Introduction

Hypertension is a major public health concern and is estimated to affect 1.13 billion people worldwide,[1] including >500 million women.[2] Despite considerable progress in antihypertensive medication treatments, hypertension is still a leading cause of cardiovascular disease (CVD) and mortality,[3-6] accounting for more than 1.16 million deaths among women in 2019.[3] Identifying risk factors associated with hypertension in women is critical for female hypertensive prevention.

Psychological stress may contribute to the development of hypertension according to the clinical practice guidelines for high blood pressure in the United States.[7] Stresses were shown to be associated with hypertension through direct or indirect pathways, such as impacting endothelial dysfunction,[8] salt consumption,[9] and addictive behaviors.[10] However, the relationship between specific chronic stressors of females, such as work demands or life events, and hypertension has not been concluded.[11–13] Besides, gender differences are not taken into account in previous studies as the work-related stress of women was hardly discriminated from life-related stress because of unpaid domestic work and care work, which might be underestimated in studies ignoring unpaid work.[14] Moreover, few studies concentrate on the financial-related stress of females.[12] In addition, although marital life is considered a risk factor for stress and cardiovascular diseases in the female population,[14] the impact of the spouse’s stress status on the development of hypertension among females has not been investigated. Hence, this cross-sectional study aimed to evaluate the associations between perceived psychological stress and hypertension with a composite measure of psychological stress factors among unpregnant married women aged 20–49 years, especially from the perspective of a family environment.

Methods Study design and participants

Data for this cross-sectional study were based on a national cohort survey in China of reproductive-aged married couples receiving free pre-conception health examinations and follow-up until 6 months after giving birth, called National Free Pre-Pregnancy Checkups Project (NFPCP). The methods of the original research have been described in previous reports.[15–17] Supported by the National Health Commission and the Ministry of Finance of the People’s Republic of China, the NFPCP has served couples from 2907 counties/districts across 27 provinces or autonomous regions, four municipalities and Xinjiang Production and Construction Corps (XPCC) in Chinese mainland since 2014. This study was approved by the institutional research review board at the National Health and Family Planning Commission (No.2009-2). Written informed consent was obtained from all participants. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guidelines.

In the current study, a total of 12,133,024 couples who participated in a free pre-conception health examination of NFPCP from January 1, 2016, to December 31, 2017, were selected as the study population. Eligible participants were wives aged 20–49 years and husbands aged 20–59 years without missing values for blood pressure (BP) and psychological stress status. Participants were excluded if they had heart disease, chronic nephritis, or mental or psychological disorders. Finally, 10,027,644 couples were included in the analysis. Detailed information about the study population recruitment is shown in Figure 1.

Figure 1:

Selection of couples for exploring association between psychological stresses and female hypertension. DBP: Diastolic blood pressure; SBP: Systolic blood pressure.

Data collection

Face-to-face interviews and medical examinations were conducted by trained healthcare staff at local family planning service agencies or maternal and childcare service centers. A structured questionnaire was administered to assess sociodemographic characteristics (including nationality, occupation, and urban/rural inhabitants according to the address of household registration), hypertensive status, psychological stress, medical history, the year of marriage, smoking status, and alcohol consumption for both wives and husbands separately. Marriage duration was defined as the number of years between the year of marriage and the year of examination (2016/2017). Higher education level was defined as high school graduate or above. Age data were categorized into five groups (20–24 years, 25–29 years, 30–34 years, 35–39 years, and ≥40 years). Alcohol drinking was categorized as no daily drinking at the time of the pre-conception health examination, 1–100 mL/day, 101–200 mL/day, 201–300 mL/day, and ≥301 mL/day. Smoking was categorized as follows: participants who did not smoke at the time of the pre-conception health examination, and those who smoked 1–5 cigarettes/day, 6–10 cigarettes/day, 11–15 cigarettes/day, or ≥16 cigarettes/day.

