Supplemental Digital Content is available in the text
Keywords: depression, hypertension, less-developed regions, nomogram, prevalence, risk factors
Abstract
Hypertensive patients commonly co-exist persistent depressive symptoms. However, these issues are not always identified, especially in primary health care, which may worsen the prognosis of hypertension. Therefore, the aim of this study was to determine the prevalence and risk factor of depression, and to develop risk nomogram of depression in hypertensive patients from primary health care Northwest China.
We used a stratified multistage random sampling method to obtain 1856 hypertensives subjects aged ≥18 years in Xinjiang between April and October 2019. The subjects were randomly divided into a training set (n = 1299) and a validation set (n = 557). Depression was evaluated by Hospital Anxiety and Depression Scale (HADS), with a cut-off score ≥8. Using the least absolute shrinkage and selection operator (LASSO) regression model, we identified optimized risk factors of depression in the training set, followed by the establishment of prediction nomogram. The discriminative ability, calibration, and clinical usefulness of nomogram were assessed. The results were verified by internal validation in validation set.
13.7% hypertensive subjects displayed depression. Seven independent risk factors of depression were identified and entered into the nomogram including age, region, ethnicity, marital status, physical activity, sleep quality, and control of hypertension. The nomogram displayed robust discrimination with an AUC of 0.760 [95% confidence interval (CI): 0.724–0.797)] and 0.761 (95%CI: 0.702–0.819), and good calibration in training set and validation set, respectively. The decision curve analysis and clinical impact curve demonstrated clinical usefulness of predictive nomogram.
There is a considerable prevalence of depression in patients with hypertension from primary care of Xinjiang, Northwest China. Our nomogram may help primary care providers assess the risk of depression in patients with hypertension.
1. Introduction
Globally, one third of adults exhibits hypertension, of whom 75% are distributed in low and middle income countries (LMICs),[1] where people have low awareness and treatment rates, limited medical resources, and consequently poor blood pressure control.[2] Therefore, the burden of cardiovascular diseases (CVD) is enormous there as well.[3–5] Previous studies indicate that depression is common in patients with hypertension,[6,7] which increases the risk of cardiovascular mortality and morbidity, if untreated.[8,9]
Accumulating hospital-based surveys indicate that the prevalence of depression in hypertension ranges from 4.8% to 62.8% in China,[10] and 20% to 30% in the Netherlands.[11,12] Nonetheless, the prevalence of depression in hypertensives is less reported in primary care settings, especially in less developed regions. In addition, depression remains largely under-recognized and under-treated at primary health care,[13] where >70% of hypertensive patients are managed in less developed region.
The co-existence of depression not only affects the quality of life, but also interferes with the treatment and prognosis of hypertension.[9,14] In fact, the United States Preventive Services Task Forces recommends screening patients with chronic diseases for depression in clinical practices that have systems in place to assure accurate diagnosis, treatment and follow-up.[15] However, it is difficult to identify depression in hypertensive patients at primary health care facilities due to lack of recognition. Furthermore, even if physicians do attempt to screen for mental disorders in hypertensive patients, primary health care facilities are faced with barriers such as overburdened work, insufficient human and material resources, and communication difficulties since physicians sometimes have to read aloud the self-reported depression scales to illiterate patients, which may lead to misunderstandings, misdiagnosis and/or inappropriate treatment.[16] Currently, the management of depression comorbid medical conditions in primary care settings has been developed and shown positive effects in developed countries.[17–19] Therefore, the availability of appropriate tools at primary health care could contribute to the quality of detection and management of depressive disorders in LIMCs, particularly in less developed regions.
Thus, we aimed to evaluate the prevalence of depression and identify the associated risk factors in hypertensive population from primary health care of Northwest China, and to develop a predictive nomogram to estimate the probability of depression in a given visit, according to 4 dimensions related to the demographic factors, socioeconomic status, health-related behaviors, and anthropometric value, in order to identify possible targets for population-based preventive interventions.
2. Methods
2.1. Site
The study was conducted at primary health care centers including the urban, agricultural and stock-raising settings in Emin County Northern Xinjiang, an underdeveloped region in China, with a total population of over 160,000. Previous survey showed local residents have higher prevalence of hypertension, and low awareness, treatment and control rates.[20] Thus, it is an ideal setting for the study about the status of depression in hypertensive population from the less-developed region.
2.2. Study population
This cross-sectional study was conducted between April to October 2019. We used a stratified-cluster sampling method (district/township-community/village-resident) to collect 6294 subjects aged ≥18 years with a response rate of 96.8% (6294/6500). The current analysis included the 1856 hypertensive patients, as in Figure 1.
Figure 1.
The flow chart of recruiting the surveyed subjects including screening, eligibility and Included.
Subjects were eligible if:
-
1.
Local inhabitants aged ≥18 years;
-
2.
they agree to participate and sign an informed consent form.
Patients with confirmed dementia/Alzheimer's disease and physical disabilities were excluded because they could not answer questions. The Ethics Committee at the People's Hospital of Xinjiang Uygur Autonomous Region approved the current study.
