Clusters of longitudinal risk profile trajectories are associated with cardiometabolic diseases: Results from the population-based KORA cohort

Fiona Niedermayer; Gunther Schauberger; Wolfgang Rathmann; Stefanie J Klug; Barbara Thorand; Annette Peters; Susanne Rospleszcz

doi:10.1371/journal.pone.0300966

. 2024 Mar 28;19(3):e0300966. doi: 10.1371/journal.pone.0300966

Clusters of longitudinal risk profile trajectories are associated with cardiometabolic diseases: Results from the population-based KORA cohort

Fiona Niedermayer ^1,^2,³, Gunther Schauberger ³, Wolfgang Rathmann ^4,⁵, Stefanie J Klug ³, Barbara Thorand ^2,⁴, Annette Peters ^1,^2,^4,⁶, Susanne Rospleszcz ^1,^2,^6,^7,^*

Editor: Peng Gao⁸

PMCID: PMC10977748 PMID: 38547172

Abstract

Background

Multiple risk factors contribute jointly to the development and progression of cardiometabolic diseases. Therefore, joint longitudinal trajectories of multiple risk factors might represent different degrees of cardiometabolic risk.

Methods

We analyzed population-based data comprising three examinations (Exam 1: 1999–2001, Exam 2: 2006–2008, Exam 3: 2013–2014) of 976 male and 1004 female participants of the KORA cohort (Southern Germany). Participants were followed up for cardiometabolic diseases, including cardiovascular mortality, myocardial infarction and stroke, or a diagnosis of type 2 diabetes, until 2016. Longitudinal multivariate k-means clustering identified sex-specific trajectory clusters based on nine cardiometabolic risk factors (age, systolic and diastolic blood pressure, body-mass-index, waist circumference, Hemoglobin-A1c, total cholesterol, high- and low-density lipoprotein cholesterol). Associations between clusters and cardiometabolic events were assessed by logistic regression models.

Results

We identified three trajectory clusters for men and women, respectively. Trajectory clusters reflected a distinct distribution of cardiometabolic risk burden and were associated with prevalent cardiometabolic disease at Exam 3 (men: odds ratio (OR)_ClusterII = 2.0, 95% confidence interval: (0.9–4.5); OR_ClusterIII = 10.5 (4.8–22.9); women: OR_ClusterII = 1.7 (0.6–4.7); OR_ClusterIII = 5.8 (2.6–12.9)). Trajectory clusters were furthermore associated with incident cardiometabolic cases after Exam 3 (men: OR_ClusterII = 3.5 (1.1–15.6); OR_ClusterIII = 7.5 (2.4–32.7); women: OR_ClusterII = 5.0 (1.1–34.1); OR_ClusterIII = 8.0 (2.2–51.7)). Associations remained significant after adjusting for a single time point cardiovascular risk score (Framingham).

Conclusions

On a population-based level, distinct longitudinal risk profiles over a 14-year time period are differentially associated with cardiometabolic events. Our results suggest that longitudinal data may provide additional information beyond single time-point measures. Their inclusion in cardiometabolic risk assessment might improve early identification of individuals at risk.

Introduction

Cardiometabolic diseases, such as cardiovascular diseases (CVD) and type 2 diabetes (T2D), are among the leading causes of mortality and morbidity worldwide. Today, these diseases represent a significant global burden on the healthcare system [1, 2].

Development of cardiometabolic diseases is multifaceted and risk factors include obesity, physical inactivity, dietary factors, dyslipidemia and hypertension, with relative risk factor contributions being different in men and women [3, 4]. Most of these risk factors are modifiable and can be targeted by primordial or primary prevention. Therefore, identification of high-risk individuals at an early stage is crucial for timely intervention and risk factor treatment. Risk assessment is usually done with established risk scores, such as the Framingham Risk Score [5] or the Finnish or German Diabetes Risk Score [6, 7], which incorporate multiple risk factors. These scores only include measures from a single time point. However, individual cardiometabolic risk builds up long-term and can change over time [8, 9]. Consequently, joint modelling of multiple risk factors allows to evaluate how risk factors longitudinally interact and influence each other.

Longitudinal trajectories convey information about the individual progression and risk development. This has been recognized in recent studies that investigated associations of blood pressure or lipid changes with CVD [10–12]. Increasing cumulative 10-year systolic blood pressure levels were associated with higher lifetime risk of CVD [10] and incident CVD [11]. In the Framingham Heart Study, long-term lipid levels over approximately 30 years were more strongly associated with elevated coronary artery calcium than single time-point lipid levels [12]. Clustering, i.e. grouping together individuals with similar trajectories, can be used to identify patterns [13] and also allow for even earlier identification of at-risk individuals beyond the already known cut-off values [14, 15].

First associations between model-based trajectory clusters and cardiometabolic events have already been demonstrated [16–20]. However, trajectory clusters based on a single risk factor cannot account for the multifactorial etiology of cardiometabolic diseases. To evaluate the interplay of multiple risk factors, Niiranen et al. [21] calculated a multivariate trajectory score based on each single identified risk factor trajectory cluster and found an association with a 2.8-fold higher risk of incident CVD. Apart from this work, reports on multivariate trajectory clusters are scarce–yet these clusters would be able to account for both the multifactorial etiology and the temporal risk factor evolution of cardiometabolic diseases. Furthermore, women develop CVD on average later in life than men [22], which cannot be attributed solely to biological differences, but may be due to the lack of sex-specific cut-offs for cardiometabolic markers leading to delayed diagnosis and initiation of treatment [3, 4]. In this regard, few studies have considered possible sex-specific clustering of cardiometabolic risk factors [19, 21, 23].

To address this issue, the present study aims to identify distinct clusters of sex-specific risk profile trajectories over a 14-year time horizon on a population-based level and evaluate their association with prevalent and incident cardiometabolic events.

Materials and methods

Study sample

We used longitudinal data from the KORA (“Cooperative Health Research in the Region Augsburg”) S4 sample, which is a population-based cohort study from Bavaria, Germany. A detailed description of recruitment, sampling, and examination protocols is presented elsewhere [24]. The original cohort was repeatedly examined at three time points (Exam 1, Exam 2, Exam 3). Briefly, 4,261 individuals participated in the baseline assessment (Exam 1, S4, 1999–2001), 3,080 of these participated in the seven-year follow-up (Exam 2, F4, 2006–2008) and 2,279 individuals of the original sample participated in the 14-year follow-up (Exam 3, FF4, 2013–2014). After Exam 3, participants were followed up until 2016 for mortality, and morbidity of several chronic diseases.

All KORA studies comply with the Declaration of Helsinki. Each participant gave written informed consent and ethical approval was granted by the Ethics Committee of the Bavarian Medical Association and the Bavarian commissioner for data protection and privacy [S4/Exam 1: EC No. 99186, F4/Exams 2 and FF4/Exam 3: EC No. 06068.]. For the current analysis, data access was granted on April 15, 2019. Authors had no means to access information that could identify individual participants during or after data collection.

For the present analysis, we used a final analytical sample of 1980 individuals with complete data until the end of follow-up 2016. We excluded 118 participants who missed Exam 2 and 181 participants were excluded because they had missing data in any of the risk factors of interest. A detailed participant flow is presented in S1 Fig in S1 File.

Risk factor assessment

Trained staff performed standardized physical examinations, blood draw and face-to-face interviews at all three examinations. A detailed description of measurement methods is described elsewhere [25–27]. Briefly, low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol and total cholesterol were measured by enzymatic photometric assays at Exam 1 and by enzymatic colorimetric Flex assays at Exam 2 and 3. Hemoglobin—A1c (HbA1c) was assessed by a turbidimetric inhibition immunoassay at Exam 1 and by a cation-exchange high performance liquid chromatographic photometric assay at Exam 2 and 3. More details are provided in S1 Table in S1 File.

