Latent profiles of coping modes in rural older patients with chronic co-morbidities: a multi-center study Authors: Hualong Ma, Zihan Li, Yishang Zhuo, Xianliang Liu, Xiaoge Liu, Dalong Chen, Jingyi Li, Jiayu Liu, Lina Guan, Yao Pei, Niu Yang, Xinyi Fang, Tanjian Li, Kaiqing Liang, Xiaoyu Wan, Yuexin Huang, Jiahui Liu, Zaoxiu Yan, Qiyuan Lyu, Qiaohong Yang Published: 2025 ABSTRACT AbstractBackgroundUnderstanding coping modes in rural elderly with chronic co-morbidities is critical due to their impact on disease management and psychological adjustment in resource-limited settings. The aim of this study is to identify latent profiles of coping modes and their influencing factors in rural older patients with chronic co-morbidities.MethodsA cross-sectional, exploratory design was adopted. Using multistage random sampling, 2521 rural elderly patients with chronic co-morbidities were recruited from ten rural areas in Guangdong and Shanxi Provinces, China between January 2024 and October 2024. Data collection tools included a demographic questionnaire, the Medicine Coping Modes Questionnaire Chinese Version, the Three-Dimensional Meaning in Life Scale, the Berkman-Syme Social Network Index, and the Family APGAR Index.ResultsA total of 2521 participants were recruited, with an average age of 71.70 years (SD = 9.47). Three distinct latent profiles of coping modes were identified: the Helpless Group (n = 642, 25.47%), Indomitable Group (n = 390, 15.47%), and Conflicting Group (n = 1489, 59.06%). Compared with the Helpless Group, individuals in the Indomitable and Conflicting Groups were younger (OR = 0.938, 0.961, 95% CI 0.924 ~ 0.952, 0.950 ~ 0.951, respectively), more educated (OR = 3.605, 2.476, 95% CI 2.399 ~ 5.419, 1.822 ~ 3.367), had shorter disease durations (OR = 0.701, 0.793, 95% CI 0.615 ~ 0.799, 0.719 ~ 0.875), reported a stronger sense of meaning in life (OR = 1.044, 1.034, 95% CI 1.037 ~ 1.052, 1.027 ~ 1.040), were more likely to cohabit with a spouse (OR = 1.870, 1.531 95% CI 1.139 ~ 3.071), offspring (OR = 2.172, 1.756, 95% CI 1.306 ~ 3.611, 1.223 ~ 2.522 ) or both (OR = 2.043, 1.693, 95% CI 1.309 ~ 3.187, 1.237 ~ 2.316), and experienced better family functioning (OR = 1.313, 1.180, 95% CI 1.100 ~ 1.567, 1.038 ~ 1.343). Moreover, the Indomitable Group displayed greater social integration (OR = 1.402, 95% CI 1.198 ~ 1.640), while the Conflicting Group had a higher incidence of co-morbidities (OR = 1.357, 95% CI 1.218 ~ 1.513). All p < .05.ConclusionsThis study highlights distinct coping profiles among rural elderly with chronic co-morbidities and underscores the critical role of psychosocial factors in shaping coping behaviors. Targeted interventions addressing psychosocial determinants could optimize disease management and improve outcomes in this vulnerable population.Clinical trial numberNot applicable.Keywords: Rural, Latent profile, Coping mode, Older, Co-morbidities. Introduction Geriatric co-morbidity refers to the co-occurrence of two or more chronic health conditions in older adults, including geriatric syndromes, psychosomatic issues, and organ-specific diseases [1]. These conditions often require prolonged and complex interventions due to their persistent nature. A study of 38,807 rural residents in China reported that the prevalence of chronic co-morbidities in rural areas has reached 28.1%, rising with age [2]. This trend highlights a significant public health challenge, compounded by China’s aging population. Beyond physical health, co-morbidities profoundly impact mental well-being and social functioning [3]. For older patients in rural areas, managing these conditions is further complicated by limited healthcare resources compared to urban settings [4]. These challenges not only hinder diagnosis and treatment but also reduce quality of life. As a result, the coping modes adopted by these patients are vital for both disease management and psychological adjustment [5]. Coping modes refer to the psychological and behavioral strategies individuals use to manage illness. Coping modes in chronic illness are shaped by the Stress, Appraisal, and Coping Framework, which emphasizes primary appraisal (assessing the stressor’s significance) and secondary appraisal (evaluating coping resources) [6]. Within this framework, coping modes are broadly conceptualized as problem-focused (addressing the stressor directly) and emotion-focused (managing emotional distress). Drawing on Feifel et al.’s empirical classification, we aim to examine three distinct modes among “medically ill patients”: Confrontation (proactive engagement, aligning with problem-focused coping), Avoidance (passive or evasive tactics like denial, a form of emotion-focused coping), and Resignation (a passive, maladaptive yielding to illness, distinct from the traditional dichotomy) [7]. Studies indicate that adaptive coping can significantly improve quality of life and support disease management [8]. The complexity of co-morbid chronic conditions in rural elderly populations—characterized by intricate disease-disease, disease-drug, and drug-drug interactions—suggests that their coping modes may differ significantly from those with a single chronic condition. Furthermore, variations in coping strategies may stem from individual characteristics, family support, and social environments [9]. Therefore, it is crucial to explore coping modes in this population to better understand their unique needs. Recent interdisciplinary research at the intersection of psychology and medicine has explored the factors shaping coping modes in patients with chronic illnesses. Key psychosocial predictors include Sense of Meaning in Life, Social Integration, and Family Functioning. Sense of Meaning in Life, a concept introduced by Viktor Frankl, refers to the perception of purpose and value in life’s events [10]. Studies show that a strong sense of meaning in life significantly supports adaptive responses to chronic diseases [11]. Social Integration reflects the quality of an individual’s interactions within their social support network. Yet, recent social changes in rural China, driven by large-scale urban migration, have resulted in a growing population of ‘left-behind elderly.’ [12] This phenomenon has led to the erosion of traditional clan-based support systems, thereby altering the dynamics of social integration. The diminished presence of younger family members often reduces direct familial support [13], potentially exacerbating feelings of isolation and impacting coping strategies among rural older adults. Family Functioning plays a crucial role in the mental health and coping behaviors of elderly patients, and urbanization has significantly altered rural family structures due to the migration of younger generations for work [14]. Despite their significance, the specific associations between these variables and coping modes in rural elderly populations remain largely unexplored. We selected them as key influencing factors, as sense of meaning in life shapes primary appraisal by enabling rural older adults to interpret chronic illness as a challenge rather than an insurmountable threat. Meanwhile, Social integration and family functioning serve as key social resources in secondary appraisal. Social integration reflects the broader community support available to individuals, while family functioning captures the immediate relational dynamics that bolster coping capacity. These factors are particularly critical for rural older adults, who often contend with limited healthcare access and social isolation [6]. Understanding the diversity of coping modes among rural elderly patients with chronic co-morbidities is crucial for designing targeted interventions. However, most studies on coping modes in chronic diseases have relied on a “variable-centered” approach, which primarily uses descriptive statistics and assumes uniformity in coping responses across all dimensions [15]. This perspective often overlooks individual differences and unique coping behaviors. Additionally, prior research is frequently limited by small sample sizes, restricting its ability to provide comprehensive insights into the coping strategies of older adults with chronic co-morbidities. Moreover, no studies have examined how Sense of Meaning in Life, Family Functioning, and Social Integration influence coping modes in this population. To fill these gaps, we propose using an “individual-centered” analytical approach through latent profile analysis (LPA). This approach allows for a more detailed exploration of coping mode heterogeneity and its influencing factors, providing richer insights compared to traditional methods. Methods Study design Our study was cross-sectional and exploratory. We follow the STROBE statement to report our study. Study setting and sampling A multistage random sampling strategy was used from January 2024 to October 2024 to recruit participants from ten rural areas in China, including the following steps: (1) Province Selection: Two provinces with different socioeconomic and geographic features, Guangdong (southern China) and Shanxi (northern China), were purposely chosen to represent rural older patients with chronic co-morbidities; (2) County Selection: Five rural counties in Guangdong and five in Shanxi were randomly selected using computer-generated numbers; (3) Township Selection: From each selected county, a list of townships was compiled, and one township per county was randomly chosen using the same method; (4) Village Selection: With the same method in Township Selection; (5) Participant Screening: In the selected villages, eligible participants were identified based on inclusion and exclusion criteria. To ensure sufficient statistical power, it is standard practice to have a sample size at least 10 times the number of variable [16]. With 27 variables and a 25% expected attrition