2.3. Data collection and measurement
Trained study staff collected data on demographic characteristics (such as age, sex, and ethnicity), socioeconomic status (occupation, education level, marital status, and family income per member), health-related behaviors (alcohol consumption, cigarette smoking, physical activity, and sleep quality), and hypertensive status (whether they had a previously diagnosed by a doctor? whether they were receiving anti-hypertensive drugs treated in the past 2 weeks?) using standardized questionnaires. Other information of comorbidities such as dyslipidemia, diabetes, cardiovascular disease (including stroke and coronary heart disease) were collected by staff as well.
2.4. Blood pressure (BP) and anthropometric variables
Each participant's BP records were measured using with the automatic sphygmomanometer OMRON HBP-1300 Professional Portable Blood Pressure Monitor (OMRON, Kyoto, Japan). All participants were measured 3 times for BP with a 30 seconds interval between each measurement. The mean value of the 3 measurements was used for analysis. Height, weight and waist circumference were also measured by trained investigators. Body mass index (BMI) was calculated as weight divided by the square of height (kg/m2).
2.5. Plasma glucose and lipid measurements
All subjects were fasting for ≥8 hours, and a 5-ml fasting blood sample was collected. Then, fasting plasma glucose (FPG), triglycerides (TG), total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) were tested at local primary care center using standard methods.
2.6. Hypertension
Hypertension is defined as systolic BP (SBP) ≥140 mm Hg, and/or diastolic BP (DBP) ≥90 mm Hg, and/or use of antihypertensive agents within 2 weeks.[21] Control of hypertension was defined as SBP <140 mm Hg and DBP<90 mm Hg.
2.7. Evaluation and definition of depression
Depression was evaluated by the Hospital Anxiety and Depression Scale (HADS). The HADS is a 14 items scale and is divided into 2 subscales directed at either depression (HADS-D) or anxiety (HADS-A). Each subject completed the Chinese version of the HADS questionnaire.[22] Depression was identified as a score of ≥8 in the HADS-D, yielding a sensitivity and specificity of approximately 80%.
2.8. Other variables and definitions
Living regions were divided into agricultural, stock-raising, and urban setting based upon subjects’ self report. Education level was categorized into: primary and lower, junior high, and senior high and higher. Occupation was defined as manual and intelligent. Family income per member was divided into 5 groups as ≤¥500/month, ¥501–1000/month, ¥1001–2000/month, ¥2001–3000/month, and >¥3000/month. Marital status was classified as single, married, or separated (divorced or widowed). Cigarette consumption was coded as yes/no. Alcohol intake was defined as yes/no. Physical activity was assessed by Global Physical Activity Questionnaire (GPAQ), which was divided into 3 categories: low physical activity, moderate, and high physical activity based on the WHO's (2010) physical activity recommendations.[23,24] Sleep quality was inquired using Pittsburgh sleep quality index (PSQI), which was classified into very good, fair good, fair bad, and very bad if PSQI score of 0–4, 5–9, 10–14 and ≥15, respectively.[25,26] The Body mass index (BMI) was further classified into normal, overweight and obesity if BMI is below 24.0 kg/m2, between 24.0 and 27.9 kg/m2, and ≥28.0 kg/m2, respectively.[27] Abdominal obesity was defined as WC ≥ 90 cm in men, and WC ≥85 cm in women. Comorbidity was defined as a combination of 2 or more diseases: dyslipidemia, diabetes, and CVD.
2.9. Statistical analysis
Descriptive analyses were conducted for all subjects between training set and validation set using SPSS 20.0 for Windows (SPSS Inc., Chicago, IL). All continuous variables were summarized as means ± standard deviations (M ± SD), and categorical variables were expressed as frequency (n) and proportions (%), and the results were compared using Student t test and the Chi-Squared test to detect the statistical significances, respectively.
The formulation and the assessment of nomogram is divided into 6 steps: First, we developed the nomogram by randomly selecting 70% of our subjects (n = 1299) to construct the model. We reserved the remaining 30% (n = 557) for validation. Second, we identified independent predictive features in the training set by nonzero coefficients in the least absolute shrinkage and selection operator (LASSO) regression model.[28,29] Third, multivariable logistic regression model of the training set was applied to construct a predicting nomogram based on the selected feature from the LASSO regression model,[30] with results presented as numbers (N), odds ratio (OR) with associated 95% confidence interval (95%CI), and corresponding P value. Fourth, we assessed performance of nomogram by the discrimination and calibration. The discrimination of nomogram was evaluated by the area under the receiver operating characteristic curve (AUC). The calibration was assessed by comparing the nomogram-predicted probability with the actual probability, which was visualized by calibration curves plot using 1000 bootstrap resamples procedures and tested by Hosmer-Lemeshow test.[31] Fifth, we conducted internal validation of the nomogram in the validation set. The AUC, and calibration curves and Hosmer-Lemeshow test of nomogram were assessed in the validation set. Finally, the decision curve analysis and clinical impact curve were conducted to determine the clinical usefulness of the nomogram in the validation set, which quantified the net benefits at different threshold probabilities in the study.[32] The nomogram and the bootstrap analysis were performed using the package of “rms” in R version 3.5.1. A P value <.05 was considered to indicate significance.