Height and weight were measured by Seca’s measuring systems (Seca GmbH & Co, KG, Hamburg). Body-Mass Index (BMI) was calculated as weight divided by squared height (kg/m²). Waist circumference (WC) was determined by using an inelastic tape at the level midway between the lower rib margin and the iliac crest.

Systolic and diastolic blood pressure (BP) were measured on the right arm after the participant rested for at least five minutes. An OMRON type HEM-705CP device was used at all examinations. The measurement was repeated three times with a three-minute interval in-between, and the mean of the second and third measurement was used as the final value.

Medication intake, physical activity, smoking behavior and alcohol consumption were self-reported, as described before [28–30]. Participants were categorized into never-smoker, ex-smoker, or smoker and the number of pack years were calculated. Physical activity was estimated based on participants reported duration of leisure times spent on sport activities based on four possible answers: (1) >2 hours per week, (2) 1–2 hours per week, (3) <1 hour per week and (4) none, separately for winter and summer. A total score was obtained by summing the numbers (1)-(4) relating to summer and winter. Participants were classified as physically active if their total score was < 5; otherwise, they were classified as “physically inactive” [29]. The Framingham Risk score was calculated according to the original formula at all three examinations [5].

Outcome assessment

Outcomes of interest were twofold. 1) Any first cardiometabolic event defined as cardiovascular mortality, nonfatal and fatal myocardial infarction and stroke, or a diagnosis of T2D after Exam 1, but before, or at, Exam 3. These were evaluated as prevalent events at Exam 3. 2) Any first incident cardiometabolic event after Exam 3 until the end of follow up in 2016. These were evaluated as incident events after Exam 3. Data on causes of deaths were obtained from official death certificates, collected by health departments, physicians, or coroners. Mortality due to any cardiovascular event was defined by the ICD-9 codes (390–459; 798). Myocardial infarction of individuals aged below 75 were registered by the Augsburg Hospital Myocardial Registry. Data on stroke or myocardial infarction of participants aged older than 75 or of participants moving out of the study area were identified by self-report assessed in regular follow-up questionnaires or telephone interviews. All self-reported events were validated by contacting treating physicians or by sighting health records. T2D was defined as a self-reported diagnosis by a physician or self-reported intake of antidiabetic medication and again validated by contacting treating physicians or by sighting health records.

Statistical analysis

The statistical analysis workflow is presented in S2 Fig in S1 File. Briefly, after descriptive statistics of the study sample, we derived longitudinal trajectory clusters, described the distribution of events and event-free survival according to clusters, and quantified the association of clusters with prevalent and incident cardiometabolic events.

Descriptive statistics

All analyses were stratified by sex. Individual risk factors at each examination are presented as means and standard deviations for continuous variables and as counts and percentages for categorical variables. For skewed variables, medians are additionally provided. Changes over time were assessed with repeated measures ANOVA (skewed variables: Friedman test) and Cochran’s Q test, respectively. To assess differences between excluded individuals and individuals from the final sample, their risk factor values at Exam 1 were compared by t-test, Wilcoxon-Test or X²-Test, where applicable.

Identification of longitudinal trajectory clusters

We used non-parametric k-means with Euclidean distance to identify sex-specific clusters of longitudinal risk factor trajectories. Since the distance calculation with categorical variables using Euclidean distance is not straightforward, we only considered continuous variables for clustering. Nine established cardiometabolic risk factors, obtained at all three examinations, were included: age, HDL cholesterol, LDL cholesterol, total cholesterol, BMI, WC, HbA1c, systolic BP and diastolic BP. Selection of these variables was based on their established role in cardiometabolic disease and their ready availability in clinical assessment. In principle, k-means groups subjects with similar trajectories, i.e., risk factor values at each Exam and similar changes over time, into a cluster. Individuals with dissimilar trajectories are assigned to different clusters. (Dis-) similarity was determined by Euclidean distance.

Since k-means is sensitive to outliers [31], all risk factor data except age were winsorized at the sex-specific 95%-quantile. Standardization per trajectory and subsequent clustering were performed with the R package “kml3d” [32]. The final number of clusters was determined by Calinski-Harabasz criterion while allowing up to eight clusters (S2 Table in S1 File). Mean risk factor levels at each exam according to cluster membership were visualized by line plots.

Association of trajectory clusters with cardiometabolic events

Distribution of cardiometabolic events according to cluster membership was visualized by bar plots. To assess the association of trajectory clusters with the first outcome of interest–incident events after Exam 1, but before, or at Exam 3, and thus prevalent at Exam 3—logistic regression models adjusted for age at baseline were calculated. Additionally, Kaplan-Meier curves were used to visualize how the cumulative incidence of first cardiometabolic events during the 14-year horizon between Exam 1 and Exam 3 evolved according to cluster membership. The sample used for visualization with Kaplan-Meier curves was identical to the sample where we analyzed the first outcome of interest, except for exclusion of n = 72 prevalent cases at baseline (S2 Fig, S4 Table in S1 File). Differences according to cluster membership were evaluated by log-rank test.

To assess the association of trajectory clusters with the second outcome of interest—incident events after Exam 3 -, all prevalent diseases at Exam 3 (n = 292, S2 Fig and S4 Table in S1 File) were excluded. Logistic regression models then were calculated and adjusted for the Framingham Risk Score at Exam 3, in order to examine whether cluster membership could provide additional information over the score.

As sensitivity analyses, CVD and T2D events were considered separately. Results of the logistic regression are given as Odds Ratios (OR), corresponding 95%- confidence intervals (CI) and p-values. A two-sided p-value <0.05 was considered to indicate statistical significance. All statistical analyses were performed with R 4.0.5 (R Core Team, Vienna, Austria).

Results

Study sample

The final sample comprised 976 men and 1004 women. At baseline, men were on average 48 years and women 47 years old. During the 14-year horizon between Exam 1 and Exam 3, all risk factors except triglycerides in women significantly changed. Mean blood pressure levels were declining, while intake of antihypertensive medication increased, and mean HDL cholesterol levels were u-shaped for both men and women. Mean BMI and WC increased in both sexes. While men had decreasing mean LDL cholesterol and total cholesterol levels, these lipid levels increased or remained stable in women. Intake of lipid-lowering medication increased in both sexes. Mean HbA1c levels significantly rose in men but declined in women. At Exam 3, more participants were physically active and fewer smoked (Table 1).

Table 1. Characteristics of male and female participants at Exam 1, Exam 2 and Exam 3.