Height, weight, and blood pressure were obtained from a physical examination conducted by trained healthcare staff, following the standard protocol entitled “Guidelines for Physical Examination of the NFPCP.” Body mass index (BMI) was calculated as weight in kilograms divided by height in square meters (kg/m2), and body weight and height were measured while participants wore light indoor clothes without shoes or other accessories. Underweight was defined as BMI <18.5 kg/m2, normal weight was defined as BMI ≥18.5 kg/m2 and BMI <24.0 kg/m2, overweight was defined as BMI ≥24.0 kg/m2 and BMI <28.0 kg/m2, and obesity was defined as BMI ≥28.0 kg/m2, in accordance with standard BMI classifications in China.[18] Fasting plasma glucose (FPG), which was obtained from the clinical examination, was determined using the glucose oxidase or hexokinase methods within 24 hours for each woman. Average salt consumption among provinces was classified into three groups according to per capita iodized salt consumption data released by China Public Health Statistical Yearbook 2019: less consumption (Gansu, Jiangxi, Tianjin, Chongqing, Hubei, Ningxia, Guizhou, Shanxi, Fujian, and Jilin), moderate consumption (Sichuan, Heilongjiang, Hunan, Henan, Hebei, Guangxi, Zhejiang, Shaanxi, Shandong, and Shanghai), and higher consumption (Hainan, Inner Mongolia, Anhui, Guangdong, Beijing, Jiangsu, Yunnan, Qinghai, Liaoning, Xinjiang, and Xizang).

All data were uploaded and transferred remotely and stored in the NFPCP medical service information system, which was supported by the National Research Institute for Family Planning.

Perceived psychological stress

Perceived psychological stress was defined in terms of three domains: work/life stress, economic stress, and overall stress. Husbands and wives were asked, “How much stress from work or life have you felt in recent times?” and “How much stress from financial issues have you felt in recent times?” at the pre-conception health examination. Perceived work/life stress or economic stress levels before wives’ pregnancy were measured using a single-item measure of stress symptoms[19] and a 5-point Likert-type scale.[20] To decrease the margin of error related to the small sample size of extremely stressed population, these variables were recoded into four categories as follows: 1 = “not at all,” 2 = “slightly,” 3 = “moderately,” and 4 = “a lot or extremely.” Perceived overall stress was reunified according to the higher categories of work/life stress and economic stress, which was also classified into four levels from “not at all” to “a lot or extremely.” To combine the psychological stress status of both spouses, individuals were further classified into two groups: stressed (for categories 2 to 4) and non-stressed (for category 1). Then, couples were divided into four groups: (1) neither wife nor husband was stressed, (2) only wife was stressed, (3) only husband was stressed, and (4) both were stressed.

Hypertension

Measurements of BP in the right arm of the participants were conducted by trained physicians using an automated BP monitor on a single occasion after participants rest for at least 10 minutes. Hypertension was defined as systolic BP/diastolic BP ≥140/90 mm Hg, or a self-reported history of hypertension.[21]

Statistical analyses

Continuous variables were described as mean values (standard deviations [SDs]) or medians (Q1, Q3). Categorical variables were summarized as numbers and percentages. The differences in demographic characteristics of normotensive and hypertensive participants were tested using an independent t-test or the Wilcoxon test for continuous variables, and the chi-squared test for categorical variables.

Age-adjusted and multivariable-adjusted logistic regression models were used to assess the associations between individual or combined stress of couples and the prevalent hypertension of wives (odds ratios [ORs] and corresponding 95% confidence intervals [95% CIs]). By treating a categorical variable as an ordinal variable in the regression models, we tested the linear trend. Covariates in the multivariable-adjusted logistic regression model were mainly comprised of wives’ characteristics, including age, nationality, education level, occupation, rural/urban inhabitants, hypertension history, BMI, FPG, smoking status, and drinking status.[7]

We examined whether the association between married couples’ stress status and hypertension of women differed in different subgroups of women, including age (<35 years, ≥35 years), nationality (Han/others), rural/urban inhabitants, BMI category (underweight, normal weight, overweight, and obesity), fasting plasma glucose (≤6.1 mmol/L, >6.1 mmol/L), and sodium intake (lower, medium, higher). To test the robustness of the main results, we calculated ORs in each study region of China, respectively, and then constructed a meta-analysis model to calculate the combined odds ratio as sensitivity analyses. I2 statistics were used to assess the between-province heterogeneity of ORs, and mixed-effects models were used to obtain the summarized OR estimates if I2 ≥50%, otherwise fixed-effects models were used. All analyses were performed using R software (version 3.5.0, https://www.r-project.org/). A two-sided P value of <0.05 was considered to be statistically significant.

Results

We collected data from 27 provinces or autonomous regions, four municipalities, and Xinjiang Production and Construction Corps (XPCC) [Supplementary Table 1, https://links.lww.com/CM9/B948]. A total of 10,027,644 women with a median age of 27 years (Q1, Q3: [24–32] years), and their spouses with a median age of 29 years (Q1, Q3: 26–34 years) were included in our analysis, in which 87.96% (8,820,399/10,027,644) of women were ethnically Han and 87.40% (8,764,381/10,027,644) of women resided in rural areas. Besides, 261,098 women were documented to have hypertension, and the prevalence of hypertension was 2.60%. Compared with normotensive women, those with hypertension were more likely to be older, located in urban areas, heavy drinkers, and heavy smokers, as well as had higher fasting plasma glucose and BMI. Besides, women with hypertension had a longer duration of marriage and a higher probability of exposure to their own or their spouses’ stress compared with those in the normotensive group. Women’s characteristics according to their hypertensive status are presented in Table 1.

