Unhealthy diet and lifestyle factors linked to female androgenetic alopecia: a community-based study from Jidong study, China

Abstract

Background

To explore unhealthy diet and lifestyle factors associated with female androgenetic alopecia (FAGA) in a community-based study.

Methods

A total of 3,008 participants were recruited from the Jidong community (Tangshan, Hebei, China). FAGA was assessed by the Savin scale. Data collection was conducted through structured questionnaires. Univariate and multivariate logistic regression analyses were employed to identify potential factors. Multivariable models were built by selecting covariates with P < 0.05 in univariable analyses and removing collinear variables identified by a variance inflation factor > 5. The predictive performance was evaluated using the receiver operating characteristic curve.

Results

The prevalence of FAGA in this study was 6.85% (206/3008). Sleep snoring (adjusted odds ratio [AOR] = 1.398, 95% confidence interval [CI] 1.032–1.894, P = 0.031), frequent consumption of bacon and preserved meat (AOR = 2.205, 95% CI 1.181–4.118, P = 0.013), frequently drinking unboiled water (AOR = 1.984, 95% CI 1.156–3.406, P = 0.013), age (AOR = 1.063, 95% CI 1.050–1.076, P < 0.001) and suboptimal health status score (AOR = 1.033, 95% CI 1.020–1.047, P < 0.001) were identified as independent factors associated with FAGA.

Conclusions

In conclusion, our data suggest that females who snore during sleep, frequently consume bacon and preserved meats, and regularly drink unboiled water are more susceptible to androgenetic alopecia. Additionally, our analysis revealed that age and suboptimal health status score are significantly correlated with FAGA.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-025-21560-7.

Introduction

Female androgenetic alopecia (FAGA) is the predominant form of alopecia among women [1]. FAGA is characterized by diffuse thinning of the mid-frontal scalp, while the frontal hairline is usually retained [1, 2]. According to prevailing social norms, women's hair is often associated with femininity, sexuality, attractiveness, health, and reproductive potential [3]. Previous research has demonstrated that female patients with androgenetic alopecia experience a diminished quality of life, lower emotional intelligence scores, and an increased vulnerability to anxiety relative to their male counterparts [4]. Although FAGA may not constitute a significant health risk, it exerts a substantial negative impact on women's quality of life [5].

Polygenic susceptibility and increased androgen action are the primary drivers of FAGA [6]. Additionally, lifestyle and dietary factors significantly influence the condition. Factors such as chronic stress, obesity, high-fat diets, glucose intolerance, alcohol misuse, and smoking can elevate androgen levels and accelerate hair loss in women with FAGA [7]. A cross-sectional study conducted in Iran identified that women with a genetic predisposition to FAGA may benefit from dietary modifications. Specifically, decreasing the intake of pro-inflammatory foods, such as trans fats and saturated fats, while increasing the consumption of anti-inflammatory foods, including fruits and vegetables, may aid in the prevention of hair loss [8]. A cross-sectional survey conducted in China examined the relationship between FAGA and behavioral factors. The results indicated that avoiding alcohol, refraining from tight hairstyles like ponytails, properly managing scalp oil, and improving sleep quality may help prevent the deterioration of hair loss [9]. However, there is a notable lack of studies specifically addressing the relationship between diet, lifestyle, and FAGA, highlighting a critical gap in the literature that warrants further exploration.

Suboptimal Health Status (SHS) is a physical state between health and disease, characterized by the perception of health complaints, general weakness, chronic fatigue and low energy levels [10]. Research has indicated that the incidence of SHS is greater in women compared to men [11]. Internationally acknowledged, SHS is utilized as an innovative tool for the early detection of chronic diseases [11, 12]. To date, no studies have investigated the potential association between SHS and FAGA. This relationship warrants further investigation to enhance our understanding of the underlying mechanisms and to potentially identify new strategies for prevention and intervention.

This study sought to examine the relationship between modifiable lifestyle factors—including obesity, smoking, alcohol consumption, sleep disturbances, and unhealthy diet—and FAGA. Additionally, the research assessed the influence of age and SHS on FAGA. The overarching goal is to provide scientific evidence and insights to reduce the incidence and progression of hair loss in women, thereby improving their quality of life.