	Men (n = 976)				Women (n = 1004)
	Exam 1 (1999–2001)	Exam 2 (2006–2008)	Exam 3 (2013–2014)	p-value	Exam 1 (1999–2001)	Exam 2 (2006–2008)	Exam 3 (2013–2014)	p-value
Age [years]	47.5 ± 12.6	54.7 ± 12.5	61.1 ± 12.5		46.9 ± 12.0	54.1 ± 12.0	60.5 ± 12.0
Blood pressure (BP)
Systolic BP [mmHg]	132.6 ± 16.4	127.4 ± 17.0	124.2 ± 16.4	<0.001	120.6 ± 17.7	115.7 ± 16.7	114.2 ± 17.2	<0.001
Diastolic BP [mmHg]	83.2 ± 10.1	78.1 ± 9.6	75.1 ± 10.1	<0.001	77.5 ± 9.9	72.9 ± 9.2	70.7 ± 8.8	<0.001
Anthropometric factors
Body-Mass Index [kg/m²]	27.3 ± 3.8	27.8 ± 4.1	28.2 ± 4.4	<0.001	26.4 ± 4.9	26.9 ± 5.1	27.5 ± 5.4	<0.001
Weight [kg]	84.3 ± 12.3	86.5 ± 14.0	87.1 ± 15.0	<0.001	69.6 ± 13.0	71.1 ± 13.5	72.0 ± 14.4	<0.001
Waist circumference [cm]	96.4 ± 10.2	99.2 ± 12.0	102.8 ± 12.1	<0.001	83.4 ± 11.7	86.8 ± 12.9	91.3 ± 13.6	<0.001
Lipids
Total cholesterol [mg/dl]	226.3 ± 41.4	213.1 ± 37.0	210.5 ± 39.1	<0.001	223.2 ± 40.6	217.8 ± 39.3	223.3 ± 38.9	<0.001
HDL cholesterol [mg/dl]	51.3 ± 13.5	50.2 ± 11.8	58.1 ± 15.5	<0.001	65.3 ± 17.4	62.2 ± 14.4	73.2 ± 18.5	<0.001
LDL cholesterol [mg/dl]	139.7 ± 39.4	137.6 ± 32.6	134.0 ± 34.9	<0.001	129.5 ± 39.7	133.9 ± 35.5	136.3 ± 36.0	<0.001
Triglycerides [mg/dl]^a ^b	157.4 ± 105.7 Median: 129.5	141.3 ± 104.9 Median: 118.0	135.9 ± 79.7 Median: 115.7	<0.001	120.4 ± 65.3 Median: 106.0	100.9 ± 55.6 Median: 88.0	108.4 ± 51.9 Median: 96.8	0.292
Glucose metabolism
HbA1c [%]	5.5 ± 0.6	5.5 ± 0.6	5.6 ± 0.7	0.014	5.6 ± 0.5	5.5 ± 0.5	5.5 ± 0.6	<0.001
2-hour glucose [mg/dl]^a^,^b	120.0 ± 49.4 Median: 108.0	111.9 ± 35.9 Median: 100.0	116.7 ± 42.5 Median: 107.0	<0.001	119.1 ± 44.5 Median: 109.0	107.1 ± 35.8 Median:100.0	109.8 ± 39.8 Median: 101.0	<0.001
Fasting blood glucose [mg/dl]^a	103.6 ± 16.1	100.9 ± 18.2	106.5 ± 22.6	<0.001	98.4 ± 14.3	93.2 ± 13.9	99.2 ± 20.4	<0.001
Inflammation
C-reactive protein [mg/L]^b	2.0 ± 3.6 Median: 0.9	1.9 ± 3.0 Median: 1.0	2.2 ± 3.7 Median: 1.1	<0.001	2.7 ± 5.0 Median: 1.2	2.3 ± 4.0 Median: 1.1	2.6 ± 5.1 Median: 1.3	<0.001
Medication
Antihypertensive	126 (12.9%)	271 (27.8%)	364 (37.3%)	<0.001	126 (12.6%)	232 (23.1%)	350 (34.9%)	<0.001
Lipid-lowering	57 (5.8%)	121 (12.4%)	188 (19.3%)	<0.001	42 (4.2%)	95 (9.5%)	142 (14.1%)	<0.001
CVD risk
Framingham risk score [%]	13.7 ± 12.4	17.3 ± 14.5	19.3 ± 14.4	<0.001	5.6 ± 6.0	6.7 ± 6.8	7.9 ± 7.6	<0.001
Lifestyle factors
Alcohol consumption [g/day]^b	23.7 ± 24.7 Median: 17.1	21.9 ± 24.2 Median: 17.0	21.7 ± 24.0 Median: 15.8	0.003	8.8 ± 12.0 Median: 2.9	8.1 ± 12.1 Median: 2.9	8.5 ± 12.8 Median: 2.9	<0.001
Pack years^b	14.7 ± 21.7 Median: 5.9	16.3 ± 23.4 Median: 6.6	17.1 ± 24.3 Median: 7.8	<0.001	6.0 ± 10.2 Median: 0.0	6.8 ± 11.5 Median: 0.0	7.3 ± 12.3 Median: 0.0	<0.001
Physically active	518 (53.1%)	558 (57.2%)	550 (56.4%)	0.030	541 (53.9%)	597 (59.5%)	594 (59.2%)	<0.001
Smoking				<0.001				<0.001
never-smoker	333 (34.1%)	333 (34.1%)	333 (34.1%)		526 (52.4%)	524 (52.2%)	521 (51.9%)
ex-smoker	399 (40.9%)	459 (47.0%)	486 (49.8%)		274 (27.3%)	328 (32.7%)	347 (34.6%)
smoker	244 (25.0%)	184 (18.9%)	157 (16.1%)		204 (20.3%)	152 (15.1%)	136 (13.6%)

Open in a new tab

Continuous variables are presented as mean ± standard deviation, p-values are calculated by repeated measures ANOVA indicating whether mean values differ significantly on at least one exam. Categorical variables are presented as counts and percentages, p-values are calculated by Cochran’s Q test and indicating whether counts differ significantly on at least one exam.

^aSample size deviate (Men: Fasting blood glucose: N_Exam1 = 293, N_Exam2: 971, N_Exam3 = 965; 2-hour glucose: N_Exam1 = 291, N_Exam2: 896, N_Exam3 = 831; triglycerides: N_Exam1 = 360; Women: Fasting blood glucose: N_Exam1 = 268, N_Exam2: 997, N_Exam3 = 990; 2-hour glucose: N_Exam1 = 260, N_Exam2: 942, N_Exam3 = 874; triglycerides: N_Exam1 = 327).

^bFriedman-Test was used due to skewed distribution

Compared to participants retained in the final analysis, excluded participants had a worse cardiometabolic profile at baseline (S3 Table in S1 File). On average, these participants were older, had higher mean levels in systolic blood pressure, BMI, WC, blood lipids, C-reactive protein and were less physically active. In addition, more subjects were taking antihypertensive medications, and more subjects were diagnosed with T2D or experienced a stroke before Exam 1.

Characterization of longitudinal trajectory clusters

Based on Calinski-Harabasz criterion, the optimal number of longitudinal risk profile trajectory clusters was three for both men and women (S2 Table in S1 File). These clusters differed in mean risk factor levels and changes over time.