3. Results
3.1. Subject characteristics
The baseline characteristics of the subjects in the training and validation sets are summarized in Table 1 and Supplementary Table S1, http://links.lww.com/MD/F609. Totally, subjects with depression accounted for 13.7% (254/1856), with 13.8% (179/1299) and 13.5% (75/557) in the training and validation sets, respectively, with no significant difference (P = .883). In addition, the significant differences were unobserved in the baseline characteristics of the subjects between the training set and the validation set (P value range from .065 to 1.000).
Table 1.
Characteristics of the subjects in training and validation sets.
| Training set(N = 1299) | Validation set(N = 557) | Total(N = 1856) | P value | |
| Depression (n, %) | ||||
| No | 1120 (86.2) | 482 (86.5) | 1602 (86.3) | .883 |
| Yes | 179 (13.8) | 75 (13.5) | 254 (13.7) | |
| Age (years) | 53.42 ± 12.46 | 54.45 ± 12.45 | 53.7 ± 12.5 | .105 |
| 18–44 | 285 (21.9) | 114 (20.5) | 399 (21.5) | .335 |
| 45–59 | 627 (48.3) | 258 (46.3) | 885 (47.7) | |
| ≥60 | 387 (29.8) | 185 (33.2) | 572 (30.8) | |
| Gender (n, %) | ||||
| Women | 556 (42.8) | 247 (44.3) | 803 (43.3) | .540 |
| Men | 743 (57.2) | 310 (55.7) | 1053 (56.7) | |
| Regions (n, %) | ||||
| Agriculture | 756 (58.2) | 320 (57.5) | 1076 (58.0) | .800 |
| Stock-raising | 172 (13.2) | 70 (12.6) | 242 (13.0) | |
| Urban | 371 (28.6) | 167 (30.0) | 538 (29.0) | |
| Education attainment status (n, %) | ||||
| Primary and lower | 562 (43.3) | 251 (45.1) | 813 (43.8) | .232 |
| Junior high | 461 (35.5) | 207 (37.2) | 668 (36.0) | |
| Senior high and higher | 276 (21.2) | 99 (17.8) | 375 (20.2) | |
| Occupation (n, %) | ||||
| Manual | 975 (75.1) | 428 (76.8) | 1403 (75.6) | .443 |
| Intelligent | 324 (24.9) | 129 (23.2) | 453 (24.4) | |
| Ethnicity (n, %) | ||||
| Han | 635 (48.9) | 297 (53.3) | 932 (50.2) | .158 |
| Kazakh | 455 (35.0) | 171 (30.7) | 626 (33.7) | |
| Others | 209 (16.1) | 89 (16.0) | 298 (16.1) | |
| Family income per member | ||||
| ≤¥500/month | 316 (24.3) | 129 (23.2) | 445 (24.0) | .770 |
| ¥501–1000/month | 343 (26.4) | 150 (26.9) | 493 (26.6) | |
| ¥1001–2000/month | 340 (26.2) | 159 (28.5) | 499 (26.9) | |
| ¥2001–3000/month | 141 (10.9) | 59 (10.6) | 200 (10.8) | |
| >¥3000/month | 159 (12.2) | 60 (10.8) | 219 (11.8) | |
| Marital status (n, %) | ||||
| Single | 60 (4.6) | 18 (3.2) | 78 (4.2) | .090 |
| Married | 1058 (81.4) | 443 (79.5) | 1501 (80.9) | |
| Separated | 181 (13.9) | 96 (17.2) | 277 (14.9) | |
| Cigarette consumption (n, %) | ||||
| No | 777 (68.9) | 350 (62.8) | 1127 (60.7) | .233 |
| Yes | 522 (40.2) | 207 (37.2) | 729 (39.3) | |
| Alcohol intake (n, %) | ||||
| No | 1109 (85.4) | 494 (88.7) | 1603 (86.4) | .065 |
| Yes | 190 (14.6) | 63 (11.3) | 253 (13.6) | |
| Physical activity (n, %) | ||||
| Low | 225 (17.3) | 102 (18.3) | 327 (17.6) | .576 |
| Median | 727 (56.0) | 297 (53.3) | 1024 (55.2) | |
| High | 347 (26.7) | 158 (28.4) | 505 (27.2) | |
| PSQI score | 4.78 ± 3.78 | 5.02 ± 3.88 | 4.8 ± 3.8 | .216 |
| sleep quality (n, %) | ||||
| Very good | 867 (66.7) | 351 (63.0) | 1218 (65.6) | .458 |
| Fair good | 305 (23.5) | 142 (25.5) | 447 (24.1) | |
| Fair bad | 113 (8.7) | 57 (10.2) | 170 (9.2) | |
| Very bad | 14 (1.1) | 7 (1.3) | 21 (1.1) | |
| Body mass index (kg/m2) | 27.37 ± 4.51 | 27.34 ± 4.35 | 27.4 ± 4.5 | .889 |
| BMI: <23.9 kg/m2 | 285 (21.9) | 126 (22.6) | 411 (22.1) | .704 |
| BMI: 24.0–27.9 kg/m2 | 486 (37.4) | 197 (35.4) | 683 (36.8) | |
| BMI: ≥28.0 kg/m2 | 528 (40.6) | 234 (42.0) | 762 (41.1) | |
| waist circumference (cm) | 91.96 ± 11.88 | 92.21 ± 11.75 | 92.0 ± 11.8 | .683 |
| Abdominal obesity (n, %) | 818 (63.0) | 364 (65.4) | 1182 (63.7) | .343 |
| Comorbidity (n, %) | 129 (9.9) | 55 (9.9) | 184 (9.9) | 1.000 |
| Blood pressure (mm Hg) | ||||
| Systolic blood pressure | 150.78 ± 19.18 | 150.58 ± 20.45 | 150.7 ± 19.6 | .837 |
| Diastolic blood pressure | 91.17 ± 12.97 | 90.70 ± 13.81 | 91.0 ± 13.2 | .475 |