Methods

Study design and population

In this cross-sectional study, participants were recruited from the Jidong Alopecia Cohort Study (JACS), based in Tangshan, a prominent industrial city in Hebei province, China, in 2015. From the total pool of potential participants, a random sample of 3,008 individuals was selected. All chosen participants were females aged 18 years or older, possessed comprehensive health information, and had provided informed consent to partake in the study. Individuals who did not meet these criteria were excluded from the study.

Data collection and measurement

All participants completed structured, self-administered questionnaires with assistance from well-trained researchers (see Supplementary File S1 for the questionnaire).

This structured instrument was utilized to gather data on demographic and social characteristics including sex, age, marital status, and educational level. Age was determined by subtracting the birth date on the participant's identification card from the study enrollment date. Marital status was categorized into three groups: unmarried, married, and divorced or widowed. Educational attainment was classified into three levels: primary education, secondary education, and higher education. Anthropometric measurements, including weight, height, and waist and hip circumferences, were recorded. Waist circumference was measured with an inelastic tape at the midpoint between the last rib and the iliac crest along the midaxillary line. It was categorized as either less than 80 cm or 80 cm and above, with 80 cm or more indicating abdominal obesity in women [13]. The Body Mass Index (BMI) was calculated by dividing weight in kilograms (kg) by the square of height in meters (m²). BMI status was determined according to the Chinese cut-off points: underweight (BMI < 18.5), normal weight (18.5 ≤ BMI < 24), overweight (24 ≤ BMI < 28), and obesity (BMI ≥ 28) [14].

Unhealthy diet and lifestyle habits

Data pertaining to unhealthy dietary and lifestyle habits were gathered using the structured instrument (Supplementary File S1). These variables included current smoking, passive smoking, current drinking, staying up late, taking sleep medications, sleep snoring, consumption of bacon and preserved meats, consumption of fried and spicy foods, and drinking unboiled water. The frequency of these unhealthy diet habits was categorized into three groups: never, occasionally (≤ 1–2 times per week) and frequently (≥ 3–5 times per week).

Suboptimal health status

SHS was assessed using the Suboptimal Health Status Questionnaire-25 (SHSQ-25) (Supplementary File S1) [10].

FAGA diagnosis and severity

FAGA was diagnosed through clinical examination of the hair and scalp, which revealed non-scarring alopecia in a characteristic pattern [15]. The severity of FAGA was assessed using the Savin Scale [16], a widely recognized instrument in the field, with classification verification performed by on-site medical professionals. The specifics of the Savin Scale are illustrated in Fig. 1, which assesses the general thinning of the crown scalp through 8 images depicting varying degrees of FAGA. The Frontal stage in the scale specifically indicates frontal anterior recession. In accordance with established the guideline and the Savin Scale, participants were categorized into two groups: non-FAGA (I-1) and FAGA (I-2, I-3, I-4, II-1, II-2, III, Advanced, Frontal).

Fig. 1.

The Savin Scale

The severity classification of female androgenetic alopecia reflects a range from no hair loss to severe hair loss, including stages I-1, I-2, I-3, I-4, II-1, II-2, III, and advanced. The Frontal stage in the scale specifically indicates frontal anterior recession.

Statistical analysis

Continuous variables were reported as mean ± standard deviation (SD) and group differences were assessed using t-tests. Categorical variables were presented as frequency and percentage (%), with group differences analyzed using chi-square tests or Fisher's exact probability method. For variables that demonstrated significance in the univariate analysis, a collinearity assessment was conducted using the variance inflation factor (VIF), where a VIF greater than five indicated significant collinearity. For multiple logistic regression analyses, covariates were selected with P < 0.05 in univariable analyses and removing collinear variables identified. The adjusted odds ratio (AOR), 95% confidence interval (CI), and P value was reported. Predictive performance was evaluated through a receiver operating characteristic (ROC) curve analysis. Additionally, age-stratified analysis was conducted (< 50 and ≥ 50 years). All statistical analyses were executed using R version 4.0.3. Statistical significance was determined by a P value < 0.05.

Results

Baseline characteristics comparison between FAGA and Non-FAGA subjects

Table 1 presents a summary of the demographic characteristics, unhealthy diet and lifestyle-related factors, and SHS scores of the participants, categorized by the presence of FAGA. A comparison between participants with FAGA and those without revealed significant differences in age, marital status, education level, waist circumference, BMI and SHS score (P < 0.05). Regarding unhealthy lifestyles, there were significant differences between the two groups in the use of sleep medications and the occurrence of snoring during sleep (P < 0.05). With respect to unhealthy diet, significant differences were noted between the groups in the consumption of bacon and preserved meat and drinking unboiled water (P < 0.05). The results of the collinearity analyses (Table S1) indicated that no collinearity was found between the variables (VIF < 5).