For men, Cluster I comprised 310 subjects (32% of the male sample). Subjects were on average the youngest at baseline (39.7 ± 11.3 years), had the lowest mean levels of BMI, waist circumference, systolic blood pressure and HbA1c and the highest mean levels of HDL cholesterol over the entire study period (Fig 1 and S5 Table in S1 File). Mean BMI ranged from normal weight to borderline overweight (Exam 1: 24.5 ± 2.2 kg/m²; Exam 3: 25.4 ± 2.5 kg/m²) according to the WHO guidelines [33]. Mean systolic blood pressure was elevated to normal and mean diastolic blood pressure was normal (<80 mm/Hg) according to the American Heart Association blood pressure guidelines [34]. Subjects had on average increasing LDL cholesterol levels (Exam 1: 108.1 ± 25.7 mg/dl; Exam 3: 122.7 ± 26.2 mg/dl) within an optimal lipid range according to the Adult Treatment Panel guidelines for treatment of high blood cholesterol [35]. Total cholesterol levels (Exam 1: 193.1 ± 27.5 mg/dl; Exam 3: 200.3 ± 29.8 mg/dl) increased from normal to borderline high but were still the lowest at Exam 1 and Exam 2 compared to the other clusters. Cluster II included 345 participants (35% of the male sample). Individuals had highest mean values of LDL cholesterol and total cholesterol at all examinations ranging from borderline high to high lipid levels (LDL cholesterol: 130–189 mg/dl, total cholesterol: 200 - ≥ 240 mg/dl [35]). In addition, mean values of HDL cholesterol, HbA1c, BMI, waist circumference, systolic blood pressure and diastolic blood pressure at baseline resided between Cluster I and Cluster III. At Exam 3, mean diastolic blood pressure (76.4 ± 9.1 mmHg) was the highest for subjects included in this cluster, although still normal, while mean systolic blood pressure ranged from stage 1 hypertension (130–139 mmHg) to elevated (120–129 mmHg) [34]. According to the WHO definition [33], these subjects were on average overweight. Cluster III comprised 320 men (33% of the male sample). Throughout the study period, individuals had the highest mean levels of BMI, waist circumference, HbA1c, systolic blood pressure and the lowest mean levels of HDL cholesterol (Fig 1 and S5 Table in S1 File). Moreover, subjects were on average the oldest (54.3 ± 10.7 years) and had the highest diastolic blood pressure (86.4 ± 9.4 mmHg), defined as stage 1 hypertension, at baseline. Cluster III was also characterized by the second highest mean LDL cholesterol and total cholesterol levels. In addition, individuals in this cluster had the strongest decline of blood pressure, e.g., systolic blood pressure decreased from high to elevated mean levels (Exam 1: 138.6 ± 15.5 mmHg; Exam 3: 126.1 ± 15.5 mmHg), probably due to increased intake of antihypertensive medication. Similarly, LDL cholesterol (Exam 1: 140.3 ± 34.2 md/dl; Exam 3: 116.8 ± 30.0 mg/dl) and total cholesterol levels (Exam 1: 226.6 ± 35.5 mg/dl; Exam 3: 189.9 ± 33.5mg/dl) declined over time from borderline high to optimal levels, accompanied by an increased intake in lipid-lowering medication. Mean BMI was consistently in the obese category (≥ 30 kg/m²). In conclusion, in men, Cluster I reflected the lowest, Cluster II medium and Cluster III the highest risk factor burden.

Fig 1 — Presented are mean risk factor levels at each examination according to cluster membership. Clusters are based on male participants’ longitudinal risk profile trajectories determined by k-means clustering.

For women, Cluster I comprised 371 subjects (37% of the female sample). Individuals had the lowest mean levels of BMI, waist circumference, systolic and diastolic blood pressure, LDL cholesterol and the highest HDL cholesterol levels throughout study period (Fig 2 and S6 Table in S1 File). Mean HbA1c values decreased to the lowest mean values at Exam 3 (5.3 ± 0.4%). According to established thresholds for obesity [33], hypertension [34] and dyslipidemia [35], subjects had on average normal weight (18.5–25 kg/m²), normal blood pressure (systolic: <120 mmHg; diastolic: <80 mmHg) and near optimal LDL cholesterol levels (100–129 mg/dl), but borderline high mean total cholesterol levels (200–239 mg/dl). Cluster II included 231 subjects (23% of the female sample). Mean levels of BMI, waist circumference, LDL cholesterol and systolic blood pressure resided between Cluster I and Cluster III at every examination (Fig 2 and S6 Table in S1 File). Moreover, subjects were characterized by the lowest mean values of HDL cholesterol (56.5 ± 12.9 mg/dl), total cholesterol (201.3 ± 32.1 mg/dl; borderline high) and were on average the youngest at baseline (39.3 ± 8.0 years). Mean HbA1c was lowest at baseline but slightly increased after Exam 2, resulting in higher mean values at Exam 3 compared to Cluster I. Diastolic blood pressure only decreased between Exam 1 and Exam 2 and then plateaued this level (73.6 ± 7.4 mmHg; normal blood pressure) until Exam 3. Mean BMI was consistently in the overweight category (25.0–29.9 kg/m²). Cluster III comprised 402 participants (40% of the female sample). In general, participants were characterized by the highest age at baseline (56.5 ± 8.5 years) and by highest mean levels of HbA1c, total cholesterol, LDL cholesterol, BMI, waist circumference and systolic blood pressure over the study period (Fig 2 and S6 Table in S1 File). In contrast to Cluster II, individuals had higher mean HDL cholesterol levels at all examinations. In addition, Cluster III was the only cluster with steadily decreasing mean levels of total cholesterol (Exam 1: 245.8 ± 32.7 mg/dl; Exam 3: 233.6 ± 37.3 mg/dl; range: high to borderline high) and LDL cholesterol (Exam 1: 153.2 ± 30.1 mg/dl; Exam 3: 147.6 ± 33.8; borderline high category) and steadily increasing HbA1c levels over time. The highest decrease in blood pressure was also observed, e.g., mean levels of systolic blood pressure decreased from elevated to normal levels (Exam 1: 129.7 ± 15.4 mmHg; Exam 3: 120.4 ± 15.3 mmHg). In conclusion, in women, Cluster I reflected the lowest, Cluster II medium and Cluster III the highest risk factor burden.

Fig 2 — Presented are mean risk factor levels at each examination according to cluster membership. Clusters are based on female participants’ longitudinal risk profile trajectories determined by k-means clustering.

It is important to note that men and women exhibited different behavior in some risk factors, especially for the high-risk Cluster III. While diastolic blood pressure, total cholesterol and LDL cholesterol decreased substantially for men in Cluster III to the lowest mean levels at Exam 3, these levels also decreased for women in Cluster III, but were still at a high-risk level at Exam 3 and highest compared to the other female clusters (Figs 1 and 2). In addition, Cluster II represented decreasing total and LDL cholesterol levels in men but increasing levels in women. The decreasing trend for diastolic blood pressure was stronger in male Cluster II than in female Cluster II. Furthermore, sex differences were also observed for HbA1c in Cluster I and Cluster II.

Association between trajectory clusters and cardiometabolic events

At Exam 3, there were 170 prevalent cardiometabolic events in men and 112 in women. The majority of events had occurred in Cluster III (Fig 3 and S4 Table in S1 File). As shown in Fig 4, cumulative event-free survival up to Exam 3 was highest in Cluster I and lowest in Cluster III with significant differences between clusters (Log-Rank Test: p<0.001). In logistic regression, for men Cluster II and Cluster III were associated with a 2.7-fold and 16.7-fold higher risk of a prevalent event at Exam 3 compared to Cluster I. This association remained significant only for Cluster III after adjusting for age at Exam 1 (OR = 10.5 (4.8–22.9)). For women, Cluster III was associated with a 11.2-fold higher risk of a prevalent event at Exam 3, which attenuated after adjusting for age at Exam 1 but remained highly statistically significant (OR = 5.8 (2.6–12.9)). For women, Cluster II did not confer a significantly higher risk of a prevalent cardiometabolic event (OR: 1.4 (0.5–4.0)).

Fig 4 — On the y-axis: probability of event-free survival without any cardiometabolic event, stratified by cluster membership. On the x-axis: Time from Exam 1 until und of follow-up (3 years after Exam 3) in days. Periods of Exam 2 and 3 are marked by dashed grey lines. Marked decreases during the examination periods are mainly due to newly identified T2D cases that were ascertained during the OGTT performed as part of the examinations. For the visualization of Kaplan-Meier curves, prevalent cardiometabolic events at Exam 1 were excluded, resulting in a sample size of n = 926 for men and n = 982 for women.

