The mediating effect of perceived stress on the relationship between big five personality traits and suboptimal health status in Chinese population: a nationwide survey in the framework of predictive, preventive, and personalized medicine

Abstract

Background

The effects of psychological factors on suboptimal health status (SHS) have been widely described; however, mechanisms behind the complex relationships among the Big Five personality traits and SHS are unclear. Identifying people with specific traits who are susceptible to SHS will help improve life quality and reduce the chronic disease burden under the framework of predictive, preventive, and personalized medicine (PPPM / 3PM). This study investigated the relationships among personality traits and SHS. It also explored whether perceived stress plays a mediating role in SHS development.

Method

A nationwide cross-sectional survey based on multistage random sampling was conducted in 148 cities in China between June 20 and August 31, 2022. Personality traits, perceived stress, and SHS were evaluated using the Big Five Inventory-10 (BFI-10), the 4-item Perceived Stress Scale (PSS-4), and the Short-Form Suboptimal Health Status Questionnaire (SHSQ-SF), respectively. Pearson’s correlation analysis was employed to examine the associations between personality traits, perceived stress, and SHS. Structural equation modeling (SEM) was used to discern the mediating role of perceived stress in the relationships among personality traits and SHS.

Result

A total of 22,897 participants were enrolled in this study, among whom the prevalence of SHS was 52.9%. SHS was negatively correlated with three trait dimensions (i.e., extraversion, agreeableness, and conscientiousness) but positively correlated with neuroticism. Meanwhile, stress was negatively correlated with extraversion, agreeableness, conscientiousness, and openness, whereas it was positively correlated with neuroticism. The SEM results showed that, when adjusting for covariates (i.e., gender, age, BMI, educational level, current residence, marital status, and occupational status), higher agreeableness (β =  − 0.049, P < 0.001) and conscientiousness (β =  − 0.103, P < 0.001) led to lower SHS prevalence, higher neuroticism (β = 0.130, P < 0.001), and openness (β = 0.026, P < 0.001) caused SHS to be more prevalent. Perceived stress played a partial mediating role in the relationships among personality traits and SHS, respectively, contributing 41.3%, 35.9%, and 32.5% to the total effects of agreeableness, conscientiousness, and neuroticism on SHS. Additionally, the mediating impact of stress was significant even though extraversion had no direct effect on SHS.

Conclusion

This study revealed a high prevalence of SHS in Chinese residents. Personality traits significantly influenced SHS rates, which perceived stress tended to mediate. From a PPPM perspective, early screening and targeted intervention for people with neuroticism (as well as stress alleviation) might contribute to health enhancement and chronic disease prevention.

Supplementary Information

The online version contains supplementary material available at 10.1007/s13167-023-00349-x.

Introduction

Suboptimal health status (SHS) is an intermediate state between optimal health and illness, characterized by health complaints, perceived general weakness and lethargy, and no clear clinical diagnosis [1]. Although SHS cannot be labeled as a specific disease, it has been highly associated with chronic disease development among multiple populations such as East Asians, Africans, and Australians [2, 3]. Epidemiological surveys have shown that approximately half of the Chinese population experience SHS [4, 5]. SHS typically precedes chronic diseases, including type 2 diabetes mellitus [6, 7] and cardiovascular diseases (CVD) [4, 8]; that is, it represents a subclinical disease stage. International multi-disciplinary expert groups strongly recommend implementing cost-effective strategies and paradigm change from reactive medicine to predictive, preventive, and personalized medicine (PPPM/3PM), so as to reduce the public health burdens induced by preventable non-communicable diseases (NCDs) and improve health status [9–11].

Identifying SHS and its relevant determinants early will facilitate targeted prevention and personalized treatment of chronic diseases from the perspective of PPPM/3PM [12]. Adverse psychological factors have been widely described in SHS development [13]. However, it remains unclear whether personality plays a part. Studies have documented personality as a major predictor of behavior, including response patterns, movement, and health status [14]. The Big Five model of personality describes traits across five dimensions: extraversion, agreeableness, conscientiousness, neuroticism, and openness [15]. The strength of each domain can have unique effects on the human body in multiple respects [16]. Thus, personality traits have been measured to examine their potential roles in predicting health-related outcomes [17]. Extraversion [18–20], agreeableness [21, 22], conscientiousness [23], and openness [20] are positively associated with health, whereas neuroticism can contribute to potentially suboptimal health outcomes [24].

Recent work has shown that personality traits are also associated with perceived stress, which greatly affects health [25]. In particular, neuroticism is positively associated with stress; extraversion, agreeableness, conscientiousness, and openness are negatively related to this response. Furthermore, stress puts one at higher risk of SHS [26, 27]. These findings reveal a complex relationship between personality and stress in relation to SHS, which prompts developing innovative screening programs considering advanced predictive models followed by targeted prevention and personalized intervention tailored to the person at high risk of SHS [11].