| FPG (mmol/L) | 5.92 ± 2.42 | 5.82 ± 2.28 | 5.89 ± 2.38 | .423 |
| TC (mmol/L) | 5.00 ± 1.06 | 5.01 ± 1.10 | 5.00 ± 1.07 | .850 |
| TG (mmol/L) | 1.69 ± 1.44 | 1.66 ± 1.18 | 1.68 ± 1.36 | .650 |
| LDL-c (mmol/L) | 2.33 ± 0.32 | 2.35 ± 0.45 | 2.33 ± 0.36 | .220 |
| HDL-c (mmol/L) | 1.56 ± 0.20 | 1.55 ± 0.14 | 1.56 ± 0.18 | .157 |
FPG = fasting plasma glucose, TC = total cholesterol, TG = triglycerides, LDL-C = low density lipoprotein cholesterol, HDL-C = high-density lipoprotein cholesterol, PSQI = Pittsburgh sleep quality index.
3.2. Independent risk factors for depression in hypertensive subjects
We used the LASSO regression model to screen independent risk factors of depression in the training set. Of demographic variables, socioeconomic status, health-related behaviors, and anthropometric measurments, 7 potential predictors were screened out of 38 factors in the study (∼5:1 ratio; Fig. 2 A and B) and were with nonzero coefficients in the LASSO regression model. These factors included age, living setting, ethnicity, marital status, physical activity, sleep quality, and control of hypertension (Table 2).
Figure 2.
Predictor selection from demographic variables, socioeconomic status, health-related behaviors, and anthropometric measurements using the LASSO binary logistic regression model. Notes: (A) Optimal parameter (lambda) selection in the LASSO model used fivefold cross-validation via minimum criteria. The partial likelihood deviance (binomial deviance) curve was plotted versus log (lambda). Dotted vertical lines were drawn at the optimal values by using the minimum criteria and the 1 SE of the minimum criteria (the 1-SE criteria). (B)LASSO coefficient profiles of the 38 features. A coefficient profile plot was produced against the log (lambda) sequence. Vertical line was drawn at the value selected using fivefold cross-validation, where optimal lambda resulted in seven features with nonzero coefficients. Abbreviations: LASSO, least absolute shrinkage and selection operator; SE, standard error.
Table 2.
Factors associated with depression from study population by multiple logistic regression.
| Stratification | N | Prevalence (95%CI) | AdjustedOR (95%CI) | P value |
| Age (count) | – | – | 1.02 (1.01–1.03) | .033 |
| 18–44 | 36/285 | 12.6 (9.0–17.1) | 1 (Reference) | |
| 45–59 | 71/627 | 11.3 (8.9–14.1) | 1.18 (0.74–1.88) | .490 |
| ≥60 | 72/387 | 18.6 (14.8–22.8) | 1.57 (1.03–2.65) | .043 |
| Regions (vs Urban) | 26/31 | 7.0 (4.6–10.1) | 1 (Reference) | |
| Stock-raising | 29/172 | 16.8 (11.6–23.3) | 1.55 (0.84–2.87) | .160 |
| Agriculture | 14/756 | 16.4 (13.8–19.2) | 2.25 (1.42–3.56) | .001 |
| Ethnicity (vs Han) | 60/635 | 9.4 (7.2–12.0) | 1 (Reference) | |
| Kazakh | 93/455 | 20.4 (16.8–24.4) | 3.25 (2.15–4.92) | <.001 |
| Others | 26/209 | 12.4 (8.3–17.7) | 1.62 (0.97–2.70) | .066 |
| Marital status (vs Married) | 107/1058 | 10.1 (8.4–12.1) | 1 (Reference) | |
| Single | 14/60 | 23.3 (13.4–36.0) | 3.21 (1.57–6.58) | .001 |
| Separated | 58/181 | 32.0 (25.3–39.4) | 3.82 (2.54–5.75) | <.001 |
| Physical activity (vs High) | 35/347 | 10.1 (7.1–13.7) | 1 (Reference) | |
| Median | 88/727 | 12.1 (9.8–14.6) | 1.30 (0.85–1.98) | .225 |
| Low | 56/225 | 24.9 (19.4–31.1) | 2.85 (1.75–4.65) | <.001 |
| Sleep quality (vs Very good) | 104/867 | 12.0 (9.9–14.3) | 1 (Reference) | |
| Fairly good | 44/305 | 14.4 (10.7–18.9) | 1.02 (0.68–1.54) | .911 |
| Fairly bad | 24/113 | 21.2 (14.1–29.9) | 1.44 (0.83–2.51) | .195 |
| Very bad | 7/14 | 50.0 (23.0–76.9) | 5.83 (1.78–19.12) | .004 |
| Control of hypertension (vs Yes) | 12/149 | 8.0 (4.2–13.6) | 1 (Reference) | |
| No | 167/1150 | 14.5 (12.5–16.7) | 2.19 (1.15–4.18) | .018 |
The above variables were identified by LOSSO regression; Multiple logistic regression adjusted the variables including: Age (count), region, ethnicity, marital status, physical activity, sleep quality, control of HT. HT = hypertension; OR = odd ratio; CI = confidence interval.
