patients after esophageal cancer surgery and to further clarify the relationships among social support, self-efficacy, and health-promoting behaviors. At present, most domestic research on health-promoting behaviors has focused on patients with chronic diseases such as stroke, hypertension, and coronary heart disease [12–14]. Few relevant studies exist on the health-promoting behaviors of patients who have undergone esophageal cancer surgery [15, 16]. Moreover, health-promoting behaviors vary among individuals. Most domestic and international studies on patients’ health-promoting behaviors have determined their overall levels on the basis of scale scores. There remains a lack of consideration of individual differences among esophageal cancer patients. Therefore, this study aimed to explore the heterogeneous characteristics of and factors influencing health-promoting behaviors in patients after esophageal cancer surgery. To achieve this goal, we employed latent profile analysis (LPA), a method that is well suited for identifying latent categories on the basis of scale dimensions, as it allows for the identification of naturally occurring subgroups without predefined classifications [17–19]. Methods Study design A cross-sectional sampling survey was conducted at 14 cancer centers across China (Beijing, Guangdong, Shandong, Shanxi, Henan, Hebei, Shaanxi, Sichuan, Hunan, Hubei, Guangxi, Fujian, Xinjiang, and Shanghai Provinces) from January 2024 to April 2024. Patients with esophageal cancer who had received surgical treatment for esophageal cancer were selected as the research subjects. We further clarify that our sample includes patients from 14 geographically and socioeconomically diverse provinces in China (including the northern, southern, and western regions) and that the sample is diverse and representative. Participants Participants were recruited using convenience sampling. The inclusion criteria were as follows: (1) patients who were pathologically diagnosed with esophageal cancer; (2) patients who had received surgical treatment for esophageal cancer; (3) patients who were aged ≥ 18 years; (4) patients who were conscious and able to read, write, and communicate in daily life; and (5) patients who understood the purpose of the research and who agreed and voluntarily participated in this study. The exclusion criteria were as follows: (1) cognitive impairment and (2) other serious physical diseases, for example, comorbidities that could independently affect health behaviors or recovery, such as end-stage renal disease, severe heart failure, or active tuberculosis. According to Kendall’s criterion [20], ten observations are recommended per variable. A total of 35 variables were included (23 from demographic variables and 12 from scales). Considering a loss rate of 15 to 20%, the minimum sample size for this study was 420 participants. Data collection The directors of 14 Chinese cancer centers were contacted to form different collaborative units. After providing informed consent, research nurses at each cancer center were provided with standardized online training, and they distributed questionnaires to participants via QR codes on the Wenjuanxing online platform. Prior to completing the questionnaire, detailed explanations of the methods for filling it out and matters needing attention were provided. The questionnaire included a sociodemographic and disease-related information collection form, the revised version of the Health-Promoting Lifestyle Profile-II (HPLP-II R), the Social Support Rating Scale (SSRS), and the Chinese version of the Cancer Self-Efficacy Scale (Chinese version of the Strategies Used by People to Promote Health (C-SUPPH) questionnaire). Disease-related information was collected by researchers via the Hospital Information System (HIS). Participants completed the questionnaire independently. Participants without smartphones or digital literacy completed the survey using paper questionnaires with the assistance of research nurses to ensure they were not excluded because of technical barriers. Research nurses promptly answered any questions that arose during the completion process. For participants who were unable to complete the questionnaire (due to visual impairments or objective reasons, such as receiving intravenous infusions), research nurses assisted them in completing the questionnaire by asking questions verbally on their behalf. The Wenjuanxing online platform set all fields as “required” and allowed anonymous submission of responses, with each IP address limited to one submission. Incomplete or duplicate questionnaires were excluded. A total of 450 questionnaires were distributed, and 438 valid questionnaires were recovered, for an effective recovery rate of 97.3%. Materials A sociodemographic and disease-related information collection form A sociodemographic and disease-related information collection form was designed by the researchers, and sociodemographic and disease-related information, including age, sex, marital status, education level, place of residence, number of children, family monthly income per person, family history of cancer, preoperative and postoperative eating habits, tumor