Working hypothesis and study aims in the PPPM/3PM framework

The underlying causes of SHS have not been documented clearly. Lifestyles, such as poor dietary intake, cigarette smoking, alcohol drinking, and physical inactivity, play important roles in development of SHS [28]. Moreover, psychological determinants are associated with SHS and always applied for measurement of SHS in mental domain [29]. Mental health promotion is crucial for health management of Chinese residents who have been experiencing an increasing prevalence of mental disorder [30]. Although mental health promotion programs have been widely developed, the health challenges resulting from personality traits are often overlooked. From the perspectives of PPPM/3PM, baseline information of SHS, as well as the relationship between personality traits and SHS, is urgently needed to be investigated.

In the current study, we hypothesized that personality traits and perceived stress might be instrumental to predict SHS for personalized supervision of specific determinants susceptible to SHS. The aim of this study is therefore to investigate the relationship between personality traits and SHS, and to discuss the feasibility of improving health status in Chinese residents according to PPPM/3PM concepts.

Methods

Study design and study participants

This study was a nationwide cross-sectional survey conducted in 148 cities in China [31]. A multistage random sampling method was adopted to recruit participants: 1) probability sampling at the provincial, municipal, district/county, and community/village levels; and 2) quota sampling from the community/village level to the individual level. Sampling ratios were established in 23 provinces, five autonomous areas, and four municipalities based on population proportions from the Seventh National Census Data of China. The Shaanxi Health Culture Research Center ethics review board approved this research protocol, which has been entered into the Chinese Clinical Trial Registry (registration number ChiCTR2200061046).

Participants who met the following criteria were included in this study: 1) ≥ 16 years old; 2) permanent residents with Chinese nationality; and 3) possessed the necessary skills to read and comprehend the questionnaires. Exclusion criteria covered 1) individuals currently diagnosed with somatic diseases; 2) individuals with current psychiatric abnormalities or cognitive impairment; and 3) individuals participating in other clinical investigations.

Parameter measurement

Investigators who participated in face-to-face interviews were publicly recruited from each city’s local university. All investigators completed training courses in quality control, research tools, and sampling techniques and were subsequently tested in accordance with the predefined training protocol. All study participants provided written informed consent.

Demographic characteristics

Participants’ demographic characteristics, including gender, age, body mass index (BMI), educational level, current residence (urban/rural), marital status, and occupational status, were acquired through face-to-face interviews.

Big Five Inventory-10 items (BFI-10)

BFI-10, a self-report questionnaire whose items are scored on a 5-point Likert-type scale, was used to measure participants’ personality traits (Table S1). This scale includes 10 items representing five trait dimensions (i.e., extraversion, agreeableness, conscientiousness, neuroticism, and openness). The higher one’s score on a given domain, the more prominent that trait is. BFI-10 has shown good reliability and validity in our investigation across China [32].

Perceived Stress Scale-4 items (PSS-4)

PSS-4 was adopted to measure participants’ perceived stress (Table S2). Four items across two dimensions are rated on a 5-point Likert-type scale. The total score ranges between 4 and 20, with a higher score indicating greater perceived stress. This scale has demonstrated sound validity and reliability in previous studies [33].

Short-Form Suboptimal Health Status Questionnaire (SHSQ-SF)

The SHSQ-SF contains nine items scored on a 5-point Likert-type scale (Table S3). This instrument was applied to assess participants’ health status. Each item’s score ranges from 0 (almost none) to 4 (almost always). A total score ≥ 11 indicates SHS; otherwise, optimal health is confirmed.

Pilot study

Four hundred participants were enrolled in this study’s three rounds of pre-investigation, which were completed June 5–8, June 10–13, and June 15–18, 2022. Feedback and guidance obtained from the participants and researchers during the pilot study were promptly compiled. The questionnaire was finalized following three rounds of pre-investigation.

Statistical analysis

Descriptive statistics elucidated participants’ sociodemographics. Pearson’s correlation was used to examine correlations among personality traits, stress, and SHS. All measures’ reliability and validity were determined based on Cronbach’s α and confirmatory factor analysis (CFA). In CFA, results for the hypothesized model showed good fit indices overall: root mean square error of approximation = 0.039 < 0.080, comparative fit index = 0.991 > 0.950, Tucker–Lewis index = 0.954 > 0.950. A mediation analysis and structural equation modeling (SEM) were carried out to ascertain the mediating effect of stress on the relationship between personality and SHS.

In the SEM procedure on SHS, the five personality traits were taken as independent variables with stress serving as a mediator. In addition, the impacts of covariates (i.e., gender, age, BMI, educational level, current residence, marital status, and occupational status) were controlled. The maximum-likelihood method was applied for asymptotically unbiased, consistent, and efficient estimators. Path significance was verified via bias-corrected bootstrapping with 5000 iterations. Data analyses were completed in IBM SPSS (version 26.0, Armonk, NY, USA) and MPLUS (version 8.3, Linda, Bengt, LA). All statistical tests were two sided with statistical significance denoted at P < 0.05.

Results

Participant recruitment and characteristics

A total of 30,505 participants were recruited for this survey, among whom 7608 were excluded for the following reasons: 1) 1182 with chronic diseases; 2) 4352 with unavailable SHS data; and 3) 2074 either under age 16 or over age 60. Ultimately, 22,897 participants were enrolled. The flowchart of recruitment appears in Fig. 1; participants’ regional distribution is displayed in Fig. 2.

Fig. 1.