Table 1 - Characteristics of participants who took part in the National Free Pre-Pregnancy Checkups Project according to wives’ hypertensive status. Variables Normotensive (N = 9,766,546) Hypertensive (N = 261,098) χ 2/W P value Age (years)* 27 (24, 32) 32 (27, 39) 8.5769e+11† <0.001 Age groups* 133385.56‡ <0.001 20–24 years 2,527,593 (25.88) 37,111 (14.21) 25–29 years 3,866,510 (39.59) 68,092 (26.08) 30–34 years 1,780,367 (18.23) 50,910 (19.50) 35–39 years 971,676 (9.95) 48,044 (18.40) ≥40 years 620,400 (6.35) 56,941 (21.81) Nationality* 13.32‡ 0.002 Han 8,591,260 (87.97) 229,139 (87.76) Others 904,144 (9.26) 24,719 (9.47) NA 271,142 (2.77) 7240 (2.77) Occupation* 23.03‡ <0.001 Farmer 6,367,262 (65.19) 166,115(63.62) Others 2,894,321(29.64) 77,127(29.54) NA 504,963 (5.17) 17,856 (6.84) Inhabitants* 1908.50‡ <0.001 Urban 1,218,583 (12.48) 39,978 (15.31) Rural 8,543,330 (87.48) 221,051 (84.66) NA 4633 (0.04) 69 (0.03) Education* 4973.34‡ <0.001 High school and above 3,696,275(37.85) 81,395(31.17) Primary school 5,638,514 (57.73) 165,621 (63.43) NA 431,757 (4.42) 14,082 (5.40) Marriage duration (years) 2 (0, 7) 6 (1, 14) 7.0173e+11† <0.001 BMI (kg/m2)* 21.48 (19.68, 23.44) 23.52 (21.22, 26.67) 8.2603e+11† <0.001 BMI groups* 142580.00‡ <0.001 <18.5 kg/m2 1,075,610 (11.01) 13,472 (5.16) 18.5–23.9 kg/m2 6,702,742 (68.63) 126,292(48.37) 24.0–27.9 kg/m2 1,562,841 (16.00) 76,436 (29.27) ≥28.0 kg/m2 407,517 (4.17) 44,170 (16.92) NA 17,836 (0.18) 728 (0.28) Drinker* 840.62‡ <0.001 0 mL/day 9,466,671 (96.93) 250,802 (96.06) 1–100 mL/day 68,749 (0.70) 2759 (1.06) 101–200 mL/day 3080 (0.03) 161 (0.06) 201–300 mL/day 1340 (0.01) 76 (0.03) ≥301 mL/day 1802 (0.02) 83 (0.03) NA 224,904(2.30) 7217(2.76) Smoker* 359.96‡ <0.001 0 cigarettes/day 9,714,598 (99.47) 259,104 (99.24) 1–5 cigarettes/day 16,734 (0.17) 617 (0.24) 6–10 cigarettes/day 9683 (0.10) 441 (0.17) 11–15 cigarettes/day 1254 (0.01) 65 (0.02) ≥16 cigarettes/day 3289 (0.03) 195 (0.07) NA 20,988 (0.21) 676 (0.26) FPG (mmol/L)* 4.80 (4.38, 5.20) 4.98 (4.51, 5.36) 9.0371e+11† <0.001 Wives’ overall stress 1511.86‡ <0.001 Not at all 8,042,550 (82.35) 210,358 (80.57) Slightly 945,805 (9.68) 24,611 (9.43) Moderately 707,924 (7.25) 23,396 (8.96) A lot or extremely 70,267 (0.72) 2733 (1.05) Wives’ work/life stress 1274.42‡ <0.001 Not at all 8,238,042 (84.35) 215,411 (82.50) Slightly 868,794 (8.90) 23,534 (9.01) Moderately 612,051 (6.27) 20,307 (7.78) A lot or extremely 47,659 (0.49) 1846 (0.71) Wives’ economic stress 1218.65‡ <0.001 Not at all 8,405,800 (86.07) 220,871 (84.59) Slightly 796,461 (8.15) 21,025 (8.05) Moderately 523,659 (5.36) 17,557 (6.72) A lot or extremely 40,626 (0.42) 1645 (0.63) Husbands’ overall stress 1114.38‡ <0.001 Not at all 7,775,634 (79.61) 204,998 (78.51) Slightly 1,075,259 (11.01) 26,822 (10.27) Moderately 807,283 (8.27) 25,493 (9.76) A lot or extremely 108,370 (1.11) 3785 (1.45) Husbands’ work/life stress 1151.54‡ <0.001 Not at all 8,008,097 (82.00) 212,067 (81.22) Slightly 988,843 (10.12) 24,074 (9.22) Moderately 693,124 (7.10) 22,267 (8.53) A lot or extremely 76,482 (0.78) 2690 (1.03) Husbands’ economic stress 998.15‡ <0.001 Not at all 8,091,239 (82.85) 214,398 (82.11) Slightly 967,653 (9.91) 23,818 (9.12) Moderately 637,272 (6.53) 20,425 (7.82) A lot or extremely 70,382 (0.72) 2457 (0.94)