3.3. Development of a predictive nomogram
Based on the results of multivariable logistic regression analysis, which included the above independent predictors (Table 2), the model was built and presented as the nomogram (Fig. 3). We observed very bad sleep quality corresponded to the highest risk score of 100 points, second was separated and single marital status (75 points and 67 points, respectively), and followed by Kazakh (68 points), low physical activity (58 points), living in agricultural setting (46 points) and controlled hypertension (43 points).
Figure 3.
Proposed nomogram for depression in patients with hypertension. Nomogram predicting depression risk in hypertensive patients.
3.4. Performance and internal validation of the nomogram
The AUC for the predictive nomogram was 0.760 (95% CI: 0.724–0.797) in the training set, and was confirmed to be 0.761 (95% CI: 0.702–0.819) through internal validation in the validation set (Fig. 4), which indicated the model's good discrimination. The calibration of depressive nomogram demonstrated good agreement when performing the calibration curve plot in training set and validation set, respectively (Fig. 5A and B). The nomogram-predicted probability and actual probability are highly consistent by using Hosmer-Lemeshow test (training set, P = .960; validation set, P = .368).
Figure 4.
ROC curve of the nomogram. AUC: area under the curve.
Figure 5.
Calibration curves of the depressive nomogram prediction in the study. Notes: (A) Calibration curve of the depressive nomogram prediction in the training set. (B) Calibration curve of the depressive nomogram prediction in the validation set. The x-axis represents the predicted depressive risk. The y-axis represents the actual identified patients with depression. The diagonal dotted line represents a perfect prediction by an ideal model. The solid line represents the performance of the nomogram, of which a closer fit to the diagonal dotted line represents a better prediction.
3.5. Clinical use
The DCA for the depressive nomogram showed that when the threshold probabilities of depression in hypertensive patients ranged from 8% to 82%, the use of this nomogram to predict the depressive risk yielded more net benefit than the scheme, which indicated the nomogram to be clinically useful (Fig. 6).
Figure 6.
Decision curve analysis for the nomogram for predict depression. Notes: The y-axis measures the net benefit. The red line represents the depressive risk nomogram. The thin solid line represents the assumption that all patients are non-depression. Thin thick solid line represents the assumption that no patients are non-depression to medication. The decision curve showed that if the threshold probability of a patient and a doctor is ranged from 8% to 82%, using this nomogram for depression in the current study to predict depression risk adds more benefit than the intervention-all-patients scheme or the intervention-none scheme.
3.6. Clinical impact curve
The clinical effectiveness of the nomogram was demonstrated by the clinical impact curve, which predicted the probability stratification of 1000 subjects with the bootstrap technique. The clinical impact curve displayed the number of high-risk subjects (the number of positive cases predicted by the nomogram) and the number of high-risk subjects with events (the number of true-positive cases) under each threshold probability. The cost benefit ratios of the depressive nomogram was determined in clinical impact curve, as shown in Figure 7. For example, the dotted vertical line illustrates a tentative cut-off point (20% risk of depression) at which 340 of 1000 subjects (34%) would be filtered by the nomogram, with about 140 of these (41.2%) being true depression cases.
Figure 7.
The clinical impact curves of the nomogram predicting risk stratification of 1,000 people and showing the loss-benefit ratio. The red line, number high risk: under each threshold probability, the number of people classified by the nomogram as positive (high risk). The blue line, number high risk with outcome: the number of people with true positive under each threshold probability. The x-axis represents the net benefit; the y-axis represents the high risk (out of 1000) of depression in hypertensive patients. For example, the dotted vertical line illustrates a tentative cut-off point (20% risk of depression) at which 340 of 1000 subjects (34%) would be filtered by the nomogram, with about 140 of these (41.2%) being true depression cases.
4. Discussion
In this study, the prevalence of depression among hypertensive patients in an economically less developed region of Northwest China is 13.7%. Moreover, the presence of depression in hypertensive patients is associated with age, living setting, ethnicity, marital status, physical activity, sleep quality and uncontrolled hypertension. A nomogram was generated based on 7 variables to predict depression in hypertensive patients. Furthermore, the nomogram showed good discrimination, excellent calibration, and satisfactory validity and clinical utility, indicating good performance.