Table 1.

Baseline characteristics between FAGA and Non-FAGA Subjects

Characteristics	Overall	Non-FAGA	FAGA	P-value
	(n = 3008)	(n = 2802)	(n = 206)
Age, mean ± SD, y	43.78 ± 13.41	42.98 ± 13.26	54.61 ± 10.35	< 0.001
Marital status, No (%)
Unmarried	157 (5.2)	155 (5.5)	2 (1.0)	< 0.001
Married	2813 (93.5)	2617 (93.4)	196 (95.1)
Divorced/widowed	38 (1.3)	30 (1.1)	8 (3.9)
Education level, No (%)
Primary school education level	211 (7.0)	176 (6.3)	35 (17.0)	< 0.001
Secondary school education level	1151 (38.3)	1034 (36.9)	117 (56.8)
Higher school education level	1646 (54.7)	1592 (56.8)	54 (26.2)
Hip circumference, mean ± SD, cm	97.93 ± 10.07	97.76 ± 10.08	100.35 ± 9.60	0.727
Waist circumference, No (%)
< 80 cm	1385 (46.0)	1318 (47.0)	67 (32.5)	< 0.001
≥ 80 cm	1623 (54.0)	1484 (53.0)	139 (67.5)
BMI, No(%)
Underweight	138(4.6)	134(4.8)	4(2.0)	< 0.001
Normal weight	1654 (55.0)	1566(55.9)	88 (42.7)
Overweight	875 (29.1)	795 (28.4)	80 (38.8)
Obesity	341 (11.3)	307 (10.9)	34 (16.5)
Current smoking, No (%)	52 (1.7)	46 (1.6)	6 (2.9)	0.283
Passive smoking, No (%)	980 (32.6)	901 (32.2)	79 (38.3)	0.067
Current drinking, No (%)	168 (5.6)	158 (5.6)	10 (4.9)	0.636
Staying up late, No (%)
Never or occasionally	2269 (75.4)	2107 (75.2)	162 (78.6)	0.268
Frequently	739 (24.6)	695 (24.8)	44 (21.4)
Taking sleep medications, No (%)	84 (2.8)	69 (2.5)	15 (7.3)	< 0.001
Sleep snoring, No (%)	1202 (40.0)	1082 (38.6)	120 (58.3)	< 0.001
Consumption of bacon and preserved meat, No (%)
Never or occasionally	2899 (96.4)	2707 (96.6)	192 (93.2)	0.012
Frequently	109 (3.6)	95 (3.4)	14 (6.8)
Consumption of fried foods, No (%)
Never or occasionally	2669 (88.7)	2487 (88.8)	182 (88.3)	0.858
Frequently	339 (11.3)	315 (11.2)	24 (11.2)
Consumption of spicy foods, No (%)
Never or occasionally	2239 (74.4)	2076 (74.1)	163 (79.1)	0.11
Frequently	769 (24.6)	726 (25.1)	43 (18.9)
Drinking unboiled water, No (%)
Never or occasionally	2871 (95.4)	2684 (95.8)	187 (90.8)	0.001
Frequently	137 (4.6)	118 (4.2)	19 (7.2)
SHS score, mean ± SD	13.14 ± 9.99	12.78 ± 9.93	18.08 ± 9.49	< 0.001

FAGA Female androgenetic alopecia, SD Standard deviation, SHS Suboptimal Health Status, BMI Body Mass Index

The prevalence of FAGA in the study population

A total of 3008 subjects were enrolled in the study (mean age 43.8 ± 13.4 years). The overall prevalence of FAGA was determined to be 6.85% (206/3008). Subtype I-2 exhibited the highest prevalence at 4.79% (144/3008), followed by subtype I-3 at 1.1% (33/3008) and subtypes II-2 at 0.4% (12/3008). Figure 2 illustrates the prevalence of FAGA and its subtypes among study participants.

Fig. 2.

The prevalence of female androgenetic alopecia and its subtypes among participants

Associations of unhealthy diet and lifestyle factors with FAGA

The results of the multivariate analysis are presented in Table 2. Sleep snoring (AOR = 1.398, 95% CI 1.032–1.894, P = 0.031), frequent consumption of bacon and preserved meat (AOR = 2.205, 95% CI 1.181–4.118, P = 0.013), frequently drinking unboiled water (AOR = 1.984, 95% CI 1.156–3.406, P = 0.013) were identified as independent factors significantly associated with FAGA.