Flowchart of participants recruitments. A total of 30,505 participants were recruited for this survey, among whom 7608 were excluded for the following reasons: 1) 1182 with chronic diseases; 2) 4352 with unavailable SHS data; and 3) 2074 either under age 16 or over age 60. Ultimately, 22,897 participants were enrolled

Fig. 2.

Regional distribution of study participants. This study was a nationwide cross-sectional survey conducted in 148 cities in China. A multistage random sampling method was adopted to recruit participants: 1) probability sampling at the provincial, municipal, district/county, and community/village levels; and 2) quota sampling from the community/village level to the individual level. Sampling ratios were established in 23 provinces, five autonomous areas, and four municipalities based on population proportions from the Seventh National Census Data of China

Regarding participants’ characteristics, 9841 (43.0%) were men and 13,056 (57.0%) were women. Of them, 10,827 (47.3%) were younger than 24, and 4667 (20.4%) were older than 45. Geographically, 5662 (24.7%) participants were from rural areas and 17,235 (75.3%) were from urban areas. Participants’ details are summarized in Table 1.

Table 1.

Sociological characteristics of participants

Variables		N	Health (N = 10,789)	SHS (N = 12,108)	χ2	P
Gender	Male	9841	4971 (50.5)	4870 (49.5)	79.762	< 0.001
	Female	13,056	5818 (44.6)	7238 (55.4)
Age (years)	≤ 24	10,827	4613 (42.6)	6214 (57.4)	173.470	< 0.001
	25–	4046	2037 (50.4)	2009 (49.6)
	35–	3357	1688 (50.3)	1669 (49.7)
	≥ 45	4667	2451 (52.5)	2216 (47.5)
BMI categories	Thin	3763	1578 (41.9)	2185 (58.1)	55.823	< 0.001
	Normal	14,090	6805 (48.3)	7285 (51.7)
	Overweight	4020	1945 (48.4)	2075 (51.6)
	Obese	1024	461 (45.0)	563 (55.0)
Educational level	Middle school and below	3493	1972 (56.5)	1521 (43.5)	159.789	< 0.001
	Senior high school	5816	2764 (47.5)	3052 (52.5)
	College	12,673	5630 (44.4)	7043 (55.6)
	Master and above	915	423 (46.2)	492 (53.8)
Marital status	Single	12,823	5531 (43.1)	7292 (56.9)	208.477	< 0.001
	Married	9651	5085 (52.7)	4566 (47.3)
	Divorced or widowed	423	173 (40.9)	250 (59.1)
Occupational status	Student	10,244	4361 (42.6)	5883 (57.4)	159.947	< 0.001
	Employed	10,877	5511 (50.7)	5366 (49.3)
	Retired	474	223 (47.0)	251 (53.0)
	Unemployment	1302	694 (53.3)	608 (46.7)
Place of residence	Rural	5662	2654 (46.9)	3008 (53.1)	0.182	0.669
	Urban	17,235	8135 (47.2)	9100 (52.8)

Qualitative variables are presented as count (%). Of 22,897 eligible participants, 10,789 (47.1%) individuals were healthy and 12,108 (52.9%) displayed SHS

SHS, suboptimal health status

SHS prevalence

As listed in Table 1, the prevalence of SHS was 52.9% (12,108/22,897). This rate was significantly higher in women (55.4%; 7238/13,056) than in men (49.5%; 4870/9841). Compared with participants over age 45 (47.5%; 2216/4667), those under age 24 had a higher SHS prevalence (57.4%; 6214/10,827). As for marital status, divorced or widowed participants were more susceptible to SHS (59.1%; 250/423) than married participants (47.3%; 4566/9651). Participants with a college education had a relatively high prevalence of 55.6% (7043/12,673). Participants were further classified into four groups in accordance with Chinese BMI criteria [34]: thin (BMI < 18.5), normal (18.5 ≤ BMI < 24), overweight (24 ≤ BMI < 28), and obese (BMI ≥ 28). The prevalence of SHS was highest among underweight participants (58.1%; 2185/3763). No significant differences were observed between rural and urban participants.

Participants’ personality traits

Standardized scores were calculated on the five trait dimensions of extraversion, agreeableness, conscientiousness, neuroticism, and openness. As shown in Fig. 3, participants with SHS scored significantly higher on neuroticism than those in the healthy group; the same participants had lower scores on extraversion, agreeableness, and conscientiousness than healthy participants. The groups’ scores on openness were not significantly different.

Fig. 3.

Box plot of BFI-10 scores of healthy and SHS participants. The data were expressed as minimum, P25, median, P75, and maximum. BFI-10, Big Five Inventory-10 items; SHS, suboptimal health status

Participants’ perceived stress

Table 2 indicates that men with SHS had scores of 11.40 ± 2.12 on the PSS-4, significantly higher than for healthy men (9.83 ± 2.58). Similarly, among women, PSS-4 scores were significantly high for those with SHS. Participants with SHS therefore had stronger perceived stress than healthy participants.

Table 2.