After Exam 3, 28 men and 21 women had an incident cardiometabolic event until the end of follow-up. The majority of incident events after Exam 3 occurred in individuals assigned to Cluster III (Fig 3 and S4 Table in S1 File). Cluster III was significantly associated with a higher risk of incident cardiometabolic events in men (OR = 7.5 (95%-CI: 2.4–32.7)) and in women (OR = 8.0 (2.2–51.7)), Table 2. This association attenuated but remained significant after adjusting for the Framingham risk score at Exam 3 (men: OR_ClusterIII = 5.2 (1.5–24.0); women: OR_ClusterIII = 6.0 (1.4–41.4)).

Table 2. Association between cluster membership and cardiometabolic events assessed by logistic regression models.

Sex	Outcome		Adjustment Model	Cluster I	Cluster II
Sex	Outcome		Adjustment Model		OR (95%-CI)	p-value	OR (95%-CI)	p-value
*Prevalent events at Exam 3*
Men	cardiometabolic		Crude	Reference	2.7 (1.2–6.1)	0.019	16.7 (7.9–35.2)	<0.001
	cardiometabolic		Crude + Age at Exam 1	Reference	2.0 (0.9–4.5)	0.114	10.5 (4.8–22.9)	<0.001
	T2D		Crude	Reference	2.4 (0.8–6.7)	0.105	19.3 (7.7–48.7)	<0.001
	T2D		Crude + Age at Exam 1	Reference	1.7 (0.6–4.8)	0.344	11.4 (4.4–29.8)	<0.001
	CVD		Crude	Reference	3.3 (0.9–12.1)	0.066	11.7 (3.5–38.9)	<0.001
	CVD		Crude + Age at Exam 1	Reference	2.6 (0.7–9.6)	0.154	7.8 (2.2–27.5)	0.001
Women	cardiometabolic		Crude	Reference	1.4 (0.5–4.0)	0.505	11.2 (5.3–23.5)	<0.001
	cardiometabolic		Crude + Age at Exam 1	Reference	1.7 (0.6–4.7)	0.332	5.8 (2.6–12.9)	<0.001
	T2D		Crude	Reference	1.6 (0.3–8.1)	0.560	23.0 (7.1–74.0)	<0.001
	T2D		Crude + Age at Exam 1	Reference	1.7 (0.3–8.6)	0.514	16.8 (4.9–58.1)	<0.001
	CVD		Crude	Reference	1.3 (0.3–4.9)	0.707	4.1 (1.5–11.0)	0.006
	CVD		Crude + Age at Exam 1	Reference	2.1 (0.5–8.2)	0.301	1.2 (0.4–3.42)	0.781
*Incident events after Exam 3*
Men		cardiometabolic	Crude	Reference	3.5 (1.1–15.6)	0.056	7.5 (2.4–32.7)	0.002
Men		cardiometabolic	Crude + FRS at Exam 3	Reference	2.6 (0.8–12.1)	0.161	5.2 (1.5–24.0)	0.017
Women		cardiometabolic	Crude	Reference	5.0 (1.1–34.1)	0.051	8.0 (2.2–51.7)	0.006
Women		cardiometabolic	Crude + FRS at Exam 3	Reference	4.7 (1.1–32.6)	0.058	6.0(1.4–41.4)	0.027

Open in a new tab

Odds Ratios are derived from logistic regression models and denotes the fold-change in risk. For the regression models with prevalent events at Exam 3, prevalent events at Exam 1 were excluded, resulting in a sample size of n = 926 for men and n = 982 for women. For the regression models with incident events after Exam 3, prevalent events at Exam 3 were excluded, resulting in a sample size of n = 806 for men and n = 892 for women. Abbreviations: OR = Odds ratio; CI = Confidence interval; FRS = Framingham Risk Score

In sensitivity analyses, we evaluated T2D and CVD separately and found that trajectory clusters were significantly associated with prevalent T2D at Exam 3 for men and women (Table 2, S3 Fig in S1 File). For CVD, Cluster III was associated with a higher risk of prevalent CVD for men, while there was no significant association for women (Table 2, S5 Fig in S1 File). Similar to the main analysis, event-free survival for both T2D and CVD and was highest in Cluster I and lowest in Cluster III (S4 and S6 Figs in S1 File). Log-rank test indicated significant differences across clusters for both T2D and CVD (all p-values <0.001). Due to the low number of incident events after Exam 3, logistic models were not run separately for CVD and T2D.

Discussion

We used a non-parametric cluster approach to identify sex-specific cardiometabolic risk profile trajectory clusters in a population-based cohort over a 14-year time period. Our findings are threefold. First, risk profiles significantly changed over time with different mean lipid trajectories for men and women. Second, we identified three distinct clusters for men and women, which reflected low, moderate, and high cumulative cardiometabolic risk factor burden. Third, trajectory clusters were associated with prevalence of cardiometabolic events at Exam 3 and with incident cardiometabolic events within three years after Exam 3. Furthermore, our results suggest that clusters yielded additional information beyond the Framingham risk score.

Our results thus underscore the clinical importance of considering the longitudinal evolution of multiple risk factors to identify individuals at risk for T2D and CVD early. Cardiometabolic diseases development is multifactorial, and there is substantial interplay between the individual risk factors [36]. However, studies reporting on the joint development of multiple risk factors are scarce. As an exception, Niiranen et al. [21] calculated a multivariate trajectory score based on single risk factor trajectory clusters in the Framingham study. They found a significant association of this score with incident CVD, beyond single-point measures. The importance of longitudinal trajectories is corroborated by further analyses from a subset of the current cohort. Trajectory clusters reflecting a high risk factor burden were associated with worse cardiac function [37] and increased adipose tissue [27], as measured by magnetic resonance imaging. These measures indicate early stages of fully-fledged cardiometabolic diseases. Again, longitudinal trajectories gave additional information beyond single time-point measures [27, 37].

Longitudinal trajectories of single risk factors have already been reported. Previous studies reported mainly increasing blood pressure levels over time [17, 38, 39], in contrast to our findings. This may have different reasons; first, the proportion of subjects on antihypertensive medication might be higher in the KORA cohort compared to other cohorts. Since 2003 systolic blood pressure levels >130 mmHg are classified as prehypertensive [40], potentially leading to earlier treatment of elevated blood pressure. Second, characteristics of participants, especially age and time of enrollment, might differ in other studies. In comparison, studies that reported mainly increasing levels of blood pressure were either conducted earlier [39], when antihypertensive treatment was potentially given only for advanced hypertension, or the sample was younger and therefore, mean blood pressure levels and the number of participants on antihypertensive medications were lower [38].

Additionally, we found the highest risk for cardiometabolic events in Cluster III for men and women. These clusters had the steepest decline of blood pressure and the highest proportion of subjects on antihypertensive medication. This is in good agreement with results from previous studies [17, 41], that demonstrated highest risk of CVD events in clusters with steepest blood pressure decline. Here, the intake of antihypertensive medication seems to represent a proxy for unfavorable blood pressure levels and thus characterize individuals at increased risk.

We observed decreasing levels of lipid parameters in men. This is likely due to increasing number of subjects on lipid-lowering medication. Previous studies provided evidence for this link [42, 43]. So far, three studies identified longitudinal risk factor trajectory clusters of lipid parameters [16, 23, 44]. While trajectories with overall higher HDL cholesterol levels seem to be associated with a favorable cardiovascular profile, trajectories with higher levels of LDL cholesterol or total cholesterol are associated with higher CVD risk. In addition, sex-specific differences in lipid trajectories have already been observed [23]. Recent studies assume this sex difference in lipids can be assigned to sex hormones [45]. Female sex hormones, like estrogens, seem to be protective against cardiovascular diseases and lipid abnormalities. However, this effect is weakened with menopause, which could explain rising blood lipid levels over time [45]. Analogous, we observed increasing total cholesterol and LDL cholesterol levels in Cluster I and Cluster II for women. In these clusters, women were on average 39 and 41 years old at baseline, implying they potentially reached menopause during Exam 2 and Exam 3. However, other explanations for sex-specific trajectories may include differences in the treatment of CVD risk factors and adherence to therapy. Women are still underrepresented in clinical trials [46–48]; they are more often likely to drop out or discontinue lipid-lowering treatment due to side effects, and physicians are less likely to prescribe statin-based or antihypertensive medications to women than to men [47, 49]. This is supported by our findings, where the reduction of total and LDL cholesterol is higher in men than in women and the proportion of lipid-lowering medication use is higher in men. Moreover, there is still a lack of sex-specific treatment guidelines, e.g., whether target values for CVD risk factors achieved in treatment should be the same for men and women [46, 47].