The current study reports that the prevalence of depression in patients with hypertension in primary care setting is 2 times greater than in a study in the Norwegian population, of whom 6.5% of hypertensives have depression using the same cut-off ≥8.[33] However, this is not consistent with the result of a recent meta- analysis of 41 studies.[10] Divergent results may partly be explained by study subjects from different regions, and various assessment tools and cut-off scores for depression. Additionally, previous studies show that the depression complicates the management of hypertension and is closely related to adverse health outcomes.[34–36] Therefore, primary health care physicians should be aware of the importance of early and regular assessment for depression among patients with hypertension.[37]
Results of the current study show that the presence of depression in hypertensive patients is higher in those aged ≥60 years, living in agricultural areas, Kazakh, with single and separated marital status, with low physical activity, with poor sleep quality, and in those who experience uncontrolled hypertension status, which are consistent with the previous study results.[38–41] The persons living in agricultural setting are mostly peasants, with poor medical access, little education and low income, especially in the elderly, compared with those living in urban region.[42] Therefore, they cannot receive a prompt diagnosis and treatment when suffering from hypertension. Poor health and mental status make them more prone to depression. Furthermore, hypertensive patients who are single and divorced/widowed often feel isolated and are less willing to communicate with other people, exacerbating the likelihood of depression.[41] Unfortunately, low physical activity and poor sleep quality also lead to depression, which corresponds to previous reports of low physical activity and poor sleep quality in hypertensive patients with depressive symptoms.[38,39] Moreover, depression can influence the efficacy of anti-hypertensive pharmaceutical treatment. A previous study found that depression can cause the dysfunction of the autonomic nervous systems hypothalamic-pituitary-adrenal axis, which can increase vascular tone and resistance to the control of blood pressure.[43] In turn, uncontrolled hypertension can aggravate the depressive symptoms of patients with hypertension.
Some studies have found the non-adherence to medication (contributing to untreated and uncontrolled hypertension) plays a significant role in the increased morbidity and mortality in hypertension and comorbid depression.[44] Therefore, it is very imperative to identify, detect and treat depression in hypertension. Our study is the first to develop and validate a nomogram for depression among hypertensive patients using 7 readily available variables, which might be useful in improving the patient outcomes with individualized risk prediction and interventions. This prediction tool has relatively accurate discrimination and calibration power, which can be widely and accurately used in primary health care among less developed region with similar populations and economies. And it may assist primary care providers in early identification of hypertensive patients at high risk for depression. Previous studies indicate that interventions such as health education, regular symptom monitoring, support of society and families, and regular follow-up could improve the treatment and prognosis of patients co-existed hypertension and depression.[18]
Current study harbors several strengths. First, we established a nomogram of predicting depression for hypertensive patients to make individualized screening possible for the first time. Second, our study contains relatively larger sample, comprehensive variables and strict data collection procedure, and wider age range, which may merit the data quality and generalizability. It makes the report one of the valuable information for public health sectors and clinical setting. Inevitably, this study has some limitations. First, the study sample is selected from one site of Xinjiang, China, which may hamper the representativeness of study results. Second, although the robustness of our nomogram was examined extensively with internal validation, we failed to conduct external validation. Therefore, the further study of the generalizability to hypertensive populations in other regions and countries is warranted. Finally, the control rate of hypertension in our sample population is 11%, significantly lower than that in high-income countries,[1] and accordingly, the extrapolation of this nomogram was limited to some extent.
5. Conclusions
There is considerable prevalence of depression in patients with hypertension at primary health care, Northwest China. Seven risk factors are associated depression in patients with hypertension. The nomogram developed may help primary care provider assess the risk of depression in hypertensives, and its external evaluation in wider hypertensive populations are warranted.
Acknowledgments
We thank all the individuals who participated in the present study. We thank The Ministry of Science and Technology of the People's Republic of China and The Department of Science and Technology of Xinjiang Uygur Autonomous Region of China for funding the project.
Author contributions
NL was involved in the study design. NL, LW, MH, ML, FP, ZY, ZW, and RA paticipated in the data collection. LW and MH designed the survey and performed the statistical analysis. LW drafted the manuscript. MH critically revised the manuscript. NL, ML, FP, ZY, ZW, and RA gave important suggestions and did significant changes. All authors reviewed and approved the final version of the paper. We also thank all the participants of the survey including the population.
Conceptualization: Lin Wang, Nanfang Li, Mulalibieke Heizhati.
Data curation: Lin Wang, Nanfang Li, Mulalibieke Heizhati, Fengyu Pan, Zhikang Yang, Zhongrong Wang, Reyila Abudereyimu.
Formal analysis: Lin Wang, Nanfang Li, Mulalibieke Heizhati.
Funding acquisition: Nanfang Li.
Investigation: Lin Wang, Mulalibieke Heizhati, Mei Li, Fengyu Pan, Zhikang Yang, Zhongrong Wang, Reyila Abudereyimu.
Methodology: Lin Wang, Mulalibieke Heizhati.
Resources: Lin Wang.
Software: Lin Wang.
Writing – original draft: Lin Wang.
Writing – review & editing: Mulalibieke Heizhati.