Stratified analysis of PSS-4 among healthy and SHS groups

Groups		n	PSS-4
Men	Health	4971	9.83 ± 2.58
	SHS	4870	11.40 ± 2.12*
Women	Health	5818	9.86 ± 2.52
	SHS	7238	11.40 ± 2.34*
Total		22,897	10.67 ± 2.52

Data of continuous variables were expressed as mean ± standard deviation

PSS-4, Perceived Stress Scale-4 items; SHS, suboptimal health status; *P < 0.001 compared with healthy group

Correlation analyses

As presented in Table 3, three traits (i.e., extraversion, agreeableness, and conscientiousness) were negatively correlated with SHS, whereas neuroticism was positively correlated with it. Extraversion, agreeableness, conscientiousness, and openness were negatively associated with perceived stress; neuroticism was positively associated with it. A significant positive relationship was also observed between perceived stress and SHS.

Table 3.

Means, standard deviations, ranges, and inter-correlations of study variables

Variables	M	SD	Range	Skewness (SE)	Kurtosis (SE)	1	2	3	4	5	6	7
1. BFI-10-Ext	6.25	1.67	2–10	0.102 (0.016)	0.112 (0.032)	1
2. BFI-10-Agr	6.98	1.48	2–10	0.035 (0.016)	− 0.122 (0.032)	0.003	1
3. BFI-10-Con	6.59	1.65	2–10	0.211 (0.016)	− 0.156 (0.032)	0.189**	0.260**	1
4. BFI-10-Neu	5.85	1.57	2–10	− 0.093 (0.016)	0.436 (0.032)	− 0.189**	− 0.234**	− 0.219**	1
5. BFI-10-Ope	6.64	1.56	2–10	0.169 (0.016)	0.002 (0.032)	0.204**	0.078**	0.024**	− 0.009	1
6. PSS-4	10.67	2.52	4–20	− 0.385 (0.016)	0.422 (0.032)	− 0.226**	− 0.268**	− 0.348**	0.348**	− 0.096**	1
7. SHSQ-SF	11.60	7.52	0–36	0.619 (0.016)	0.378 (0.032)	− 0.107**	− 0.172**	− 0.233**	0.253**	− 0.008	0.345**	1

Pearson’s correlation was used to examine correlations among personality traits, stress, and SHS. An inspection of skewness and kurtosis values suggested that all measurements were normally distributed (being included in the conventional cut off of ± 3)

BFI-10-Agr, Big Five Inventory-10-Agreeableness; BFI-10-Com, Big Five Inventory-10-Conscientiousness; BFI-10-Ext, Big Five Inventory-10-Extraversion; BFI-10-Neu, Big Five Inventory-10-Neuroticism; BFI-10-Ope, Big Five Inventory-10-Openness; M, mean; PSS-4, Perceived Stress Scale-4 items; SD, standard deviation; SE; standard error; SHSQ-SF, Short-Form Suboptimal Health Status Questionnaire; *P < 0.05; **P < 0.01 (two-tailed)

Mediation analyses of stress on personality traits and SHS

Table 4 reveals that extraversion had a direct effect on stress (β =  − 0.126), while stress had a direct impact on SHS (β = 0.250). No significant direct effect manifested between extraversion and SHS. Stress thus fully mediated this relationship.

Table 4.

SEM pathways of big five personality traits, stress and SHS

Model pathways		Crude coefficient	95% CI	P	Adjust coefficient	95%CI	P
Extraversion
Direct effects	Ext → SHS	− 0.011	[− 0.023, 0.001]	0.068	− 0.011	[− 0.023, 0.001]	0.080
Indirect effects	Ext → Stress	− 0.126	[− 0.140, − 0.113]	< 0.001	− 0.126	[− 0.140, − 0.113]	< 0.001
	Stress → SHS	0.250	[0.237, 0.262]	< 0.001	0.252	[0.239, 0.266]	< 0.001
	Indirect effect/total effect	1			1
Agreeableness
Direct effects	Agr → SHS	− 0.049	[− 0.063, − 0.037]	< 0.001	− 0.052	[− 0.065, − 0.040]	< 0.001
Indirect effects	Agr → Stress	− 0.147	[− 0.160, − 0.133]	< 0.001	− 0.145	[− 0.158, − 0.131]	< 0.001
	Stress → SHS	0.250	[0.237, 0.262]	< 0.001	0.252	[0.239, 0.266]	< 0.001
	Indirect effect/total effect	0.429			0.413
Conscientiousness
Direct effects	Con → SHS	− 0.103	[− 0.117, − 0.091]	< 0.001	− 0.098	[− 0.112, − 0.084]	< 0.001
Indirect effects	Con → Stress	− 0.233	[− 0.246, − 0.219]	< 0.001	− 0.218	[− 0.232, − 0.204]	< 0.001
	Stress → SHS	0.250	[0.237, 0.262]	< 0.001	0.252	[0.239, 0.266]	< 0.001
	Indirect effect/total effect	0.361			0.359
Neuroticism
Direct effects	Neu → SHS	0.130	[0.117, 0.143]	< 0.001	0.123	[0.111, 0.136]	< 0.001
Indirect effects	Neu → Stress	0.239	[0.225, 0.253]	< 0.001	0.235	[0.222, 0.249]	< 0.001
	Stress → SHS	0.250	[0.237, 0.262]	< 0.001	0.252	[0.239, 0.266]	< 0.001
	Indirect effect/total effect	0.315			0.325
Openness
Direct effects	Ope → SHS	0.026	[0.015, 0.038]	< 0.001	0.014	[0.001, 0.026]	0.030
Indirect effects	Ope → Stress	− 0.051	[− 0.064, − 0.037]	< 0.001	− 0.056	[− 0.069, − 0.044]	< 0.001
	Stress → SHS	0.250	[0.237, 0.262]	< 0.001	0.252	[0.239, 0.266]	< 0.001
	Indirect effect/total effect	-			-