location, clinical status, cancer type, degree of differentiation, and surgical method, was collected. The revised version of the Health-Promoting Lifestyle Profile-II (HPLP-II R) This scale was developed by Walker [5], and the Chinese version revised by Cao was used [21]. The scale includes 6 dimensions—interpersonal relationships, health responsibility, spiritual growth, physical exercise, stress management, and nutrition—with a total of 40 items, scored using a 4-point Likert scale ranging from “1 ~ 4” (never–always). The total score ranges from 40 to 160; the higher the total score is, the greater the level of health-promoting behavior and the healthier an individual’s lifestyle. Cronbach’s alpha coefficient of the scale was 0.837 in this study. The Social Support Rating Scale (SSRS) The SSRS, developed by the Chinese scholar Xiao Shuiyuan [22], is widely used to measure the type and utilization of social support and was used in this study. The level of social support can be used to predict the physical and mental health status of individuals with cancer. The scale includes the subjective support dimension (Items 1, 3, 4, and 5), objective support dimension (Items 2, 6, and 7), and utilization of support dimension (Items 8, 9, and 10), with a total of 10 items. The total score ranges from 12 to 66 points, with a higher score indicating a better level of social support. In this study, Cronbach’s alpha coefficient of this scale was 0.852. The Chinese version of the Cancer Self-Efficacy Scale (Chinese version of the Strategies Used by People to Promote Health (C-SUPPH) questionnaire) This scale was developed by Lev and Oven [23], and Yuan [24] translated the English version of the SUPPH questionnaire into Chinese. It includes three dimensions, namely, positive attitude, self-stress reduction, and self-decision-making, with scores ranging from “never confident to very confident.” Positive scores ranging from 1 to 5 are assigned. The higher the score is, the stronger the patient’s self-efficacy, and this score is suitable for measuring the self-efficacy of cancer patients with a Chinese cultural background [24]. In this study, Cronbach’s alpha coefficient of this scale was 0.879. Ethical considerations This study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of Sun Yat-sen University Cancer Center. (Approval number: B2024-159-01). Statistical data analysis The latent profile model was established using Mplus 8.3 software. There are three types of fit indicators for model adaptation testing [25]. The information evaluation indicators include the Akaike information criterion (AIC), Bayesian information criterion (BIC), and adjusted Bayesian information criterion (aBIC); the classification index is entropy, with a value ranging from 0 to 1; and the likelihood ratio test index includes the Lo-Mendell-Ruben (LMR) adjusted likelihood ratio test and bootstrap likelihood ratio test (BLRT) based on bootstrapping. The smaller the AIC, BIC, and aBIC are, the better the model fit; generally, an entropy > 0.7 is needed, and an entropy ≥ 0.8 means that the classification accuracy is greater than 90%. When the LMR test and BLRT are significant, that is, at P < 0.05, the K categories are better than the model of the K-1 categories [25]. SPSS 26.0 was used to describe and analyze the data. Normally distributed measurement data are described by means and standard deviations; count data are described by frequencies and composition ratios. The distributions of the latent categories of health-promoting behaviors in patients who underwent esophageal cancer surgery were compared using the chi-square test or Fisher’s exact probability method, and the sociodemographic, social support, and self-efficacy scores were compared between behaviors in each category using one-way analysis of variance or the Kruskal–Wallis H test. An ordered multiclass logistic regression analysis was used to explore the factors influencing each category of health-promoting behaviors. P < 0.05 indicated a statistically significant difference. Results Profile analysis of potential health-promoting behaviors in patients after esophageal cancer surgery In this study, 1 to 5 latent class models were constructed in sequence on the basis of the six dimensions of the revised version of the HPLP-II R (interpersonal relationships, health responsibility, stress management, nutrition, spiritual growth, and physical exercise). See Table 1 for details. As the number of categories increased, the AIC, BIC, and aBIC continued to decrease. When 3 categories were retained, the entropy value was the greatest, and both the LMR test and the BLRT reached significance. By combining the practical significance and research significance of classification, 3 categories were ultimately selected in this study. Following the principle of parsimony [26], we sought the simplest model that adequately captured the heterogeneity in health-promoting behaviors. Although the fit indices continued to improve with 4 and 5 classes, the improvement was marginal compared with the added complexity. The LMR test was significant for the 3-class model (P < 0.001), indicating that the fit of the 3-class model was