Table 2.

Adjusted Odds Ratios of influencing factors for FAGA

Associations between FAGA and unhealthy diet and lifestyle factors, age and SHS score by multiple logistic regression analyses. Multivariable models were built by selecting covariates with P < 0.05 in univariable analyses and removing collinear variables identified by a variance inflation factor > 5. FAGA Female androgenetic alopecia, SHS Suboptimal Health Status, AOR adjusted odds ratio, CI confidence intervals

Associations of age and SHS with FAGA

Age (AOR = 1.063, 95% CI 1.050–1.076, P < 0.001) and SHS score (AOR = 1.033, 95% CI 1.020–1.047, P < 0.001) were identified as independent factors significantly associated with FAGA.

ROC curve analysis of multivariate logistic regression model

The predictive performance was analyzed using an ROC curve, which is presented in Fig. 3. The area under the curve (AUC) of the logistic regression model was 0.780 (95% CI: 0.753–0.806), indicating good discriminative ability.

Fig. 3.

The receiver operating characteristic curves of the multivariate logistic regression model. A ROC curve of model. The receiver operating characteristic curve demonstrates the predictive performance. Area under the curve (AUC) = 0.780 (95% CI: 0.753–0.806). B ROC curve of frequent consumption of bacon and preserved meat. AUC = 0.517. C ROC curve of frequently drinking unboiled water. AUC = 0.525. D ROC curve of sleep snoring. AUC = 0.598. E ROC curve of age. AUC = 0.748. F ROC curve of SHS score. AUC = 0.673

Age-stratified analysis

The prevalence of FAGA rises with age, peaking after 50 years [1]. Our study found a prevalence of 3.4% in those under 50 and 13.1% in those 50 and older, a significant increase. We stratified participants into two age groups (< 50 and ≥ 50 years) to assess the impact of lifestyle and SHS scores on FAGA. Baseline characteristics and collinearity analyses are detailed in Tables S2-S5.

In the under-50 group, education level (higher school education level) (AOR = 0.557, 95% CI 0.333–0.934, P = 0.026), sleep snoring (AOR = 2.594, 95% CI 1.572–4.281, P < 0.001), frequent consumption of bacon and preserved meat (AOR = 2.429, 95% CI 1.073–5.500, P = 0.033), and SHS score (AOR = 1.035, 95% CI 1.013–1.059, P = 0.002) were identified as independent factors associated with FAGA (Table 3). The logistic regression model's AUC was 0.729, indicating acceptable discriminative ability (Fig. 4).

Table 3.

Adjusted Odds Ratios of influencing factors for FAGA (Age < 50y)

Variables	AOR	95%CI	P value
SHS score	1.035	1.013–1.059	0.002
Education level (Higher school education level)	0.557	0.333–0.934	0.026
Sleep snoring	2.594	1.572–4.281	< 0.001
Consumption of bacon and preserved meat (Frequently)	2.429	1.073–5.500	0.033

Multivariable models were built by selecting covariates with P < 0.05 in univariable analyses and removing collinear variables identified by a variance inflation factor > 5. FAGA Female androgenetic alopecia, SHS Suboptimal Health Status, AOR Adjusted odds ratio, CI confidence intervals

Fig. 4.

The receiver operating characteristic curves of the multivariate logistic regression model (Age < 50y). A ROC curve of model. The receiver operating characteristic curve demonstrates the predictive performance. Area under the curve (AUC) = 0.729. B ROC curve of education level (higher school education level). AUC = 0.420. C ROC curve of frequent consumption of bacon and preserved meat. AUC = 0.540. D ROC curve of sleep snoring. AUC = 0.628. (E) ROC curve of SHS score. AUC = 0.660

For the 50 and older group, frequently drinking unboiled water (AOR = 2.248, 95% CI 1.182–4.273, P = 0.013), and SHS score (AOR = 1.035, 95% CI 1.018–1.052, P < 0.001) were identified as independent factors associated with FAGA (Table 4). The AUC of the logistic regression model was 0.644, indicating certain discriminative ability (Fig. 5).

Table 4.