It is worth noting that we observed stronger effects for T2D compared to CVD in sensitivity analysis, which is likely due to the higher numbers of T2D events. As we included only individuals who participated in all examinations, all fatal CVD cases between the exams were excluded. Since these individuals were naturally at a high risk for CVD, our reported effect of trajectory clusters will underestimate the true effect. Moreover, particularly in women, CVD events might have been underdiagnosed [50]. In contrast, new diabetes cases were diagnosed at each exam by OGTT (S2 Fig in S1 File). In line with this point, participants who dropped out during the study period had a worse cardiometabolic risk profile and had more T2D diagnoses and stroke events at baseline compared to the analytical sample, which may explain the relatively stable cluster trajectories over time.

Our findings are relevant for both clinical and epidemiological applications. The importance of longitudinal data in population-based research on chronic diseases research has been recognized [51]. However, longitudinal data is not only provided in research cohorts, but also in clinical health check-ups and regular visits at physicians, providing a direct practical application in the clinical setting. Using the collected data in a systematic longitudinal approach can inform clinical decision-making and help the physician with early therapeutic and lifestyle intervention for the individual patient. To transfer this into practice, longitudinal data should be included for improving established scores commonly used in clinical settings to identify individuals at risk. This may be particularly relevant for women, since these are at higher risk of not being timely diagnosed and treated due to their different presentation of CVD, like atypical MI symptoms, less inflammatory and fibrotic cardiac markers, and lower prevalence of hypertrophy [22]. On the other side, women tend to have more cardiometabolic disturbances jointly [3, 52, 53]. Therefore, monitoring cardiometabolic risk factors over time allows the identification of sex-specific adverse trajectories and can add valuable information on sex-specific cut-off values for cardiometabolic markers.

Limitations of our study need to be considered. First, although the measurement period with 14 years provided longitudinal data from three repeated examinations, the follow-up was comparatively short with only three years after the last examination. Thus, the number of incident events after Exam 3 was rather low and prevented us from fitting fully adjusted models of incident events, and from conducting separate analyses for incident CVD and T2D. As abovementioned, by including only participants with available data at all examinations, we excluded all who died from CVD in between, which will underestimate the true effect and bias the estimates towards the Null. Finally, we did not conduct any formal modeling whether the cluster membership improve the risk prediction compared to the Framingham risk score. Therefore, the advantage of longitudinal information over single time-point risk scores needs to be further investigated and confirmed in future studies.

Conclusions

In conclusion, we identified and characterized cardiometabolic longitudinal risk profile trajectories over a 14-year time horizon based on a population-based cohort. Sex-specific risk profile clusters presented a clear distribution of cardiometabolic risk burden and were associated with cardiometabolic events. Our results emphasize the need of considering longitudinal information in both clinical risk assessment for the individual patient as well as in epidemiological risk prediction models.

Supporting information

S1 File

(DOCX)

pone.0300966.s001.docx^{(1.2MB, docx)}

Acknowledgments

We thank all participants for their long-term commitment to the KORA study, the staff for data collection and research data management and the members of the KORA Study Group (https://www.helmholtz-munich.de/en/epi/cohort/kora) who are responsible for the design and conduct of the study.

List of abbreviations

ANOVA: Analysis of Variance
BMI: Body-Mass-Index
BP: Blood Pressure
CI: Confidence Interval
CVD: Cardiovascular diseases
FRS: Framingham Risk Score
HDL: High-density lipoprotein
HR: Hazard Ratio
KORA: Cooperative Health Research of Augsburg
LDL: Low-density lipoprotein
OR: Odds Ratio
T2D: Type 2 diabetes

Data Availability

The datasets analyzed during the current study are not publicly available due national data protection laws, since the informed consent given by KORA study participants does not cover data posting in public databases. Data are available upon request by means of a project agreement from KORA. Requests should be sent to kora.passt@helmholtz-muenchen.de and are subject to approval by the KORA Board. Analysis codes are available from the authors upon reasonable request.

Funding Statement

This project has been financed in part through HGF Future Topic AMPro. The KORA study was initiated and financed by the Helmholtz Zentrum München – German Research Center for Environmental Health, which is funded by the German Federal Ministry of Education and Research (BMBF) and by the State of Bavaria. Furthermore, KORA research was supported within the Munich Center of Health Sciences (MC-Health), Ludwig-Maximilians-Universität, as part of LMUinnovativ. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