superior to that of the 2-class model. The nonsignificant LMR for the 4-class model (P > 0.05) suggested that additional classes did not significantly improve model fit [27]. Sensitivity analyses by age/center confirmed consistent profiles. The model provided optimal classification results for the health-promoting behaviors of patients after esophageal cancer surgery. Table 1Fit indicators for profile analysis of potential health-promoting behaviors in patients after esophageal cancer surgeryNumber of categoriesAICBICaBICEntropyLMR(p)BLRTClass probability1C5402.3515451.3385413.256----2C4664.2584741.8204681.5230.810734.834 (0.0003)752.093 (< 0.001)0.463/0.5373C4342.1084448.2464365.7350.858328.435 (< 0.001)336.150 (< 0.001)0.276/0.562/0.1624C4260.4064395.1194290.3940.81893.506 (0.0230)95.702 (< 0.001)0.125/0.358/0.380/0.1375C4174.0174337.3054210.3650.84698.085 (0.0284)100.389 (< 0.001)0.136/0.376/0.310/0.036/0.142AIC Akaike information criterion, BIC Bayesian information criterion, aBIC sample-size adjusted Bayesian information criterion, LMR(p) p value from the Lo–Mendell–Rubin adjusted likelihood ratio test, and BLRT bootstrap likelihood ratio test Table 1Fit indicators for profile analysis of potential health-promoting behaviors in patients after esophageal cancer surgeryNumber of categoriesAICBICaBICEntropyLMR(p)BLRTClass probability1C5402.3515451.3385413.256----2C4664.2584741.8204681.5230.810734.834 (0.0003)752.093 (< 0.001)0.463/0.5373C4342.1084448.2464365.7350.858328.435 (< 0.001)336.150 (< 0.001)0.276/0.562/0.1624C4260.4064395.1194290.3940.81893.506 (0.0230)95.702 (< 0.001)0.125/0.358/0.380/0.1375C4174.0174337.3054210.3650.84698.085 (0.0284)100.389 (< 0.001)0.136/0.376/0.310/0.036/0.142AIC Akaike information criterion, BIC Bayesian information criterion, aBIC sample-size adjusted Bayesian information criterion, LMR(p) p value from the Lo–Mendell–Rubin adjusted likelihood ratio test, and BLRT bootstrap likelihood ratio test Fit indicators for profile analysis of potential health-promoting behaviors in patients after esophageal cancer surgery AIC Akaike information criterion, BIC Bayesian information criterion, aBIC sample-size adjusted Bayesian information criterion, LMR(p) p value from the Lo–Mendell–Rubin adjusted likelihood ratio test, and BLRT bootstrap likelihood ratio test Characteristics and naming of potential categories of health-promoting behaviors in patients after esophageal cancer surgery On the basis of the latent classification results, the scores of the three categories on the six dimensions of the HPLP-II R are plotted, as shown in Fig. 1. Group C1 included 121 people, accounting for 27.6% of the total sample. The average score for each dimension of health-promoting behavior was generally low; thus, this group was named the “low-level health-promoting behavior group.” Group C2 included 246 people, accounting for 56.2% of the total sample. The average score for each dimension item regarding health-promoting behavior was at a medium level; thus, this group was named the “medium-level health-promoting behavior group.” Group C3 included 71 people, accounting for 16.2% of the total sample, and the average score on each dimension item for health-promoting behavior was high, indicating that patients who fell into this category after surgery for esophageal cancer had a high level of health-promoting behavior; thus, this group was named the “high-level health-promoting behavior group.” The details are shown in Fig. 1.Fig. 1Three profile categories of potential health-promoting behaviors in patients after esophageal cancer surgery Three profile categories of potential health-promoting behaviors in patients after esophageal cancer surgery Single-factor analysis of potential health-promoting behaviors in patients after esophageal cancer surgery Univariate analysis revealed significant differences in age, education level, place of residence, number of children, family per capita monthly income, marital status, degree of differentiation, esophageal cancer location, preoperative neoadjuvant status, and level of social support among postoperative patients with different esophageal cancer subtypes. The differences in the distribution of characteristics such as the social support level and self-efficacy level were statistically significant (P < 0.05). See Table 2 for details. Table 2Single-factor analysis of the health-promoting behavior profiles of patients after esophageal cancer surgeryProjectCategoryClassification of health-promoting behaviors in patients after esophageal cancer surgeryTotal number of patients \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\chi$$\end{document}χ2 P Negative Group (n = 121)Medium Group (n = 246)Positive Group (n = 71)Gender1.5740.455 Male10319961363 Female18471075Age group134.587 < 0.001 18 ~ < 4510164369 45 ~ < 60426816126 > 606916212243Education level115.292 < 0.001 Elementary school or lower62994165 Junior high school16382680 Senior high school/technical secondary school34846124 College or higher9253569Marital status6.6700.036 Single/Divorced/Widowed1111931 Married11023562407Place of residence34.917 < 0.001 City5211359224 Countryside6913312214Number of children78.461 < 