rate, the minimum required sample size was calculated to be 338 cases ((27 × 10)*(1 + 0.25) ≈ 338), which was adequate for reliable statistical analysis. Prior to the main study, we conducted a pilot test with 21 representative older adults at one of the sampling sites to evaluate the survey’s feasibility and clarity. As a reward, we provide basic supplies (eggs, bread, toothbrush, towel, shower gel, shampoo or laundry detergent) to each elderly person who completes the survey. Recognizing potential difficulties with digital surveys among elderly participants, we optimized the electronic format by increasing font sizes and implementing a touch-friendly interface to accommodate age-related declines in vision and manual dexterity. Each survey group was supported by a designated researcher mentor who provided onsite assistance to ensure accurate and complete responses. Our electronic survey platform featured real-time monitoring that enabled the immediate detection of issues, such as incomplete response rates exceeding 10% or uniformly identical responses, with follow-up outreach conducted to encourage survey completion and validate key responses. Inclusion and exclusion criteria Participants were eligible if they met the following criteria: (1) Registered as rural residents either through household or permanent residence status; (2) 60 years or older; (3) Meeting the WHO definition of multiple co-morbidities, characterized by two or more chronic diseases. Examples include cardiovascular conditions (e.g., hypertension, coronary heart disease), respiratory disorders (e.g., COPD, asthma), metabolic diseases (e.g., diabetes, obesity), neurological conditions (e.g., stroke sequelae, Parkinson’s), or other chronic illnesses requiring ongoing management; (4) Able to provide informed consent and willingly participate during the study period; (5) Able to describe coping modes directly or with input from family/caregivers. Participants were excluded if they met any of the following conditions: (1) Advanced Alzheimer’s or similar conditions preventing completion of study questionnaires; (2) Acute Illnesses: Conditions in the acute phase, such as infectious shock or acute myocardial infarction; (3) Diagnoses like schizophrenia or bipolar disorder that hinder participation; (4) Enrollment in other studies, such as drug trials, within the last six months. Instruments The instruments used in this study were in electronic form to mirror the approach used in the China Health and Retirement Longitudinal Study (CHARLS) when conducting multi-center study. Each researcher was equipped with a phone to record responses. All participants were required to complete an electronic questionnaire; however, 92.66% received assistance from family members or trained interviewers due to physical limitations or poor literacy. Rigorous interviewer training and built-in logic checks on the electronic platform ensured high data quality and minimized potential bias. Additionally, key responses provided via proxy were verified through a follow-up process. This study utilized several instruments to collect data: (1) A demographic questionnaire capturing variables like age and ethnicity; (2) The Medicine Coping Modes Questionnaire, originally developed by Feifel et al., [7] was adapted into Chinese by Shen et al., [17] and assessing coping through three dimensions: Confrontation, Avoidance, and Resignation, with higher scores indicating a greater tendency to adopt the corresponding coping modes. Confirmatory factor analysis (CFA) was conducted in our study, with CFI = 0.912, TLI = 0.903, and RMSEA = 0.063 (90% CI: 0.052–0.074), supporting its suitability for rural elderly cohorts in China; (3)The Three-Dimensional Meaning in Life Scale, first created by Martela et al., [18] was translated into Chinese by Miao et al., [19] measuring life purpose across Goal, Importance, and Consistency, with higher scores indicate a stronger sense of meaning in life; (4) The Berkman-Syme Social Network Index, developed by Berkman et al., [20] evaluates social integration, with a total score of 0 ~ 12. Social integration is categorized into four levels: scores of 0 ~ 5 indicate level I, 6 indicates level II, 7 ~ 8 indicate level III, and 9 ~ 12 indicate level IV. Higher scores reflect greater social integration; and (5) the Family APGAR Index, crafted by Smilkstein et al., [21] was adapted into Chinese by Lv et al., [22] to assess family functioning across Adaptability, Cooperation, Maturity, Emotion, and Intimacy. Total scores are interpreted as follows: 0 ~ 3 indicates severe family dysfunction, 4 ~ 6 indicates moderate family dysfunction, and 7 ~ 10 indicates good family functioning. Instruments two through five demonstrated robust reliability and validity. Ethical considerations This study strictly adhered to the principles of medical ethics and the guidelines outlined in the Declaration of Helsinki. All participants