Values are medians (Q1, Q3) or n (%). *The analyzed variables of age, nationality, occupation, rural/urban inhabitants, education level, BMI, FPG, drinking status, and smoking status were female. †χ2 values; ‡W values. Higher education means high school graduate or above. BMI: Body mass index; FPG: Fasting plamsa glucose; NA: Not available.

Regarding the three different chronic stress types, the prevalence of stress in husbands was higher than in wives (work/life-related stress: 1,807,480 [18.02%] vs. 1,574,191 [15.70%], χ2 = 4,134,520; economic stress: 1,722,007 [17.17%] vs. 1,400,973 [13.97%], χ2 = 4,013,232; overall stress: 2,047,012 [20.41%] vs. 1,774,736 [17.70%], χ2 = 4,328,672, all P <0.001). Compared with the non-stressed wives, the multivariable-adjusted ORs of slightly, moderately, and extremely overall stress for women’s hypertension were 1.13 (95% CI:1.11–1.15), 1.29 (95% CI:1.27–1.31), and 1.31 (95% CI:1.25–1.37), respectively; and compared with the non-stressed husbands, the multivariable-adjusted ORs of husbands’ slightly, moderately, and extremely overall stress were 1.04 (95% CI:1.02–1.05), 1.23 (95% CI:1.22–1.25), and 1.24 (95% CI: 1.20–1.29), respectively. Similar increasing trends were also observed among wives and husbands with work/life stress and economic stress (all P-values for trend <0.001, Table 2). After further adjustment for women’s stress, a decreased risk of women’s hypertension was observed when husbands had slight stress (OR: 0.95 [95% CI: 0.94–0.97]), slight work/life stress (OR: 0.89 [95% CI: 0.87–0.90), and slight economic stress (OR: 0.94 [95% CI: 0.93–0.96]) [Supplementary Table 2, https://links.lww.com/CM9/B948]. Moreover, even if women were not under stress, the risks of female hypertension were increased to 1.19 times (95% CI: 1.17–1.21), 1.24 times (95% CI: 1.22–1.27), and 1.19 times (95% CI: 1.17–1.22) if their husbands had overall stress, life/work-stress, and economical stress, respectively [Table 3].

Table 2 - Multivariable-adjusted logistic regression models and meta-analysis model for analysing the association between stress status and prevalent hypertension of female participants (N = 10,027,644). Stress groups Participants, n (%) Case, n (%)* Age-adjusted† Multivariable-adjusted‡ Meta-analysis§ Age-adjusted† Multivariate-adjusted‡ OR (95% CI) OR (95% CI) Pooled OR (95% CI) Pooled OR (95% CI) Overall stress of wives Not at all 8,252,908 (82.30) 210,358 (2.55) 1.00 1.00 1.00 1.00 Slightly 970,416 (9.68) 24,611 (2.54) 1.07 (1.06, 1.09) 1.13 (1.11, 1.15) 1.13 (0.97, 1.32) 1.14 (0.98, 1.34) Moderately 731,320 (7.29) 23,396 (3.20) 1.26 (1.24, 1.27) 1.29 (1.27, 1.31) 1.29 (1.13, 1.47) 1.22 (1.08, 1.38) A lot or extremely 73,000 (0.73) 2733 (3.74) 1.41 (1.35, 1.46) 1.31 (1.25, 1.37) 1.39 (1.18, 1.65) 1.19 (1.03, 1.37) P for trend <0.001 <0.001