0.001Nulliparous2116191 child3465211202 children53106301893 or more children32744110Per capita monthly household income (yuan)33.523 < 0.001 < 300047774128 3000 ~ 5000468626158 > 5000288341152Family relationship status3.0980.541 Harmonious10121665382 General1929654 Terrible1102Smoking before surgery2.1220.713 Yes519832181 No409626162 Former smoker and quit smoking30521395Alcohol consumption before surgery7.9080.094 Yes479928174 No4210135178 Former drinker and quit drinking3246886Preoperative eating habits3.5650.467 Light taste15391670 Average taste7213736245 Strong taste347019123Diet structure before surgery0.1250.939 Carnivorism6513540240 Vegetarianism5611131198Smoking after surgery2.3700.667 Yes48416 No10521263380 Former smoker and quit smoking1226442Alcohol consumption after surgery2.9830.560 Yes2619 No10421166381 Former drinker and quit drinking1529448Postoperative eating habits8.1310.086 Light taste8414953286 Average taste328818138 Strong taste59014Diet structure after surgery0.2220.894 Carnivorism347521130 Vegetarianism8717150308Differentiation degree of esophageal cancer15.3330.017 Undifferentiated715325 Poorly differentiated485614118 Moderately differentiated5614944249 Highly differentiated10261046Esophageal cancer classification4.9120.555 Squamous cell carcinoma10320258363 Adenocarcinoma15391064 Small cell carcinoma0202 Other3339Esophageal cancer location15.7710.015 Neck segment1236 Upper chest12131035 Mid chest6013341234 Lower chest489817163Surgical method4.8650.301 Open chest1835659 Thoracolaparoscopy9317556324 Robot-assisted surgery1036955Preoperative neoadjuvant therapy status8.3420.015 No5314534232 Yes6810137206Family history of cancer2.9090.233 No9620663365 Yes2540873Body mass index (BMI)22.28 ± 3.2422.51 ± 3.1822.77 ± 2.75-0.5410.582Social support35.49 ± 5.7540.24 ± 5.5944.77 ± 5.38-64.246 < 0.001Self-efficacy67.12 ± 20.6296.85 ± 18.59111.34 ± 18.66-145.050 < 0.001 Table 2Single-factor analysis of the health-promoting behavior profiles of patients after esophageal cancer surgeryProjectCategoryClassification of health-promoting behaviors in patients after esophageal cancer surgeryTotal number of patients \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\chi$$\end{document}χ2 P Negative Group (n = 121)Medium Group (n = 246)Positive Group (n = 71)Gender1.5740.455 Male10319961363 Female18471075Age group134.587 < 0.001 18 ~ < 4510164369 45 ~ < 60426816126 > 606916212243Education level115.292 < 0.001 Elementary school or lower62994165 Junior high school16382680 Senior high school/technical secondary school34846124 College or higher9253569Marital status6.6700.036 Single/Divorced/Widowed1111931 Married11023562407Place of residence34.917 < 0.001 City5211359224 Countryside6913312214Number of children78.461 < 0.001Nulliparous2116191 child3465211202 children53106301893 or more children32744110Per capita monthly household income (yuan)33.523 < 0.001 < 300047774128 3000 ~ 5000468626158 > 5000288341152Family relationship status3.0980.541 Harmonious10121665382 General1929654 Terrible1102Smoking before surgery2.1220.713 Yes519832181 No409626162 Former smoker and quit smoking30521395Alcohol consumption before surgery7.9080.094 Yes479928174 No4210135178 Former drinker and quit drinking3246886Preoperative eating habits3.5650.467 Light taste15391670 Average taste7213736245 Strong taste347019123Diet structure before surgery0.1250.939 Carnivorism6513540240 Vegetarianism5611131198Smoking after surgery2.3700.667 Yes48416 No10521263380 Former smoker and quit smoking1226442Alcohol consumption after surgery2.9830.560 Yes2619 No10421166381 Former drinker and quit drinking1529448Postoperative eating habits8.1310.086 Light taste8414953286 Average taste328818138 Strong taste59014Diet structure after surgery0.2220.894 Carnivorism347521130 Vegetarianism8717150308Differentiation degree of esophageal cancer15.3330.017 Undifferentiated715325 Poorly differentiated485614118 Moderately differentiated5614944249 Highly differentiated10261046Esophageal cancer classification4.9120.555 Squamous cell carcinoma10320258363 Adenocarcinoma15391064 Small cell carcinoma0202 Other3339Esophageal cancer location15.7710.015 Neck segment1236 Upper chest12131035 Mid chest6013341234 Lower chest489817163Surgical method4.8650.301 Open chest1835659 Thoracolaparoscopy9317556324 Robot-assisted surgery1036955Preoperative neoadjuvant therapy status8.3420.015 No5314534232 Yes6810137206Family history of cancer2.9090.233 No9620663365 Yes2540873Body mass index (BMI)22.28 ± 3.2422.51 ± 3.1822.77 ± 2.75-0.5410.582Social support35.49 ± 5.7540.24 ± 5.5944.77 ± 5.38-64.246 < 0.001Self-efficacy67.12 ± 20.6296.85 ± 18.59111.34 ± 18.66-145.050 < 0.001 Single-factor analysis of the health-promoting behavior profiles of patients after esophageal cancer surgery Multiple logistic regression analysis of potential health-promoting behaviors in patients after esophageal cancer surgery Taking the health-promoting behavior categories of patients after esophageal cancer surgery as the dependent variable, the indicators with statistically significant differences in the single-factor analysis were included as independent variables in the multiple logistic regression analysis. Assignments were made