provided written informed consent voluntarily after being fully informed. Inclusion and exclusion criteria were applied equitably, ensuring that no individual was excluded based on sex, race, religious beliefs, socioeconomic status, or cultural background et al. The present study was approved by the IRB of The First Affiliated Hospital of Jinan University (No. JNUKY-2024-0045). Data analysis Data analysis was performed using Python (Version 3.12) and Mplus (Version 8.3). Data with less than 20% missingness were handled with multiple imputation. Continuous variables with a normal distribution were shown as mean ± standard deviation, while others were presented as medians (interquartile range). Categorical variables were reported as frequencies (N) and percentages (%). LPA was used to classify participants based on their coping modes. Model selection was guided by Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), adjusted BIC (aBIC), p-values from the Lo-Mendell-Rubin test (p-LMR), and Bootstrap Likelihood Ratio Test (p-BLRT). Lower AIC [23], BIC [24], and aBIC values indicated better fit [25]. A p-LMR [26] or p-BLRT [27] value < 0.05 showed the model with k classes was better than one with k-1 classes. Entropy > 0.80 signified high classification accuracy [28]. Models were excluded if any group represented ≤ 8% of the sample to ensure representativeness. Group differences were assessed using one-way ANOVA for continuous variables and chi-square or Fisher’s exact tests for categorical variables. Significant variables were included in multinomial logistic regression, with latent profile groups as the dependent variable. Statistical significance was set at p <.05 [29]. A sensitivity analysis was performed to compare the latent profiles with or without patients with more than 10% missing data (34 participants, accounting for 1.33%). The results demonstrated the robustness of the latent profiles and logistic regression. Therefore, in this study, patients with more than 10% missing data were excluded to ensure that the data were as representative of real-world conditions as possible. Results General information The 2521 participants had a mean age of 71.7 years (SD = 9.47), and 1122 (44.5%) were female. Among them, 337 (13.4%) had two chronic diseases, 742 (29.4%) had three, 974 (38.6%) had four, and 468 (18.6%) had five or more chronic conditions. Detailed statistics are in Table 1. Table 1Detailed descriptive informationVariablesOverall(N = 2521)Latent profiles (M ± SD/N(%))F p Helpless Group(G0, N = 642)Indomitable Group(G1, N = 390)Conflicting Group(G2, N = 1489)Age71·70 ± 9·4774·80 ± 8·4368·94 ± 9·6071·09 ± 9·5256·334< ·001Sex assigned at birth1·975·372 Female1122 (44·51)300 (46·73)175 (44·87)647 (43·45) Male1399 (55·49)342 (53·27)215 (55·13)842 (56·55)Ethnicity1·928·381 The Han2347 (93·1)590 (91·9)365 (93·59)1392 (93·49) Other174 (6·9)52 (8·1)25 (6·41)97 (6·51)Marital status4·156·385 Married1677 (66·52)436 (67·91)257 (65·9)984 (66·08) Widowed655 (25·98)155 (24·14)111 (28·46)389 (26·12) Single/Others189 (7·5)51 (7·94)22 (5·64)116 (7·79)Educational level132·794< ·001 Primary school or below705 (27·97)220 (34·27)80 (20·51)405 (27·2) Junior high school1037 (41·13)332 (51·71)148 (37·95)557 (37·41) Senior high school or technical secondary school598 (23·72)82 (12·77)115 (29·49)401 (26·93) Bachelor’s degree or above181 (7·18)8 (1·25)47 (12·05)126 (8·46)Occupation (including previous occupations)11·671·166 Farmer984 (39·03)281 (43·77)146 (37·44)557 (37·41) Corporate executive491 (19·48)120 (18·69)72 (18·46)299 (20·08) Service industry561 (22·25)134 (20·87)88 (22·56)339 (22·77) Freelancer322 (12·77)73 (11·37)60 (15·38)189 (12·69) Other163 (6·47)34 (5·3)24 (6·15)105 (7·05)Religion5·417·067 Having233 (9·24)51 (7·94)28 (7·18)154 (10·34) Not having2288 (90·76)591 (92·06)362 (92·82)1335 (89·66)Living arrangement18·434·018 Alone338 (13·41)114 (17·76)43 (11·03)181 (12·16) With spouse370 (14·68)98 (15·26)51 (13·08)221 (14·84) With offspring477 (18·92)123 (19·16)76 (19·49)278 (18·67) With spouse and offspring916 (36·33)206 (32·09)151 (38·72)559 (37·54) Other420 (16·66)101 (15·73)69 (17·69)250 (16·79)Number of family members11·377·077 1363 (14·4)102 (15·89)62 (15·9)199 (13·36) 2416 (16·5)112 (17·45)57 (14·62)247 (16·59) 31052 (41·73)281 (43·77)155 (39·74)616 (41·37) ≥ 4690 (27·37)147 (22·9)116 (29·74)427 (28·68)Average monthly income15·272·054 ≤1000 CNY539 (21·38)123 (19·16)80 (20·51)336 (22·57) 1001ཞ3000 CNY1087 (43·12)313 (48·75)157 (40·26)617 (41·44) 3001ཞ5000 CNY477 (18·92)110 (17·13)76 (19·49)291 (19·54) 5001ཞ10000 CNY312 (12·38)69 (10·75)61 (15·64)182 (12·22) ≥ 10,001106 (4·2)27 (4·21)16 (4·1)63 (4·23)Payment method6·214·400 Urban Employee Basic Medical Insurance617 (24·47)157 (24·45)102 (26·15)358 (24·04) New Rural Cooperative Medical System936 (37·13)248 (38·63)135 (34·62)553 (37·14) Self-pay729 (28·92)186 (28·97)121 (31·03)422 (28·34) Other239 (9·48)51 (7·94)32 (8·21)156 (10·48)Number of chronic diseases72·008< ·001 2337 (13·37)98 (15·26)70 (17·95)169 (11·35) 3742 (29·43)237 (36·92)139 (35·64)366 (24·58) 4974 (38·64)205 (31·93)120 (30·77)649 (43·59) ≥ 5468 (18·56)102 (15·89)61 (15·64)305 (20·48)Longest duration of chronic disease39·005< ·001 ≤1 year311 (12·34)54 (8·41)63 (16·15)194 (13·03) 1ཞ5 years804 (31·89)178 (27·73)139 (35·64)487 (32·71) 5ཞ10 years486 (19·28)120 (18·69)73 (18·72)293 (19·68) ≥ 10 years920 (36·49)290 (45·17)115 (29·49)515 (34·59)Number of medicines11·644·070 0331 (13·13)64 (9·97)54 (13·85)213 (14·3) 1919 (36·45)228 (35·51)153 (39·23)538 (36·13) 2371 (14·72)103 (16·04)58 (14·87)210 (14·1) ≥ 3900 (35·7)247 (38·47)125 (32·05)528 (35·46)Primary caregiver15·709·047 None317 (12·57)80 (12·46)47 (12·05)190 (12·76) Spouse824 (32·69)239 (37·23)132 (33·85)453 (30·42) Offspring431 (17·1)116 (18·07)67 (17·18)248 (16·66) Nanny687 (27·25)147 (22·9)109 (27·95)431 (28·95) Other262 (10·39)60 (9·35)35 (8·97)167 (11·22)Social integration26·824< ·001 I336 (13·33)94 (14·64)39 (10·0)203 (13·63) II984 (39·03)273 (42·52)121 (31·03)590 (39·62) III881 (34·95)199 (31·0)168 (43·08)514 (34·52) IV320 (12·69)76 (11·84)62 (15·9)182 (12·22) Sense of meaning in life42·28 ± 19·8832·93 ± 16·1349·12 ± 21·0644·51 ± 19·66112·716< ·001Family function12·008·017 Severe impairment644 (25·55)181 (28·19)90 (23·08)373 (25·05) Moderate impairment942 (37·37)254 (39·56)134 (34·36)554 (37·21) Good935 (37·09)207 (32·24)166 (42·56)562 (37·74)Medical coping modes Confrontation21·62 ± 4·0316·31 ± 2·3924·99 ± 2·4723·03 ± 2·522069·848< ·001 Avoidance16·32 ± 3·6020·82 ± 2·3313·82 ± 2·2815·03 ± 2·501519·620< ·001 Resignation12·46 ± 2·4515·14 ± 1·538·60 ± 1·0912·32 ± 1·362785·719< ·001Method of questionnaire completion2.8970.235 Self-completed185 (7.34)159 (6.32)188 (7.45)179 (7.11) With assistance from family member/researcher2336 (92.66)2362 (93.68)2333 (92.55)2342 (92.89) Detailed descriptive information Latent profile analysis LPA was conducted on scaled scores for three coping modes: Confrontation, Avoidance, and Resignation (Table 2). Participants most favored “Confrontation” (score: 0.57), followed by “Resignation” (0.53) and “Avoidance” (0.44). Model performance metrics, including AIC, BIC, and aBIC, were negative but valid for comparative purposes [30]. From two (Model B) to three classifications (Model C), entropy slightly declined but remained acceptable (> 0.8), while AIC, BIC, and aBIC values decreased, and p-LMR and p-BLRT values were < 0.05, confirming Model C as superior. Adding a fourth class (Model D) reduced entropy below 0.8, and Model E further underperformed. Thus, the three-class solution (Model C) was optimal. Table 2Results of model fitModelClassificationAICBICaBICp-LMRp-BLRTEntropyProportion (%)A1−4746·764−4711·770−4730·833///1B2−7507·692−7449·367−7481·140< ·001< ·0010·92126·50/73·50C3−7782·309−7700·655−7745·136< ·001< ·0010·81015·47/59·06/25·47D4−7811·203−7706·220−7763·411< ·001< ·0010·62930·58/24·95/26·34/18·13E5−7822·573−7694·260−7764·160< ·001< ·0010·65321·02/24·67/17·49/1·94/34·88 Results of model fit Figure 1 shows the three-mode scores for the three patient groups. Helpless Group (red; 𝑁=642, 25.5%): Lowest “Confrontation” scores, moderate “Avoidance,” and highest “Resignation.” These patients were more likely to submit to and occasionally avoid dealing with their co-morbidities. Indomitable Group (blue; 𝑁=390, 15.5%): Highest “Confrontation” scores, moderate “Avoidance,” and lowest “Resignation.” These patients actively confronted their co-morbidities with minimal avoidance or resignation. Conflicting Group (green; 𝑁=1489, 59.1%): Balanced “Confrontation” and “Resignation” scores, with low “Avoidance.” This group alternated between confronting and yielding to their conditions. Latent profiles of Coping modes Factors influencing latent profiles Univariate tests showed significant differences in age, education, living arrangement, number of chronic diseases, primary caregiver, social integration, sense of meaning in life, and family functioning among three latent profiles (all p <.05; Table 1). Key factors were included in a multinomial logistic regression, validated by a likelihood ratio test (p <.001; Tables 3 and 4): (1) Indomitable Group vs. Helpless Group: Patients were younger (OR = 0.938, p <.001), had shorter chronic disease durations (OR = 0.701, p <.001), greater social integration (OR = 1.402, p <.001), higher life meaning (OR = 1.044, p <.001), and better family function (OR = 1.313, p =.003). Higher education levels (Senior high or technical secondary school: OR = 3.605; Bachelor’s or above: OR = 18.344; both p <.001) and cohabiting with a spouse (OR = 1.870, p =.013), offspring (OR = 2.172, p =.003), or both (OR = 2.043, p =.002) were more common; (2) Conflicting Group vs. Helpless Group: Patients were younger (OR = 0.961, p <.001), had more chronic diseases (OR = 1.357, p <.001) but shorter durations (OR = 0.793, p <.001), higher