Adjusted Odds Ratios of influencing factors for FAGA (Age ≥ 50y)

Variables	AOR	95%CI	P value
SHS score	1.035	1.018–1.052	< 0.001
Drinking unboiled water (Frequently)	2.248	1.182–4.273	0.013

Multivariable models were built by selecting covariates with P < 0.05 in univariable analyses and removing collinear variables identified by a variance inflation factor > 5. FAGA Female androgenetic alopecia, SHS Suboptimal Health Status, AOR Adjusted odds ratio, CI Confidence intervals

Fig. 5.

The receiver operating characteristic curves of the multivariate logistic regression model (Age ≥ 50y). A ROC curve of model. The receiver operating characteristic curve demonstrates the predictive performance. Area under the curve (AUC) = 0.644. B ROC curve of frequently drinking unboiled water. AUC = 0.527. C ROC curve of SHS score. AUC = 0.632

Discussion

In the cross-sectional analysis, approximately 7% of the participants were found to have FAGA. Multivariate analysis identified significant associations with sleep snoring, frequent consumption of bacon and preserved meat, frequently drinking unboiled water, age and SHS score. ROC analyses illustrate that the model (AUC of 0.780) had good predictive performance. While FAGA may not pose a major health risk, it significantly affects women's quality of life. Table 5 summarizes the primary risk factors and provides recommendations for a healthy lifestyle to prevent FAGA. Our findings provide compelling new evidence for the relationship between unhealthy diet and lifestyle factors and FAGA.

Table 5.

Practical Suggestions to Prevent FAGA in Women

Risk factor	Practical Suggestions
Sleep snoring	Lie on your side to reduce snoring
	Use lip tape or nasal dilators to improve nasal breathing
	In severe cases, consider mouth protection (oral device treatment) or sleep apnea machines (positive airway pressure treatment)
Consumption of bacon and preserved meat (Frequently)	Reduce intake of processed meats and opt for healthier alternatives like fresh fruits, vegetables, and lean proteins
Drinking unboiled water (Frequently)	Ensure the safety of drinking water by boiling it if necessary before consumption

FAGA Female androgenetic alopecia

Our findings indicate that sleep snoring, frequent consumption of bacon and preserved meats, and habitual consumption of unboiled water are independent factors contributing to FAGA in relation to diet and lifestyle. Chinese bacon, a traditional Chinese food, typically refers to smoked and preserved meat, which falls under the category of processed meats [17]. Bazmi et al. identified that a reduction in the consumption of pro-inflammatory foods is associated with a decreased incidence of FAGA [8]. The findings of our study align with this conclusion, indicating that the intake of bacon and preserved meat, which are recognized as pro-inflammatory [18], significantly elevates the risk of FAGA. Additionally, the risk is more significant in women younger than 50 years old. These results reinforce the hypothesis that dietary factors are pivotal in the development and progression of FAGA.

Our study identified the frequent consumption of unboiled water as an independent risk factor for FAGA. Notably, this risk appears to be more pronounced among the elderly population (individuals over 50 years of age). While modern water treatment facilities in developed countries and regions generally ensure the provision of safe drinking water, numerous developing areas, including China and South Asia, still rely on boiling water as a primary method to ensure its safety [19–21]. A systematic assessment of the sanitary status of drinking water in China from 2007 to 2018 revealed that sanitary conditions remain suboptimal, with microbial contamination posing the most significant threat to water safety [22]. The consumption of unboiled water has also been reported to carry a heightened risk of microbial contamination, potentially leading to the presence of pathogens such as Escherichia coli and Cryptosporidium, consequently increasing the likelihood of gastroenteritis [23–25]. Regular consumption of unboiled water may lead to modifications in the gut microbiome. Interestingly, it has been reported that the gut microbiome plays a significant role in the development of androgenetic alopecia, although the exact mechanism remains unclear and warrants further investigation [26].

Our study investigated the influence of sleep-related lifestyle factors, such as staying up late, snoring, and taking sleep medications, on FAGA. The findings suggest that women who experience snoring during sleep exhibit an increased risk of developing FAGA, with this risk being particularly pronounced in women under the age of 50. Snoring, a primary symptom of obstructive sleep apnea, is commonly regarded as an indicator of sleep-disordered breathing [27]. Recent research has associated sleep apnea with various dermatological conditions, including male androgenetic alopecia, and suggests a potential link to hypoxia [28, 29]. However, our study lacks detailed assessments of snoring, underscoring the need for further research on the impact of sleep snoring on FAGA. We also investigated the link between alcohol, smoking, and FAGA, including passive smoking among women. Our findings showed no significant connection between alcohol or smoking and FAGA. However, earlier research in China suggested that alcohol is a risk factor for FAGA in women, while smoking is not; however, it lacked clear details on alcohol consumption levels [9]. This underscores the need for more evidence-based studies.