as follows: age, 18 to 44 years = 1, > 45 to 60 years = 2, > 60 years = 3; education level, primary school and below = 1, junior high school = 2, high school/technical secondary school = 3, undergraduate/junior college = 4, and master’s degree and above = 5; place of residence, urban = 1, rural = 2; number of children, no children = 1, 1 child = 2, 2 children = 3, and 3 or more children = 4; average monthly household income per capita, < 3,000 yuan = 1, > 3000–5000 yuan = 2, > 5000–10,000 yuan = 3, and > 10,000 yuan = 4; marital status, single/divorced/widowed = 1 and married = 2; degree of differentiation, undifferentiated = 1, poorly differentiated = 2, moderately differentiated = 3, and well differentiated = 4; tumor location, cervical segment = 1, upper thoracic segment = 2, middle thoracic segment = 3, and lower thoracic segment = 4; and preoperative neoadjuvant status, no = 1 and yes = 2; and social support and self-efficacy were substituted with actual measured values. The results revealed that esophageal cancer patients who were older than 60 years (OR = 0.073; P < 0.001) and had a primary school education or below (OR = 0.035; P = 0.003) were more likely to be classified into Group C1, which was the low-level health-promoting behavior group. Patients who were 18–44 years old (OR = 6.344, P = 0.010), lived in a city (OR = 0.179, P = 0.008), had received neoadjuvant treatment before surgery (OR = 7.02, P = 0.003), had greater social support (OR = 0.853, P = 0.002), and had greater self-efficacy (OR = 0.941, P < 0.001) were more likely to be classified into Group C3, which was the high-level health-promoting behavior group. See Table 3 for details. Table 3Multivariate logistic analysis of the health-promoting behavior profiles of patients after esophageal cancer surgeryVariableβSEWald χ2POR95% CIProfile 2: medium-level health-promoting behavior group (vs. Profile 1: low-level health-promoting behavior group) Age−2.6180.75012.185 < 0.0010.0730.017–0.317 > 60 years, ref: 18 to 44 yearsProfile 3: high-level health-promoting behavior group (vs. Profile 1: low-level health-promoting behavior group) Age1.8460.7126.7210.0106.3341.569–25.57 18 ~ 44 years, ref: > 60 years Education level−3.3631.1149.1070.0030.0350.004–0.308 Primary school and below, ref: master’s degree and above Preoperative neoadjuvant therapy status1.9490.6568.8210.0037.021.94–25.402 Yes, ref: No Social support0.2820.05823.893 < 0.0011.3261.184–1.485 Self-efficacy0.1390.01857.245 < 0.0011.1491.108–1.191Ref Reference, SE standard error, OR odds ratio, 95% CI 95% confidence interval Table 3Multivariate logistic analysis of the health-promoting behavior profiles of patients after esophageal cancer surgeryVariableβSEWald χ2POR95% CIProfile 2: medium-level health-promoting behavior group (vs. Profile 1: low-level health-promoting behavior group) Age−2.6180.75012.185 < 0.0010.0730.017–0.317 > 60 years, ref: 18 to 44 yearsProfile 3: high-level health-promoting behavior group (vs. Profile 1: low-level health-promoting behavior group) Age1.8460.7126.7210.0106.3341.569–25.57 18 ~ 44 years, ref: > 60 years Education level−3.3631.1149.1070.0030.0350.004–0.308 Primary school and below, ref: master’s degree and above Preoperative neoadjuvant therapy status1.9490.6568.8210.0037.021.94–25.402 Yes, ref: No Social support0.2820.05823.893 < 0.0011.3261.184–1.485 Self-efficacy0.1390.01857.245 < 0.0011.1491.108–1.191Ref Reference, SE standard error, OR odds ratio, 95% CI 95% confidence interval Multivariate logistic analysis of the health-promoting behavior profiles of patients after esophageal cancer surgery Ref Reference, SE standard error, OR odds ratio, 95% CI 95% confidence interval Discussion The overall health-promoting behaviors of patients after esophageal cancer surgery are moderate to low This study revealed that the overall score for health-promoting behavior in patients after esophageal cancer surgery was 104.31 ± 21.72 points, which was moderately low and slightly lower than that reported by Guo et al. [16]. The main reasons for the low level of health-promoting behaviors among patients after esophageal cancer surgery are as follows: in terms of demographics, esophageal cancer patients may be older, and most of them have low educational levels and have less disease-related information. In terms of physical symptoms, esophageal cancer surgery usually involves greater physical trauma and a longer recovery period after surgery. Patients may have difficulty actively participating in health-promoting behaviors because of postoperative pain, physical weakness, digestive system adjustments, and other issues. At the same time, patients’ eating habits and nutrient absorption capacity are usually limited after esophageal cancer surgery, and they may need to be fed through a nasogastric tube or adopt other special dietary methods after surgery, which may further lead to malnutrition and affect their overall health and recovery. It may be more difficult for these patients to take active measures [28, 29]. In terms of health behavior and psychological status, after esophageal cancer surgery, patients are generally under great pressure, usually because of concerns about disease