Glossary
Abbreviations: BMI = body mass index, BP = blood pressure, CI = confidence interval, CVD = cardiovascular diseases, DBP = diastolic blood pressure, DCA = decision curve analysis, FPG = fasting plasma glucose, GPAQ = Global Physical Activity Questionnaire, HADS = Hospital Anxiety and Depression Scale, HDL-C = high-density lipoprotein cholesterol, LASSO = least absolute shrinkage and selection operator, LDL-C = low-density lipoprotein cholesterol, OR = odds ratio, PSQI = Pittsburgh sleep quality index, SBP = systolic blood pressure, TC = total cholesterol, TG = triglycerides.
References
- [1].Mills KT, Stefanescu A, He J. The global epidemiology of hypertension. Nat Rev Nephrol 2020;16:223–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Ibrahim MM, Damasceno A. Hypertension in developing countries. Canadian J Cardiol 2014;30:527–33. [DOI] [PubMed] [Google Scholar]
- [3].GBD 2016 Causes of Death Collaborators. Global, regional, and national age-sex specific mortality for 264 causes of death, 1980-2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017;390:1151–210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Zhou M, Wang H, Zeng X, et al. Mortality, morbidity, and risk factors in China and its provinces, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. Lancet 2019;394:1145–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Mills KT, Bundy JD, Kelly TN, et al. Global disparities of hypertension prevalence and control: a systematic analysis of population-based studies from 90 countries. Circulation 2016;134:441–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Xue J, Chen S, Bogner HR, et al. The prevalence of depressive symptoms among older patients with hypertension in rural China. Int J Geriatr Psychiatry 2017;32:1411–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [7].Zhang J, Ye M, Huang H, et al. Depression of chronic medical inpatients in China. Arch Psychiatr Nurs 2008;22:39–49. [DOI] [PubMed] [Google Scholar]
- [8].Graham N, Ward J, Mackay D, et al. Impact of major depression on cardiovascular outcomes for individuals with hypertension: prospective survival analysis in UK Biobank. BMJ Open 2019;9:e024433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Axon RN, Zhao Y, Egede LE. Association of depressive symptoms with all-cause and ischemic heart disease mortality in adults with selfreported hypertension. Am J Hypertens 2010;23:30–7. [DOI] [PubMed] [Google Scholar]
- [10].Li Z, Li Y, Chen L, et al. Prevalence of depression in patients with hypertension: a systematic review and meta-analysis. Medicine (Baltimore) 2015;94:e1317. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].García-Fabela L, Melano-Carranza E, Aguilar-Navarro S, et al. Hypertension as a risk factor for developing depressive symptoms among community-dwelling elders. Rev Invest Clin 2009;61:274–80. [PMC free article] [PubMed] [Google Scholar]
- [12].Lobo-Escolar A, Roy JF, Saz P, et al. Association of hypertension with depression in community-dwelling elderly persons: results from the ZARADEMP project. Psychother Psychosom 2008;77:323–5. [DOI] [PubMed] [Google Scholar]
- [13].Lotfi L, Flyckt L, Krakau I, et al. Undetected depression in primary healthcare: occurrence, severity and co-morbidity in a two-stage procedure of opportunistic screening. Nord J Psychiatry 2010;64:421–7. [DOI] [PubMed] [Google Scholar]
- [14].Saboya PM, Zimmermann PR, Bodanese LC. Association between anxiety or depressive symptoms and arterial hypertension, and their impact on the quality of life. Int J Psychiatry Med 2010;40:307–20. [DOI] [PubMed] [Google Scholar]
- [15].O’Connor EA, Whitlock EP, Beil TL, et al. Screening for depression in adult patients in primary care settings: a systematic evidence review. Ann Intern Med 2009;151:793–803. [DOI] [PubMed] [Google Scholar]
- [16].Institute of Medicine (US) Committee on Public Health Priorities to Reduce and Control Hypertension. A Population-Based Policy and Systems Change Approach to Prevent and Control Hypertension. Washington (DC): National Academies Press (US); 2010. [PubMed] [Google Scholar]
- [17].Katon W, Von Korff M, Lin E, et al. Collaborative management to achieve treatment guidelines: impact on depression in primary care. JAMA 1995;273:1026–31. [PubMed] [Google Scholar]
- [18].Gilbody S, Whitty P, Grimshaw J, et al. Educational and organizational interventions to improve the management of depression in primary care: a systematic review. JAMA 2003;289:3145–51. [DOI] [PubMed] [Google Scholar]
- [19].Unützer J, Katon W, Callahan CM, et al. Collaborative care management of late-life depression in the primary care setting: a randomized controlled trial. JAMA 2002;288:2836–45. [DOI] [PubMed] [Google Scholar]