Our study has identified age as an independent risk factor for FAGA. This finding aligns with previous community surveys conducted in Shanghai and six other cities—Zibo, Taiyuan, Xichang, Hailar, Langfang, and Jiaozuo—which have demonstrated a progressive increase in the prevalence of FAGA with advancing age [30, 31]. Furthermore, our research indicates that higher SHS scores are associated with an increased risk of developing FAGA. Although direct evidence linking SHS to FAGA is limited, studies on suboptimal health and chronic conditions provide indirect support for our findings. Notably, research indicates that individuals with higher SHS scores have significantly elevated serum cortisol levels, highlighting stress as a key factor in SHS [32]. Elevated stress levels may impact neurotrophic factors in androgenic alopecia patients, worsening the condition [33]. Additionally, metabolomic analysis shows lower progesterone in those with SHS, linking poor health to hormonal imbalances that negatively affect hair follicles [34, 35]. Our study identifies SHS as a possible risk factor for FAGA and suggests new research avenues.

The strength of this study lies in its community-based design and large sample size. Nevertheless, our study is subject to certain limitations. Firstly, the utilization of cross-sectional survey data restricted our capacity to establish causality, as we were only able to establish statistical correlations and may have been prone to recall bias. Secondly, this study employed the SHS score as a continuous measure. The clinical examinations and SHS classification of participants may have provided a more precise assessment of the true risk of SHS, but conducting such an analysis would necessitate a larger sample size, increased time, and financial resources. Furthermore, the variables pertaining to unhealthy eating habits and lifestyle were not comprehensively addressed in this study. Specifically, the influence of physical activity was not examined. Finally, the study population was confined to a single region within the nation, which may restrict the generalizability of our findings to other areas. Consequently, additional research is warranted to explore these aspects in future investigations.

Conclusions

In summary, our study identified significant associations between the frequent consumption of bacon and preserved meats, the consumption of unboiled water, sleep snoring, and FAGA, in the context of dietary and lifestyle factors. Furthermore, our analysis demonstrated significant correlations between age, SHS score, and FAGA. Specifically, our findings are expected to contribute to mitigating the incidence and progression of hair loss in women.

Abbreviations

FAGA

Female androgenetic alopecia

JACS

Jidong Alopecia Cohort Study

BMI

Body mass index

SHS

Suboptimal Health Status

SHSQ-25

Suboptimal Health Status Questionnaire-25

VIF

Variance inflation factor

ROC

Receiver operating characteristic

AUC

Area under the curve

OR

Odds ratios

AOR

Adjusted odds ratio

CI

Confidence interval

Authors’ contributions

LP, YZ and XH contributed to the concept and design of the study. YHL and WQW cleaned the data. LP and YHL performed data statistical analyses and interpretation. LP drafted the original manuscript. YHL, XH and YZ polished and edited the manuscript. All authors have read and approved the final manuscript.

Funding

This work was supported by grants from the National Natural Science Foundation of China (No.82073452 and No.81772161), Natural Foundation Project of Shanghai Science and Technology Commission (17ZR1426300), Shanghai General Hospital Integrated Traditional Chinese and Western Medicine Special Project (ZHYY-ZXYJHZX-202002), Shanghai Tongji Hospital Clinical Research and Cultivation Key Project (ITJ(ZD)1903), Shanghai Tongji Hospital Clinical "Five New" Innovation R&D Project (ITJ(ZD)2306), Shanghai Outstanding Young Medical Talent Training Funding Program and Clinical Research Plan of SHDC (SHDC22022302), Shanghai Hospital Development Center Foundation (SHDC12024144).

Data availability

Data is provided within the manuscript and supplementary information files.

Declarations

Ethics approval and consent to participate

All research procedures are conducted in accordance with the Helsinki Declaration, which has been approved by the Ethics Committee of the Staff Hospital of Jidong Oil field of China National Petroleum Corporation (No. 2013 YILUNZI 1). These approvals are renewed every five years to ensure ongoing compliance with ethical standards. Participants also signed informed consent forms before interviews.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.