The validity was determined based on confirmatory factor analysis. The results for the hypothesized model showed good fit indices overall: root mean square error of approximation = 0.039 < 0.080, comparative fit index = 0.991 > 0.950, Tucker–Lewis index = 0.954 > 0.950. All the coefficients were standardized

Agr, Agreeableness; CI, confidence interval; Con, Conscientiousness; Ext, Extraversion; Neu, Neuroticism; Ope, Openness; SHS, suboptimal health status

Similarly, agreeableness had a direct effect on stress (β =  − 0.147), while stress had a direct impact on SHS (β = 0.250). Agreeableness had a significant direct effect on SHS (β =  − 0.049). Stress mediated roughly 42.9% of the total effect of agreeableness on SHS.

Furthermore, conscientiousness had a direct impact on SHS (β =  − 0.103). The significant mediating effect of stress accounted for 36.1% of the total effect of conscientiousness on SHS.

Neuroticism had a significant direct effect on SHS (β = 0.130). Mediation analyses showed that stress mediated 31.5% of the total effect of neuroticism on SHS.

Openness had a direct effect on SHS (β = 0.026), and its indirect effect on SHS was significant (β =  − 0.013). In other words, stress alleviated or masked the direct impact of openness on SHS. People who scored high on openness were less likely than others to perceive stress, thereby reducing the risk of SHS. Detailed results are depicted in Fig. 4.

Fig. 4.

The mediating analysis of perceived stress on the relationship between personality traits and SHS. The effects of covariates (i.e., gender, age, BMI, educational level, current residence, marital status, and occupational status) were controlled in the SEM analysis. All the coefficients in the figure were standardized. Model fit indices: RMSEA = 0.039, CFI = 0.991, TLI = 0.954. BMI, body mass index; CFI, comparative fit index; RMSEA, root mean square error of approximation; SEM, structural equation modeling; SHS, suboptimal health status; TLI, Tucker-Lewis index; ***P < 0.001

Covariate analysis

After adjusting for the effects of sociodemographic covariates, SEM confirmed the adjusted mediation model’s robustness. Table 4 indicates that the five trait dimensions had significant direct effects on stress, with stress exerting a significant influence on SHS. In addition, four dimensions had significant direct effects on SHS in the model of agreeableness (β =  − 0.052), conscientiousness (β =  − 0.098), neuroticism (β = 0.123), and openness (β = 0.014). Traits also had indirect impacts on SHS via perceived stress, respectively, contributing 100.0%, 41.3%, 35.9%, and 32.5% to the total effects of extraversion, agreeableness, conscientiousness, and neuroticism on SHS.

Covariate analysis highlighted age group as a determinant of both stress and SHS. Compared with participants > 45 years, other groups had higher stress (β = 0.037 to 0.081, P < 0.001) and less SHS (β =  − 0.059 to − 0.026, P < 0.001). Normal BMI reduced the extent of stress (β =  − 0.020, P = 0.016) and had preventive effects on SHS (β =  − 0.045, P < 0.001). Moreover, being divorced or widowed was negatively related to both higher SHS (β = 0.018, P = 0.013) and higher stress (β = 0.015, P = 0.017) (Table 5). In addition, we performed a subgroup analysis on young individuals aged 16 to 18. These stratified results indicated that perceived stress play a key mediating role in the relationship between personality (e.g., neuroticism) and SHS (Tables S4, S5, S6, and S7).

Table 5.

SEM results of covariates on the SHS model

Covariates		SHS			Stress
	β	95% CI	P	β	95% CI	P
Gender (reference: male)	0.029	[0.017, 0.042]	< 0.001	0.004	[− 0.007, 0.016]	0.475
Age (reference: ≥ 45)
≤ 24	− 0.058	[− 0.092, − 0.023]	0.001	0.081	[0.052, 0.110]	< 0.001
25–	− 0.059	[− 0.076, − 0.042]	< 0.001	0.064	[0.047, 0.081]	< 0.001
35–	− 0.026	[− 0.041, − 0.013]	< 0.001	0.037	[0.023, 0.052]	< 0.001
BMI categories (reference: thin)
Normal	− 0.045	[− 0.064, − 0.027]	< 0.001	− 0.020	[− 0.038, − 0.004]	0.016
Overweight	− 0.021	[− 0.038, − 0.004]	0.020	− 0.016	[− 0.032, 0.001]	0.062
Obese	0.001	[− 0.013, 0.015]	0.860	− 0.004	[− 0.018, 0.010]	0.602
Educational level (reference: middle school and below)
Senior high school	0.040	[0.021, 0.056]	< 0.001	0.017	[− 0.001, 0.033]	0.054
College	0.096	[0.077, 0.115]	< 0.001	− 0.001	[− 0.019, 0.018]	0.924
Master and above	0.062	[0.048, 0.077]	< 0.001	− 0.022	[− 0.035, − 0.009]	0.001
Place of residence (reference: rural)	− 0.010	[− 0.023, 0.003]	0.116	− 0.033	[− 0.044, − 0.022]	< 0.001
Marital status (reference: single)
Married	− 0.037	[− 0.063, − 0.011]	0.006	− 0.017	[− 0.039, 0.005]	0.133
Divorced/widowed	0.018	[0.004, 0.032]	0.013	0.015	[0.003, 0.027]	0.017
Occupational status (reference: retired)
Student	− 0.046	[− 0.099, 0.007]	0.085	− 0.040	[− 0.084, 0.008]	0.090
Employed	− 0.043	[− 0.089, 0.003]	0.067	− 0.005	[− 0.046, 0.036]	0.813
Unemployment	− 0.039	[− 0.063, − 0.015]	0.002	0.012	[− 0.009, 0.034]	0.295