life meaning (OR = 1.034, p <.001), and better family function (OR = 1.180, p =.012). Higher education levels (Senior high or technical secondary school: OR = 2.476, p <.001; Bachelor’s or above: OR = 9.434, p =.001) and cohabiting with a spouse (OR = 1.531, p =.017), offspring (OR = 1.756, p =.002), or both (OR = 1.693, p =.001) were also more common. Table 3.Results of comparison between “Indomitable Group” and “Helpless Group” (Reference) Results of comparison between “Indomitable Group” and “Helpless Group” (Reference) Table 4.Results of comparison between “Conflicting Group” and “Helpless Group” (Reference) Results of comparison between “Conflicting Group” and “Helpless Group” (Reference) Discussion To our knowledge, this was the first cross-sectional study exploring coping modes in rural elderly patients with chronic co-morbidities using LPA to identify latent profiles and their influencing factors, offering new insights into medical coping modes. Three latent profiles of medical coping modes The highest scaled scores on the “Confrontation” mode seems to suggest a generally positive coping approach among rural elderly patients with multiple co-morbidities, as they tend to adopt active and constructive strategies. However, a closer examination of the latent profiles reveals a nuanced picture. The high proportion of patients in the “Conflicting Group” and the smaller number in the “Indomitable Group” creates what can be described as an “optimistic illusion” when viewed through a purely cross-sectional lens. This highlights the value of advanced methods like LPA, which uncover hidden patterns missed by conventional analyses. Patients were categorized into three distinct groups: the “Conflicting Group,” “Indomitable Group,” and “Helpless Group.” These findings partly align with Kenntemich et al.’s study of German patients during COVID-19, which identified five coping groups. Similarities include the overlap between the “Indomitable Group” and “High functional coping,” as well as the “Helpless Group” and “Low coping.” [31] Both studies underscore the importance of nuanced, data-driven approaches over simple categorizations when analyzing coping behaviors. In this study, the “Conflicting Group” represented the largest share (59.1%). These patients sought support from healthcare providers, family, friends, and social media to understand their condition but also felt hopeless about recovery and saw their situation as uncontrollable. This ambivalence likely arises from distrust in the healthcare system—potentially exacerbated by structural barriers such as geographic isolation, financial constraints (e.g., travel costs to tertiary hospitals), and fragmented care access—as well as negative past medical experiences or limited understanding of medical information [32–34]. To address these challenges, interventions should focus on enhancing patient knowledge, building trust in the doctor-patient relationship, and reducing feelings of powerlessness [35]. The “Helpless Group” (25.5%) often felt powerless, indifferent to their future, and unmotivated to manage their care. They avoided discussing their illness, distracted themselves, or refrained from medical decision-making. These behaviors may stem from societal shame surrounding chronic illness and low self-efficacy—the belief in one’s ability to manage challenges [36]. Chronic conditions, with symptoms like pain and fatigue, can erode confidence, causing anxiety or depression and further reducing self-efficacy [37]. Interventions should aim to empower these patients psychologically and strengthen their self-efficacy through targeted support [38]. The “Indomitable Group” (15.5%) actively engages in understanding their illness, participating in treatment, and adopting problem-solving strategies. They seek social support, experience less anxiety and depression, and achieve better clinical outcomes and quality of life [39]. While they require minimal intervention, sustaining their positive behaviors through education and psychosocial support is beneficial [40]. These patients can also serve as peer leaders, acting as role models to guide the other groups in adopting effective coping strategies. This approach is particularly valuable in rural areas with limited healthcare resources, fostering community-based support [41]. Factors influencing potential profiles The logistic regression showed a progression from the “Indomitable” to the “Conflicting,” and finally to the “Helpless Group” as age increased, consistent with findings by Cong et al. [42] This shift aligns with changes in Locus of Control, where patients transition from internal control (self-driven) to external control (fate or others) as they age [43]. Older rural patients often adopt beliefs such as “life is determined by fate,” influenced by socialization and limited medical resources in their environments [44]. Resignation, however, is not always negative. For