treatment and prognosis and also because family pressure and financial burdens may develop. Medical staff should pay close attention to the physical and psychological states of patients after esophageal cancer surgery, focus on nutritional intake and stress management, and promptly identify and address negative emotions, thereby improving the health-promoting behaviors of postoperative patients. Characteristics of postoperative esophageal cancer patients in different health-promoting behavior categories Through LPA, this study identified three health-promoting behavior categories after esophageal cancer surgery, suggesting that the health-promoting behaviors of patients after esophageal cancer surgery are diverse. The “low-level health-promoting behavior group” accounted for 27.6% of the sample and had the lowest scores in the “spiritual growth” and “stress management” dimensions. This highlights deficiencies in their ability to find meaning in recovery (e.g., adapting to role changes postsurgery) and cope with emotional distress (e.g., anxiety about recurrence). Patients in this category were characterized mainly by advanced age (> 60 years) and a low education level (less than primary school). Owing to the decline in their ability to perform personal functions, this group of patients could not bear the physical difficulty of surgery. Moreover, it was difficult for this group of patients to actively acquire and understand health promotion-related education knowledge and methods, resulting in a low level of health-promoting behavior. These gaps underscore the need for targeted interventions, such as mindfulness-based stress reduction programs to enhance coping strategies and peer support groups to foster spiritual growth through shared experiences. In contrast, the “high-level health-promoting behavior group” accounted for 16.2% of the sample and scored highest in the “health responsibility” and “nutrition” dimensions. This likely reflects their better adherence to postoperative dietary guidance (e.g., following dysphagia-friendly meal plans) and proactive engagement in health management (e.g., regular follow-up visits and symptom self-monitoring), which is consistent with their greater social support (e.g., family assistance in meal preparation) and self-efficacy in overcoming nutrition-related barriers (e.g., managing early satiety). Patients in this group were mainly young and middle-aged (18–44 years old), lived in cities, received neoadjuvant treatment before surgery, had high social support levels, and had high self-efficacy levels. These patients had good physical functioning and a high tolerance for adverse reactions to surgery and treatment. Moreover, the objective living environment was conducive to access to public resources related to health promotion. Additionally, preoperative neoadjuvant therapy can shrink tumor lesions before surgery, somewhat reduce the incidence of postoperative complications, and is conducive to postoperative recovery. High levels of social support and self-efficacy are also favorable for improving individual health-promoting behaviors. Therefore, such patients showed higher levels of health-promoting behaviors. Medical staff should identify different types of postoperative esophageal cancer patients and develop targeted intervention programs based on particular demographic and disease characteristics, levels of social support, and levels of self-efficacy to promote the success of different types of postoperative esophageal cancer patients. Factors influencing the health-promoting behavior of patients after esophageal cancer surgery Age This study revealed that age is a factor influencing potential health-promoting behaviors in patients after esophageal cancer surgery. Patients aged between 18 and 44 years who underwent esophageal cancer surgery were more likely to be included in the active health-promoting behavior group (OR = 6.344, P = 0.010) and have higher levels of health-promoting behaviors. Postoperative esophageal cancer patients aged > 60 years were more likely to be included in the low-level health-promoting behavior group (OR = 0.073, P < 0.001), and their health-promoting behavior levels were low. Consistent with the research results of Guo et al. [16], our study results differ from the conclusion of Zhou [30] that older patients have higher levels of health-promoting behavior. The main reason is that as patients age, their functional abilities deteriorate, and their tolerance to surgical treatment decreases. Moreover, their ability to receive information and learn health education knowledge is declining. Therefore, postoperative esophageal cancer patients older than 60 years had lower levels of health-promoting behaviors. Medical staff should provide targeted health education to postoperative esophageal cancer patients of different ages. In particular, more understandable language and education methods should be used for older patients, and step-by-step health promotion education and intervention strategies should be developed. Education level This study revealed that education level influences the potential health-promoting behaviors of patients after esophageal cancer