The effects of covariates (i.e., gender, age, BMI, educational level, current residence, marital status, and occupational status) were controlled in the SEM analysis

BMI, body mass index; CI, confidence interval; SEM, structural equation modeling; SHS, suboptimal health status

Discussion

This nationwide cross-sectional study illustrated that approximately 52.9% of participants reported SHS complaints among a 22,897-person sample. Two dimensions of personality (i.e., agreeableness and conscientiousness) might protect against SHS onset. Chinese individuals with neuroticism were more likely than others to develop SHS. Additionally, perceived stress played a key mediating role in the relationship between personality and SHS.

Studies on SHS prevalence commonly yield inconsistent results due to heterogeneity in sampling and SHS measurements. More than five SHS scales have been included in Chinese investigations, such as the Suboptimal Health Status Questionnaire-25 (SHSQ-25), the Multidimensional Sub-health Questionnaire of Adolescents, the Sub-health Measurement Scale Version 1.0, the Chinese Sub-health State Evaluation Scale, and the Sub-Health Self-Rating Scale [35, 36]. Of these, the SHSQ-25 has frequently been used with East Asian, African, Oceanian, and European populations.

Two studies involving the SHSQ-25 revealed lower SHS in college students (21.0%) and a higher rate in the general adult population (69.5%) in China [37, 38]. We observed a prevalence of 52.9% across 148 Chinese cities. Notably, we used a simplified form of the SHSQ-25. This choice may introduce methodological heterogeneity across our studies even though the SHSQ-SF has been validated in northern and southern populations in China. Moreover, our sample contained many younger participants and college students, which could lead to underestimation compared with measurements of middle-aged and elderly populations [37].

Psychological factors are crucial in SHS development, and mental symptoms constitute a major factor to assess [13]. Depression and anxiety have each been shown to influence college students’ health [38]. Personality is another core contributor to quality of life and health [39]. As for how personality traits directly affect SHS, our results suggest that people with higher agreeableness and conscientiousness are more likely to experience optimal health. This finding is consistent with a previous survey using a diverse sample, which underlined agreeableness and conscientiousness as the best predictors of healthy behavior [40].

Agreeableness is also negatively related to risk-taking behavior [40]. Given that agreeableness has an inverse association with hostility [41], this outcome aligns with earlier research linking hostility to negative health habits [40, 42]. Adverse lifestyle patterns can increase the risk of somatic diseases among people with hostility, implying that positive health habits decrease the risk of SHS among people with higher agreeableness. In addition, conscientiousness improves health status through productive behavior management (i.e., maintaining healthy behavior patterns and avoiding harmful ones) [43].

Our results demonstrated that participants with neuroticism were more likely than others to develop SHS, echoing prior findings that emotional stability is a main contributor to perceived health [39, 44]. Individuals with a tendency toward neuroticism (i.e., low emotional stability) experience more negative affect and poorer emotional functioning [45], which can produce perceptions of unhealthy symptoms and negative outcomes [39]. These people also tend to display adaptive dysfunction when perceiving stress [39]. This propensity leaves them vulnerable to poor health.

Extraversion had no direct impact on SHS in our sample. Although some studies have uncovered a significant relationship between extraversion and quality of life in adults [46], those involving pediatric samples have reported either non-significant or weak results [44, 47]. People displaying openness are inclined to think and behave in nonconforming ways, which may generate diverse responses to health issues. The relationship between openness and perceived health has not been confirmed [48, 49]. However, we observed that people with an open personality had a higher prevalence of SHS.

One’s reactivity to psychosocial stressors could predict their susceptibility to mental and physical disorders associated with chronic stress (e.g., cardiovascular and infectious diseases). Personality is a primary psychological factor associated with stress resilience and health outcomes. Understanding how personality traits influence stress responses may shed light on individual differences in susceptibility to chronic stress-linked diseases. Studies have suggested that personality can modulate responses to perceived stress due to traits’ varying impacts on components of stress responses [50–52]. Stress exposure upregulates serum endothelin-1 (ET-1), which modulates a broad spectrum of processes, including regulation of physical and mental well-being, senses, pain, drug sensitivity, and healing processes. Besides, overproduction of ET-1, induced by imbalanced stress conditions, influences the development and progression of both communicable and non-communicable diseases [53]. Positive traits (e.g., extraversion, agreeableness, conscientiousness, and openness) largely protect against hormonal responses, whereas negative traits (e.g., neuroticism) inversely predict cardiovascular responses [52]. In other words, people with high extraversion or openness may be more resilient to negative environmental stimuli and stress-related symptoms. For agreeableness and conscientiousness, the tendency to maintain close interpersonal relationships may protect against excessive stress responses. In addition, together with our finding that perceived stress can positively predict SHS, high neuroticism may put people at elevated risk of chronic stress-related diseases.