- [20].Wang L, Li N, Heizhati M, et al. Prevalence, awareness, treatment, and control and related factors of hypertension in multiethnic agriculture, stock-raising, and Urban Xinjiang, Northwest China: a cross-sectional screening for 47000 adults. Int J Hypertens 2019;2019:3576853. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [21].Joint Committee for Guideline Revision. 2018 Chinese guidelines for prevention and treatment of hypertension-a report of the revision committee of Chinese guidelines for prevention and treatment of hypertension. J Geriatr Cardiol 2019;16:182–241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Leung CM, Wing YK, Kwong PK, et al. Validation of the Chinese-Cantonese version of the hospital anxiety and depression scale and comparison with the Hamilton Rating Scale of Depression. Acta Psychiatr Scand 1999;100:456–61. [DOI] [PubMed] [Google Scholar]
- [23].Bull FC, Maslin TS, Armstrong T. Global physical activity questionnaire (GPAQ): nine country reliability and validity study. J Phys Act Health 2009;6:790–804. [DOI] [PubMed] [Google Scholar]
- [24].Global Recommendations on Physical Activity for Health. Geneva: World Health Organization; 2010. [PubMed] [Google Scholar]
- [25].Buysse DJ, Reynolds CF, Monk TH, et al. The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res 1989;28:193–213. [DOI] [PubMed] [Google Scholar]
- [26].Chuang LL, Chuang YF, Hsu MJ, et al. Validity and reliability of the Traditional Chinese version of the Multidimensional Fatigue Inventory in general population. PLoS One 2018;13:e0189850. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Zhai FY, Zhang LW, Wang CR, et al. Validation of lipids on body mass index reference recommended by Obesity Working Group, International Life Science Association of China. Zhonghua Liu Xing Bing Xue Za Zhi 2004;25:117–9. [PubMed] [Google Scholar]
- [28].Alhamzawi R, Ali HTM. The Bayesian adaptive lasso regression. Math Biosci 2018;303:75–82. [DOI] [PubMed] [Google Scholar]
- [29].Sauerbrei W, Royston P, Binder H. Selection of important variables and determination of functional form for continuous predictors in multivariable model building. Stat Med 2007;26:5512–28. [DOI] [PubMed] [Google Scholar]
- [30].Balachandran VP, Gonen M, Smith JJ, et al. Nomograms in oncology: more than meets the eye. Lancet Oncol 2015;16:173–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [31].Poncet A, Perneger TV, Merlani P, et al. Determinants of the calibration of SAPS II and SAPS 3 mortality scores in intensive care: a European multicenter study. Crit Care 2017;21:85. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Van Calster B, Wynants L, Verbeek JFM, et al. Reporting and interpreting decision curve analysis: a guide for investigators. Eur Urol 2018;74:796–804. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [33].Johansen A, Holmen J, Stewart R. Anxiety and depression symptoms in arterial hypertension: the influence of antihypertensive treatment. The HUNT study, Norway. Eur J Epidemiol 2012;27:63–72. [DOI] [PubMed] [Google Scholar]
- [34].Abreu-Silva EO, Todeschini AB. Depression and its relation with uncontrolled hypertension and increased cardiovascular risk. Curr Hypertens Rev 2014;10:8–13. [DOI] [PubMed] [Google Scholar]
- [35].Rubio-Guerra AF, Rodriguez-Lopez L, Vargas-Ayala G, et al. Depression increases the risk for uncontrolled hypertension. Exp Clin Cardiol 2013;18:10–2. [PMC free article] [PubMed] [Google Scholar]
- [36].Michal M, Wiltink J, Lackner K, et al. Association of hypertension with depression in the community: results from the Gutenberg Health Study. J Hypertens 2013;31:893–9. [DOI] [PubMed] [Google Scholar]
- [37].Pálinkás A, Sándor J, Papp M, et al. Associations between untreated depression and secondary health care utilization in patients with hypertension and/or diabetes. Soc Psychiatry Psychiatr Epidemiol 2019;54:255–76. [DOI] [PubMed] [Google Scholar]
- [38].Son YJ, Park C, Won MH. Impact of physical activity and sleep duration on depressive symptoms in hypertensive patients: results from a nationally representative Korean sample. Int J Environ Res Public Health 2018;15:2611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [39].Mahmood S, Hassan SZ, Tabraze M, et al. Prevalence and predictors of depression amongst hypertensive individuals in Karachi, Pakistan. Cureus 2017;9:e1397. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [40].Gray CA, Sims OT, Oh H. Prevalence and predictors of co-occurring hypertension and depression among community-dwelling older adults. J Racial Ethn Health Disparities 2020;7:365–73. [DOI] [PubMed] [Google Scholar]
- [41].Ma C. The prevalence of depressive symptoms and associated factors in countryside-dwelling older Chinese patients with hypertension. J Clin Nurs 2018;27:2933–41. [DOI] [PubMed] [Google Scholar]
- [42].Ma L, Tang Z, Sun F, et al. Risk factors for depression among elderly subjects with hypertension living at home in China. Int J Clin Exp Med 2015;8:2923–8. [PMC free article] [PubMed] [Google Scholar]
- [43].McGirr A, Diaconu G, Berlim MT, et al. Dysregulation of the sympathetic nervous system, hypothalamic-pituitary-adrenal axis and executive function in individuals at risk for suicide. J Psychiatry Neurosci 2010;35:399–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [44].Demirtürk E, Hacihasanoğlu Aşilar R. The effect of depression on adherence to antihypertensive medications in elderly individuals with hypertension. J Vasc Nurs 2018;36:129–39. [DOI] [PubMed] [Google Scholar]