surgery. After esophageal cancer surgery, patients whose education level was less than elementary school had lower levels of health-promoting behaviors and were more likely to be included in the low-level health-promoting behavior group (OR = 0.035; P = 0.003). These findings are consistent with the results of previous studies [16, 30]. Possible reasons include (1) a lack of health knowledge. Patients with low education levels lack relevant health knowledge, which makes it difficult for them to understand health information and medical guidance, possibly leading to limited health awareness and an insufficient understanding of the importance of healthy behaviors. (2) Barriers to information access are another possible reason. Individuals with less education may have difficulty obtaining, interpreting, and utilizing health information, finding reliable health information, or evaluating the accuracy of health information. Therefore, when providing health education, medical staff need to teach patients according to their individual needs and adopt appropriate education methods. Patients with higher education levels can be educated on the principles of disease pathogenesis, influencing factors, preventive health care, etc. In comparison, patients with lower levels of education can be educated by providing disease knowledge and information on precautions in plain language, increasing the frequency of health education, and seeking help from patients’ family members when necessary to meet their needs. Place of residence This study revealed that place of residence is a factor influencing the latent categories of health-promoting behaviors among patients after esophageal cancer surgery. Esophageal cancer surgery patients who lived in cities had greater levels of health-promoting behaviors and were more likely to be included in the high-level health-promoting behavior group (OR = 0.179, P = 0.008). This may be because, compared with rural areas, urban areas usually have more advanced and abundant medical resources, which not only provide higher-quality postoperative treatment but also perform regular physical examinations and follow-up, which help patients better understand and engage in health-promoting behaviors. Additionally, urban residents generally have access to more health education opportunities and information, which may increase their awareness of the importance of a healthy lifestyle and help them better understand how to promote health through diet, physical activities, etc. At the same time, urban areas have more convenient transportation and greater utilization of medical and rehabilitation facilities. Therefore, after esophageal cancer surgery, patients who lived in cities had greater levels of health-promoting behaviors. This should prompt medical staff to pay more attention to patients from rural areas, effectively utilize medical resources, establish an integrated health information management system between primary hospitals and higher-level hospitals, reduce health information barriers, and further improve the levels of health-promoting behaviors among such patients. Neoadjuvant treatment status before surgery This study revealed that whether patients received neoadjuvant treatment before surgery was a factor influencing the potential health-promoting behaviors of patients after esophageal cancer surgery. Esophageal cancer surgery patients who received preoperative neoadjuvant treatment before surgery had greater levels of health-promoting behaviors and were more likely to be included in the high-level health-promoting behavior group (OR = 7.02; P = 0.003). On the one hand, neoadjuvant therapy may reduce the tumor volume, which enhances the surgical effect and causes mild adverse reactions. This physiological improvement may result in more positive psychological reactions and greater motivation for patients to maintain their health after surgery. On the other hand, the preoperative neoadjuvant treatment process allows patients more time to psychologically prepare for the disease and surgery, increasing their motivation to follow the lifestyle recommended by medical staff after surgery. Therefore, after esophageal cancer surgery, patients who received preoperative neoadjuvant therapy before surgery exhibited higher levels of health-promoting behaviors, which may be due to their knowledge of the disease, psychological preparation, physiological recovery, frequency of medical interactions, and expectations for recovery. These findings suggest that medical staff can continue to actively explore the positive application of neoadjuvant therapy in patients with esophageal cancer. Moreover, more health promotion-related help and support should be given to postoperative patients who are not receiving neoadjuvant therapy to improve surgical performance and patient quality of life. Social support Our finding that social support predicts increased health-promoting behaviors aligns with the findings of a Netherlands study showing that family support enhances adherence to postoperative care among esophageal cancer survivors [31]. After esophageal cancer surgery, patients who had high levels of social support had greater levels