Moreover, we identified a mediating effect of perceived stress on the relationships among personality traits and SHS. A mature personality (i.e., extraversion, agreeableness, conscientiousness, or openness) functions as an internal resource that helps people cope with stressors and other forms of psychological distress, thus promoting their health status. Conversely, neuroticism may expose people to poor stress adjustment and emotional functioning; adverse health could result.

Women are more likely than men to experience psychological distress and neuropsychiatric disorders [27, 54]. Our stratified analysis also revealed higher SHS among women, indicating that this group is more prone to mental problems and SHS due to distinct physiological and psychological characteristics [55]. Differences in men’s and women’s lifestyles, and in their pursuit of physical activities, are greatly related to their mental and physical well-being [56, 57].

Regarding covariates, we discovered that age is an important determinant of SHS and perceived stress: older participants experienced more SHS and less stress than younger ones. Consistent with research conducted with another Chinese resident sample [58], a gradual decrease in elderly people’s physical fitness and energy seems to be to blame. In terms of marital status, being divorced or widowed significantly contributed to SHS and stress. This outcome substantiates the associations of poor marital status with psychological distress and poor health status [59]. Regarding BMI classifications, the risk of SHS and stress appeared lower among normal-weight participants than among thin participants in this study. Lower BMI has negative effects on health status, such as eating disorders (e.g., anorexia nervosa), mental disorders, compromised immunity, impaired healing, cardiovascular abnormalities, and reproductive dysfunction [60]. This is consistent with the investigation on Flammer syndrome (FS), a specific manifestation characterized by low body weight and abnormal reactions toward any kind of physical (e.g., cold provocation) and mental stress. Exploring determinants of FS has been understood of great clinical utility for risk assessment of SHS [61].

Exploring the mechanisms underlying the health-to-disease transition is crucial for disease prevention and health promotion. SHS development shares several psychological risk factors, e.g., specific personality traits and psychological stress. When a person suffers from prolonged perceived stress, it reduces quality of life even though no specific disease can be diagnosed. Studies have reported that continuous stress could negatively impact the function of eye and brain due to autonomous nervous system imbalance and vascular dysregulation [62]. There are diverse manners for persons to cope with stress. If a person has no sufficient resilience capacities or skills to cope with stress, he may experience perceived fatigue, burnout, anxiety, and/or depression [62], which can thereby lead to more somatic and mental disorders.

To summarize, considering that perceived stress plays diverse mediating role in SHS development among individuals with different personalities, relaxation, psychotherapy, or other stress reduction programs should be helpful in reducing the progression of SHS. Mental relaxation, e.g., meditation, breathing exercises, and autogenic training, have been considered significant modulators of somatic and mental health [63]. Besides, it is well known that adaptation through coping is a psychological defense mechanism, especially in individuals with neuroticism. Sufficient resilience capacities and adaptive skills are crucial to coping with stress, which contribute to health enhancement and chronic disease prevention.

However, our findings could not reflect the nature of people in other countries due to ethnicity, lifestyles, and sociocultural factors. In addition, the discrepancies in situation of information transparency and potential bias of selective reporting might also influence the results. Therefore, more surveys are needed to be conducted in more countries with different ethnicities and sociocultural environment for practical PPPM/3PM implementation.

Limitations

This study has several limitations common in population-based cross-sectional surveys. First, we could not demonstrate causal relationships among personality traits, stress, and SHS due to a lack of chronological evidence. Second, since we excluded the persons with somatic diseases and psychiatric abnormalities, a large proportion of the eligible participants were young people who cannot represent the age composition of Chinese population. This might lead to selection bias. Third, all data were obtained through self-report questionnaires, which may have led to information bias. Fourth, although stress and SHS were assessed using standardized questionnaires, these measures are not equivalent to clinical diagnoses. Future research should therefore include diagnostic interviews.

To the best of our knowledge, this study is one of the first to delineate the relationships among personality traits, perceived stress, and SHS in representative Chinese populations. We covered a broad age range and considered multiple minorities and regions using data from a large-scale cross-sectional survey.

Conclusion and expert recommendations in the framework of 3P medicine

Our findings, on the basis of a nationwide survey, determined that 52.9% of participants experience SHS in China. Personality traits and perceived stress are psychological determinants of SHS and may be intervention targets for promoting health status in Chinese population. Specifically, perceived stress plays a mediating role in the relationships between personality traits and SHS.