some, it serves as an energy-saving strategy, allowing them to prioritize quality of life over ineffective treatments [45]. Whether this approach consistently improves quality of life requires further study. Additionally, the longer illness duration observed in the “Helpless Group” supports this transition pattern. Education, sense of meaning in life, and family functioning progressively increase from the “Helpless” to the “Conflicting” and the “Indomitable Group,” consistent with those of Dinh et al. [46], Kalla et al. [47], and Adesina et al. [48] Educated patients understand diseases, treatments, and health management better [49]. They also access reliable information, reducing uncertainty, and often believe that “knowledge can change destiny,” fostering proactive coping [50]. A strong sense of meaning helps patients view chronic illness as part of their life journey, encouraging personal growth and healthier behaviors [11]. As Viktor Frankl noted, “meaning” is one of humanity’s most powerful motivators in overcoming adversity [10]. Healthy family function provide emotional and practical support, easing isolation and psychological burdens [51]. Shared resources and responsibilities help patients engage more positively with their condition [52]. Meanwhile, patients in the “Indomitable” and “Conflicting” groups were more likely to live with family, highlighting the role of companionship in fostering resilience and effective coping. However, two unexpected findings emerged. First, patients in the Conflicting Group had more chronic co-morbidities but shorter illness durations than those in the Helpless Group. This may stem from their mixed coping mode—combining proactive and resigned behaviors—leading to inconsistent health management. While they may actively seek treatment during positive periods, their ambivalence can cause unsustainable efforts and chronic stress, which weakens immunity and worsens health [53]. Additionally, their proactive phases might result in earlier detection of conditions, uncovering more co-morbidities [54]. In contrast, the Helpless Group’s passive approach likely delays medical evaluations, leaving some conditions undiagnosed. Second, social integration levels were similar between the Conflicting and Helpless groups. While social integration reflects broad networks (e.g., friends, neighbors, community), it doesn’t always provide the meaningful, practical support needed to cope with chronic co-morbidities [55]. This highlights a distinction between generalized social connections and the deeper, more personal support offered by family functioning. Research limitations First, some variables that have been confirmed to have a significant association with medical coping modes are not significant influencing factors in our study, such as types of medical insurance. Future research with larger samples is still needed to obtain more robust conclusions. Second, while this study captured socioeconomic and geographic diversity through sampling in Guangdong and Shanxi provinces, the exclusion of impoverished western regions (e.g., Yunnan) due to logistical and human resource constraints may limit the generalizability of findings to all rural areas in China. Future research should incorporate broader geographic representation to account for regional disparities in healthcare infrastructure and socioeconomic conditions. Third, while formal education enhances health literacy, future studies should investigate how non-literate rural elderly acquire medical knowledge through oral traditions, community networks, or engagement with village healers. Such pathways may complement or substitute formal education in shaping coping behaviors, particularly in resource-limited settings. Forth, future research should investigate how social integration and family functioning interact—for instance, whether strong community networks mitigate the adverse effects of poor family functioning on coping behaviors. Such findings could inform policies that prioritize community-based interventions (e.g., village health committees, peer support groups) to supplement familial support systems, particularly in regions with high rates of labor migration or fragmented family structures. Conclusion This research identified three latent coping profiles in rural older adults with chronic co-morbidities: the “Helpless,” “Indomitable,” and “Conflicting” Groups. Influencing factors include age, education, living arrangement, number of chronic diseases, longest duration of diseases, social integration, sense of meaning in life and family functioning. Longitudinal studies are recommended to track changes in these coping profiles over time. Acknowledgements We also sincerely thank all the participants who participated in the study. Abbreviations adjusted BIC Akaike Information Criterion Bayesian Information Criterion Latent profile analysis Bootstrap Likelihood Ratio Test p-values from the