of health-promoting behaviors and were more likely to be included in the high-level health-promoting behavior group (OR = 0.853, P = 0.002). This may be because greater social support provides patients with interpersonal emotional support, helps reduce their risk of anxiety and depression, enhances their psychological resilience, and enables them to engage in health-promoting behaviors more actively in the face of disease and recovery. Second, regarding the level of information support, the information support that postoperative patients receive from different social networks can increase their awareness of the importance of health behaviors and guide them in engaging in health-promoting behaviors [32]. Additionally, patients with strong social support are more likely to follow doctors’ recommendations under the supervision and encouragement of family or friends, allowing them to face the disease positively and improving their compliance with health-promoting behaviors and awareness of health responsibilities. Therefore, medical staff should focus on the level of social support for patients after esophageal cancer surgery, emphasize the importance of social support, guide patients to actively seek effective social support by establishing patient relationships, increase patients’ utilization of social support, and improve patients’ health. Behavioral patterns and alleviating negative emotions can be promoted, thereby improving patients’ quality of life. Self-efficacy This study revealed that self-efficacy influences the latent categories of health-promoting behaviors among patients after esophageal cancer surgery. After esophageal cancer surgery, patients who had a high level of self-efficacy had a greater level of health-promoting behaviors and were more likely to be included in the high-level health-promoting behavior group (OR = 0.941; P < 0.001). These findings are consistent with previous research results [16, 33–36]. According to self-efficacy theory [37], people with greater self-efficacy for a certain behavior demonstrate greater motivation, effort, and persistence in behavioral activities and thus have better health outcomes. Patients with higher self-efficacy are more confident, enabling them to successfully adopt and maintain healthy behaviors. Such patients are better able to maintain a positive attitude when they encounter obstacles and challenges, such as physical discomfort, mood swings, or other difficulties. Additionally, patients with higher levels of self-efficacy are more likely to proactively seek recovery-related information and resources, including medical consultations, rehabilitation procedures, and social support, and to adopt more effective coping strategies to address postoperative life and health challenges [38]. Therefore, after esophageal cancer surgery, patients with high self-efficacy have greater levels of health-promoting behaviors because they are better able to effectively cope with challenges during recovery and proactively manage their health conditions. It is suggested that medical staff fully evaluate the health-promoting behavior of patients after esophageal cancer surgery and guide them to view the disease correctly, seek rehabilitation knowledge, and adopt positive attitudes and behaviors to solve health problems, thereby improving their self-efficacy and health-promoting behavior after surgery. Limitations The present study has several limitations that should be acknowledged when the findings are interpreted: First, generalizability across populations was limited. Although our sample included patients from 14 provinces from northern, southern, and western China, reflecting diverse geographic and socioeconomic backgrounds, generalizability to populations beyond China (e.g., Western countries) may be restricted. Second, the use of convenience sampling may have resulted in overrepresented patients with higher levels of engagement, potentially introducing selection bias that should be considered when the findings are interpreted. Additionally, the cross-sectional nature of this study precludes establishing causal relationships between variables, as it captures only a snapshot of data at a single time point. Finally, key data on postoperative complications (e.g., severity of dysphagia) and healthcare access barriers were not collected, despite their potential to influence health behaviors. Conclusion This study revealed that the overall health-promoting behavior level of patients after esophageal cancer surgery is moderate and that there are three levels of behavior: low-, medium-, and high-level health-promoting behaviors. Age distribution, education level, place of residence, preoperative neoadjuvant treatment status, social support level, and self-efficacy level varied among different types of patients after esophageal cancer surgery. Clinical medical staff should consider the abovementioned influencing factors and explore intervention plans for different types of health-promoting behaviors for patients after esophageal cancer surgery. This study provides a reference for an in-depth understanding of health-promoting behaviors and the development of targeted intervention strategies.