Screening for SHS and its relevant factors provides a window of opportunity for early prevention and targeted intervention of NCDs in the PPPM/3PM context. Adverse psychological conditions, as a main aspect of SHS, represent a target for identifying people at high risk of chronic disease. Under the PPPM/3PM framework, screening individuals with SHS-related personality traits (e.g., neuroticism) and targeted intervention will contribute to decrease in SHS prevalence, thereby reducing the incidence of SHS-induced chronic diseases.

Predictive approach

Early screening of individuals at high risk of SHS is important for health enhancement under PPPM/3PM [64]. Mental-associated risk factors are contributors to both suboptimal health status and NCDs. With regard to perceived stress, this study evaluated its mediating role in the development of SHS attributed to personality. Our study concluded that individuals with neuroticism were susceptible to SHS due to unfavorable coping measures in the face of stress. Hence, the findings of this study provided novel target for early screening of persons at high risk for NCDs.

Targeted prevention

As a subclinical stage preceding NCDs, early detection of SHS is conducive to the timely prevention of chronic diseases. Modifiable risks that are relevant for health conditions have to be explored in low- and middle-income regions, such as personality traits and psychosocial stressors. Moreover, corresponding risk-mitigating measures are recommended for targeted prevention including innovative screening programs toward individuals with different sociocultural background. We demonstrated that perceived stress partially mediates the relationships among agreeableness, conscientiousness, neuroticism, and SHS, and suggest sufficient approaches to preventing SHS among people with these traits, especially neuroticism. Stress alleviation could reduce the progression of SHS and prevent the development of chronic diseases, particularly for those with neurotic personality. In the framework of PPPM/3PM, policies and strategies are needed urgently to improve individuals’ mental health condition in China.

Personalization of medicine services

Conventional risk factors, such as unhealthy lifestyle, depression, and anxiety, cannot fully clarify all SHS occurrence. Considering that interventions should be tailored to one’s unique risk profile, this study recommend tailoring psychological intervention to individuals in their personal circumstances. Moreover, it is needed to take corresponding stress-coping measures for individuals with specific personality traits susceptible to SHS, especially for persons with neuroticism. This course of action will hopefully reduce delayed intervention, generic prevention, and ineffective treatment.

Paradigm shifts from reactive medicine to PPPM/3PM and moving beyond the state of the art

NCDs are usually treated after disease onset, representing relatively delayed management. SHS is a preclinical and reversible stage before NCDs. Early screening of SHS from people with high-risk psychological traits opened a window of opportunity for implementing effective management of preventable risk factors from the prospective of PPPM/3PM. Translational research targeting to psychological traits may be of great clinical utility to develop innovative approaches benefiting early prediction and precision treatment [11, 61]. Therefore, evaluation of mental health condition is highly recommended as a novel approach for prevention of SHS in the framework of PPPM/3PM. To shift from reactive medicine to PPPM/3PM, scholars should investigate the determinants of SHS and develop new strategic models.

Supplementary Information

Below is the link to the electronic supplementary material.

Abbreviations

Agr

Agreeableness

BFI-10-Agr

Big Five Inventory-10-Agreeableness

BFI-10

Big Five Inventory-10

BFI-10-Com

Big Five Inventory-10-Conscientiousness

BFI-10-Ext

Big Five Inventory-10-Extraversion

BFI-10-Neu

Big Five Inventory-10-Neuroticism

BFI-10-Ope

Big Five Inventory-10-Openness

BMI

Body mass index

CFA

Confirmatory factor analysis

CI

Confidence interval

Con

Conscientiousness

CSHES

Chinese Sub-health State Evaluation Scale

CVD

Cardiovascular diseases

Ext

Extraversion

M

Mean

MSQA

Multidimensional Sub-health Questionnaire of Adolescents

NCDs

Non-communicable diseases

Neu

Neuroticism

Ope

Openness

PPPM/3PM

Predictive, preventive, and personalized medicine

PSS-4

Perceived Stress Scale-4 items

SD

Standard deviation

SEM

Structural equation model

SHMS V1.0

Sub-health Measurement Scale Version 1.0

SHSQ-SF

Short-Form Suboptimal Health Status Questionnaire

SHSQ-25

Suboptimal Health Status Questionnaire-25

SHS

Suboptimal health status

SSS

Sub-Health Self-Rating Scale

T2DM

Type 2 diabetes mellitus

Author contribution

Haifeng Hou conceived the study and guided the development of research and the preparation of manuscripts. Qihua Guan, Hualei Dong, Zhihui Zhang, Zheng Guo, Zi Lin, and Hui Niu performed the material preparation and data collection. Qihua Guan, Hualei Dong, and Zhihui Zhang researched data, performed the statistical analyses, and wrote the manuscript. Haifeng Hou provided critical expert advice or critical review of the current manuscript. All authors read and approved the final manuscript.

Funding

This work was supported by Natural Science Foundation of Shandong Province (ZR2022MH082).

Data availability

The data are available from the corresponding authors on a reasonable request.

Code availability

Not applicable.

Declarations

Ethics approval

The current study was approved by the Ethics Research Committee (JKWH-2022–02).

Consent to participate

The study was conducted in accordance with the Declaration of Helsinki of the World Medical Association and written informed consent was obtained from all the participants.

Consent for publication

All authors have given consent for publication.

Competing interests

The authors declare no competing interests.