Lo-Mendell-Rubin test Authors’ contributions HLM conceptualized the study, curated and analyzed the data, and developed the methodology. He also performed software validation, formal analysis, and visualization. Hualong Ma was the primary contributor to the original draft of the manuscript and participated in its review and editing; ZHL contributed to conceptualization, formal analysis, software validation, and methodology. She was also involved in visualization and contributed to the review and editing of the manuscript; YSZ participated in the study’s conceptualization, formal analysis, software validation, methodology, and visualization, and contributed to manuscript review and editing; XLL supervised the study and was responsible for project administration. He also contributed to investigation, resource acquisition, and data curation; XGL contributed to resource acquisition, investigation, data curation, and study conceptualization; DLC contributed to resource acquisition, investigation, data curation, and study conceptualization; JYL contributed to resource acquisition, investigation, data curation, and study conceptualization; JYL contributed to resource acquisition, investigation, data curation, and study conceptualization; LNG contributed to resource acquisition, investigation, data curation, and study conceptualization; YP contributed to resource acquisition, investigation, data curation, and study conceptualization; NY contributed to resource acquisition, investigation, data curation, and study conceptualization; XYF contributed to resource acquisition, investigation, data curation, and study conceptualization.; TJL contributed to resource acquisition, investigation, data curation, and study conceptualization; KQL contributed to resource acquisition, investigation, data curation, and study conceptualization; XYW contributed to resource acquisition, investigation, data curation, and study conceptualization; YXH contributed to resource acquisition, investigation, data curation, and study conceptualization; JHL contributed to resource acquisition, investigation, data curation, and study conceptualization; ZXY contributed to the study’s conceptualization, resource acquisition, investigation, data curation, and supervision; QYL contributed to conceptualization, resource acquisition, investigation, data curation, supervision, and manuscript review and editing; QHY conceptualized the study, developed the methodology, acquired resources, supervised the study, and administered the project. Additionally, Qiaohong Yang contributed to funding acquisition and the review and editing of the manuscript. All authors had confirmed that they had full access to all the data in the study and accept responsibility to submit for publication. Funding This study was supported by the Ministry of Science and Technology of China (National Key R&D Program, “Active Health and Technological Responses to Aging,” Grant No. 2020YFC2005700), the People’s Government of Guangdong Province (Guangdong Province’s “High-Quality Development Project for Hundreds of Counties, Thousands of Towns, and Tens of Thousands of Villages”), and the Guangdong Nursing Association and the Guangdong Lingnan Nightingale Nursing Research Institute (Grant No. GDSHLXHYJYY202607). Data availability Requests for deidentified data can be directed to the first author, Hualong Ma (email: Ma980383402@outlook.com), or the corresponding author, Qiaohong Yang (email: yqiaohong@163.com). Access to the data will be granted subject to approval of a research proposal and the signing of a data access agreement, in compliance with legal and ethical considerations. Declarations Ethics approval and consent to participate The present study was approved by the IRB of The First Affiliated Hospital of Jinan University (No. JNUKY-2024-0045). This study strictly adhered to the principles of medical ethics and the guidelines outlined in the Declaration of Helsinki. All participants provided written informed consent voluntarily after being fully informed. Inclusion and exclusion criteria were applied equitably, ensuring that no individual was excluded based on sex, race, religious beliefs, socioeconomic status, or cultural background et al. Consent for publication Written consent for publication was obtained from all participants. Competing interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Contributor Information Zaoxiu Yan, Email: 349456864@qq.com. Qiyuan Lyu, Email: yuanqlv@163.com. Qiaohong Yang, Email: yqiaohong@163.com. References Associated Data Data Availability Statement Requests for deidentified data can be directed to the first author, Hualong Ma (email: Ma980383402@outlook.com), or the corresponding author, Qiaohong Yang (email: yqiaohong@163.com). Access to the data will be granted subject to approval of a research proposal and the signing of a data access agreement, in compliance with legal and ethical considerations.