1.Khan MAB, Hashim MJ, King JK, Govender RD, Mustafa H, Al Kaabi J. Epidemiology of Type 2 Diabetes—Global Burden of Disease and Forecasted Trends. J Epidemiol Glob Health. 2020;10(1):107–11. doi: 10.2991/jegh.k.191028.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Roth GA, Mensah GA, Fuster V. The Global Burden of Cardiovascular Diseases and Risks: A Compass for Global Action. J Am Coll Cardiol. 2020;76(25):2980–1. doi: 10.1016/j.jacc.2020.11.021 [DOI] [PubMed] [Google Scholar]
3.Gerdts E, Regitz-Zagrosek V. Sex differences in cardiometabolic disorders. Nat Med. 2019;25(11):1657–66. doi: 10.1038/s41591-019-0643-8 [DOI] [PubMed] [Google Scholar]
4.Mosca L, Barrett-Connor E, Wenger NK. Sex/gender differences in cardiovascular disease prevention: what a difference a decade makes. Circulation. 2011;124(19):2145–54. doi: 10.1161/CIRCULATIONAHA.110.968792 [DOI] [PMC free article] [PubMed] [Google Scholar]
5.D’Agostino RB, Sr., Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–53. doi: 10.1161/CIRCULATIONAHA.107.699579 [DOI] [PubMed] [Google Scholar]
6.Lindstrom J, Tuomilehto J. The diabetes risk score: a practical tool to predict type 2 diabetes risk. Diabetes Care. 2003;26(3):725–31. doi: 10.2337/diacare.26.3.725 [DOI] [PubMed] [Google Scholar]
7.Paprott R, Muhlenbruch K, Mensink GB, Thiele S, Schulze MB, Scheidt-Nave C, et al. Validation of the German Diabetes Risk Score among the general adult population: findings from the German Health Interview and Examination Surveys. BMJ Open Diabetes Res Care. 2016;4(1):e000280. doi: 10.1136/bmjdrc-2016-000280 [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Guo F, Garvey WT. Trends in Cardiovascular Health Metrics in Obese Adults: National Health and Nutrition Examination Survey (NHANES), 1988–2014. J Am Heart Assoc. 2016;5(7). doi: 10.1161/JAHA.116.003619 [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Hardoon SL, Whincup PH, Lennon LT, Wannamethee SG, Capewell S, Morris RW. How much of the recent decline in the incidence of myocardial infarction in British men can be explained by changes in cardiovascular risk factors? Evidence from a prospective population-based study. Circulation. 2008;117(5):598–604. doi: 10.1161/CIRCULATIONAHA.107.705947 [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Allen N, Berry JD, Ning H, Van Horn L, Dyer A, Lloyd-Jones DM. Impact of blood pressure and blood pressure change during middle age on the remaining lifetime risk for cardiovascular disease: the cardiovascular lifetime risk pooling project. Circulation. 2012;125(1):37–44. doi: 10.1161/CIRCULATIONAHA.110.002774 [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Reges O, Ning H, Wilkins JT, Wu CO, Tian X, Domanski MJ, et al. Association of Cumulative Systolic Blood Pressure With Long-Term Risk of Cardiovascular Disease and Healthy Longevity: Findings From the Lifetime Risk Pooling Project Cohorts. Hypertension. 2021;77(2):347–56. doi: 10.1161/HYPERTENSIONAHA.120.15650 [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Tsao CW, Preis SR, Peloso GM, Hwang SJ, Kathiresan S, Fox CS, et al. Relations of long-term and contemporary lipid levels and lipid genetic risk scores with coronary artery calcium in the framingham heart study. J Am Coll Cardiol. 2012;60(23):2364–71. doi: 10.1016/j.jacc.2012.09.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Warren Liao T. Clustering of time series data—a survey. Pattern Recognition. 2005;38(11):1857–74. doi: 10.1016/j.patcog.2005.01.025 [DOI] [Google Scholar]
14.Nagin DS, Tremblay RE. Developmental trajectory groups: fact or a useful statistical fiction? Criminology. 2005;43(4). doi: 10.1111/j.1745-9125.2005.00026.x [DOI] [Google Scholar]
15.Govender P, Sivakumar V. Application of k-means and hierarchical clustering techniques for analysis of air pollution: A review (1980–2019). Atmos Pollut Res. 2020;11(1):40–56. doi: 10.1016/j.apr.2019.09.009 [DOI] [Google Scholar]
16.Duncan MS, Vasan RS, Xanthakis V. Trajectories of Blood Lipid Concentrations Over the Adult Life Course and Risk of Cardiovascular Disease and All-Cause Mortality: Observations From the Framingham Study Over 35 Years. J Am Heart Assoc. 2019;8(11):e011433. doi: 10.1161/JAHA.118.011433 [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Rahman F, Yin X, Larson MG, Ellinor PT, Lubitz SA, Vasan RS, et al. Trajectories of Risk Factors and Risk of New-Onset Atrial Fibrillation in the Framingham Heart Study. Hypertension. 2016;68(3):597–605. doi: 10.1161/HYPERTENSIONAHA.116.07683 [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Yacaman-Mendez D, Trolle-Lagerros Y, Zhou M, Monteiro Ponce de Leon A, Gudjonsdottir H, Tynelius P, et al. Life-course trajectories of weight and their impact on the incidence of type 2 diabetes. Sci Rep. 2021;11(1):12494. doi: 10.1038/s41598-021-91910-z [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Ogata S, Watanabe M, Kokubo Y, Higashiyama A, Nakao YM, Takegami M, et al. Longitudinal Trajectories of Fasting Plasma Glucose and Risks of Cardiovascular Diseases in Middle Age to Elderly People Within the General Japanese Population: The Suita Study. J Am Heart Assoc. 2019;8(3):e010628. doi: 10.1161/JAHA.118.010628 [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Jin C, Chen S, Vaidya A, Wu Y, Wu Z, Hu FB, et al. Longitudinal Change in Fasting Blood Glucose and Myocardial Infarction Risk in a Population Without Diabetes. Diabetes Care. 2017;40(11):1565–72. doi: 10.2337/dc17-0610 [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Niiranen TJ, Enserro DM, Larson MG, Vasan RS. Multisystem Trajectories Over the Adult Life Course and Relations to Cardiovascular Disease and Death. J Gerontol A Biol Sci Med Sci. 2019;74(11):1778–85. doi: 10.1093/gerona/gly249 [DOI] [PMC free article] [PubMed] [Google Scholar]
22.Kessler EL, Rivaud MR, Vos MA, van Veen TAB. Sex-specific influence on cardiac structural remodeling and therapy in cardiovascular disease. Biol Sex Differ. 2019;10(1):7. doi: 10.1186/s13293-019-0223-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Dayimu A, Wang C, Li J, Fan B, Ji X, Zhang T, et al. Trajectories of Lipids Profile and Incident Cardiovascular Disease Risk: A Longitudinal Cohort Study. J Am Heart Assoc. 2019;8(21):e013479. doi: 10.1161/JAHA.119.013479 [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Holle R, Happich M, Lowel H, Wichmann HE, Group MKS. KORA—a research platform for population based health research. Gesundheitswesen. 2005;67 Suppl 1:S19–25. doi: 10.1055/s-2005-858235 [DOI] [PubMed] [Google Scholar]
25.Laxy M, Knoll G, Schunk M, Meisinger C, Huth C, Holle R. Quality of Diabetes Care in Germany Improved from 2000 to 2007 to 2014, but Improvements Diminished since 2007. Evidence from the Population-Based KORA Studies. PLoS One. 2016;11(10):e0164704. doi: 10.1371/journal.pone.0164704 [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Lorbeer R, Rospleszcz S, Schlett CL, Heber SD, Machann J, Thorand B, et al. Correlation of MRI-derived adipose tissue measurements and anthropometric markers with prevalent hypertension in the community. J Hypertens. 2018;36(7):1555–62. doi: 10.1097/HJH.0000000000001741 [DOI] [PubMed] [Google Scholar]
27.Rospleszcz S, Lorbeer R, Storz C, Schlett CL, Meisinger C, Thorand B, et al. Association of longitudinal risk profile trajectory clusters with adipose tissue depots measured by magnetic resonance imaging. Sci Rep. 2019;9(1):16972. doi: 10.1038/s41598-019-53546-y [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Ruckert IM, Heier M, Rathmann W, Baumeister SE, Doring A, Meisinger C. Association between markers of fatty liver disease and impaired glucose regulation in men and women from the general population: the KORA-F4-study. PLoS One. 2011;6(8):e22932. doi: 10.1371/journal.pone.0022932 [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Sujana C, Seissler J, Jordan J, Rathmann W, Koenig W, Roden M, et al. Associations of cardiac stress biomarkers with incident type 2 diabetes and changes in glucose metabolism: KORA F4/FF4 study. Cardiovasc Diabetol. 2020;19(1):178. doi: 10.1186/s12933-020-01117-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Meisinger C, Lowel H, Thorand B, Doring A. Leisure time physical activity and the risk of type 2 diabetes in men and women from the general population. The MONICA/KORA Augsburg Cohort Study. Diabetologia. 2005;48(1):27–34. doi: 10.1007/s00125-004-1604-3 [DOI] [PubMed] [Google Scholar]
31.Genolini C, Falissard B. KmL: k-means for longitudinal data. Computational Statistics. 2009;25(2):317–28. doi: 10.1007/s00180-009-0178-4 [DOI] [Google Scholar]
32.Genolini C, Alacoque X, Sentenac M, Arnaud C. kml and kml3d: R Packages to Cluster Longitudinal Data. J Stat Softw. 2015;65(4):1–34. doi: 10.18637/jss.v065.i04 [DOI] [Google Scholar]
33.WHO (World Health Organization). Waist Circumference and Waist–Hip Ratio: Report of a WHO Expert Consultation Geneva, 8–11 December 2008 Geneva: WHO; 2008. Available from: https://www.who.int/publications/i/item/9789241501491. [Google Scholar]
34.Whelton PK, Carey RM, Aronow WS, Casey DE Jr., Collins KJ, Dennison Himmelfarb C, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2018;71(19):e127–e248. doi: 10.1016/j.jacc.2017.11.006 [DOI] [PubMed] [Google Scholar]
35.Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486–97. doi: 10.1001/jama.285.19.2486 [DOI] [PubMed] [Google Scholar]
36.GBD Risk Factors Collaborators. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1223–49. doi: 10.1016/S0140-6736(20)30752-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Lorbeer R, Rospleszcz S, Schlett CL, Rado SD, Thorand B, Meisinger C, et al. Association of antecedent cardiovascular risk factor levels and trajectories with cardiovascular magnetic resonance-derived cardiac function and structure. J Cardiovasc Magn Reson. 2021;23(1):2. doi: 10.1186/s12968-020-00698-w [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Allen NB, Siddique J, Wilkins JT, Shay C, Lewis CE, Goff DC, et al. Blood pressure trajectories in early adulthood and subclinical atherosclerosis in middle age. JAMA. 2014;311(5):490–7. doi: 10.1001/jama.2013.285122 [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Tielemans SM, Geleijnse JM, Menotti A, Boshuizen HC, Soedamah-Muthu SS, Jacobs DR Jr., et al. Ten-year blood pressure trajectories, cardiovascular mortality, and life years lost in 2 extinction cohorts: the Minnesota Business and Professional Men Study and the Zutphen Study. J Am Heart Assoc. 2015;4(3):e001378. doi: 10.1161/JAHA.114.001378 [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Schachter M. Changes in the usage of antihypertensive drugs: implications and prospects. Br J Clin Pharmacol. 2005;60(3):231–4. doi: 10.1111/j.1365-2125.2005.02421.x [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Smitson CC, Scherzer R, Shlipak MG, Psaty BM, Newman AB, Sarnak MJ, et al. Association of Blood Pressure Trajectory With Mortality, Incident Cardiovascular Disease, and Heart Failure in the Cardiovascular Health Study. Am J Hypertens. 2017;30(6):587–93. doi: 10.1093/ajh/hpx028 [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Ford ES, Capewell S. Trends in total and low-density lipoprotein cholesterol among U.S. adults: contributions of changes in dietary fat intake and use of cholesterol-lowering medications. PLoS One. 2013;8(5):e65228. doi: 10.1371/journal.pone.0065228 [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Hulsegge G, Spijkerman AM, van der Schouw YT, Bakker SJ, Gansevoort RT, Smit HA, et al. Trajectories of Metabolic Risk Factors and Biochemical Markers prior to the Onset of Cardiovascular Disease—The Doetinchem Cohort Study. PLoS One. 2016;11(5):e0155978. doi: 10.1371/journal.pone.0155978 [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Lee YC, Chien KL, Lee BC, Lin HJ, Hsu HC, Chen MF. High-Density Lipoprotein-Cholesterol Trajectory Pattern, Associated Lifestyle and Biochemical Factors Among Taiwanese. Circ J. 2009;73(10):1887–92. doi: 10.1253/circj.cj-09-0209 [DOI] [PubMed] [Google Scholar]
45.Palmisano BT, Zhu L, Eckel RH, Stafford JM. Sex differences in lipid and lipoprotein metabolism. Mol Metab. 2018;15:45–55. doi: 10.1016/j.molmet.2018.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Verghese D, Muller L, Velamakanni S. Addressing Cardiovascular Risk Across the Arc of a Woman’s Life: Sex-Specific Prevention and Treatment. Curr Cardiol Rep. 2023;25(9):1053–64. doi: 10.1007/s11886-023-01923-5 [DOI] [PubMed] [Google Scholar]
47.Connelly PJ, Currie G, Delles C. Sex Differences in the Prevalence, Outcomes and Management of Hypertension. Curr Hypertens Rep. 2022;24(6):185–92. doi: 10.1007/s11906-022-01183-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Wilmes N, van Luik EM, Vaes EWP, Vesseur MAM, Laven S, Mohseni-Alsalhi Z, et al. Exploring Sex Differences of Beta-Blockers in the Treatment of Hypertension: A Systematic Review and Meta-Analysis. Biomedicines. 2023;11(5). doi: 10.3390/biomedicines11051494 [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Nanna MG, Wang TY, Xiang Q, Goldberg AC, Robinson JG, Roger VL, et al. Sex Differences in the Use of Statins in Community Practice. Circ Cardiovasc Qual Outcomes. 2019;12(8):e005562. doi: 10.1161/CIRCOUTCOMES.118.005562 [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Maas AH, Appelman YE. Gender differences in coronary heart disease. Neth Heart J. 2010;18(12):598–602. doi: 10.1007/s12471-010-0841-y [DOI] [PMC free article] [PubMed] [Google Scholar]
51.Brennan P, Perola M, van Ommen GJ, Riboli E, European Cohort C. Chronic disease research in Europe and the need for integrated population cohorts. Eur J Epidemiol. 2017;32(9):741–9. doi: 10.1007/s10654-017-0315-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Kautzky-Willer A, Leutner M, Harreiter J. Sex differences in type 2 diabetes. Diabetologia. 2023;66(6):986–1002. doi: 10.1007/s00125-023-05891-x [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Connelly PJ, Azizi Z, Alipour P, Delles C, Pilote L, Raparelli V. The Importance of Gender to Understand Sex Differences in Cardiovascular Disease. Can J Cardiol. 2021;37(5):699–710. doi: 10.1016/j.cjca.2021.02.005 [DOI] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0300966.r001

Decision Letter 0

Peng Gao

16 Jan 2024

PONE-D-23-38330Clusters of longitudinal risk profile trajectories are associated with cardiometabolic diseases: results from the population-based KORA cohortPLOS ONE

Dear Dr. Rospleszcz,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Mar 01 2024 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: https://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Peng Gao, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

3. In the online submission form, you indicated that "The datasets analyzed during the current study are not publicly available due national data protection laws, since the informed consent given by KORA study participants does not cover data posting in public databases. Data are available upon request by means of a project agreement from KORA. Requests should be sent to kora.passt@helmholtz-muenchen.de and are subject to approval by the KORA Board. Analysis codes are available from the authors upon reasonable request."

All PLOS journals now require all data underlying the findings described in their manuscript to be freely available to other researchers, either 1. In a public repository, 2. Within the manuscript itself, or 3. Uploaded as supplementary information.

This policy applies to all data except where public deposition would breach compliance with the protocol approved by your research ethics board. If your data cannot be made publicly available for ethical or legal reasons (e.g., public availability would compromise patient privacy), please explain your reasons on resubmission and your exemption request will be escalated for approval.

4. Your ethics statement should only appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please ensure that your ethics statement is included in your manuscript, as the ethics statement entered into the online submission form will not be published alongside your manuscript.

5. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

6. Please review your reference list to ensure that it is complete and correct. If you have cited papers that have been retracted, please include the rationale for doing so in the manuscript text, or remove these references and replace them with relevant current references. Any changes to the reference list should be mentioned in the rebuttal letter that accompanies your revised manuscript. If you need to cite a retracted article, indicate the article’s retracted status in the References list and also include a citation and full reference for the retraction notice.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: General comments:

This is a very interesting manuscript dealing with a topic of importance in the field of cardiology.

Specific comments:

Methods, Study sample, last paragraph, Page 5: the authors should state the population that was accounted for the analysis e.g. participants with the 3 exams and endpoint at follow-up, and from there indicating the excluded participants.

Methods, association of trajectory clusters with cardiometabolic events, Page 8: the authors do not state the population on which they performed kaplan meïer methods which may be different than participants with 3 exams.

Methods, association of trajectory clusters with cardiometabolic events, Page 8: in a more general aspect definition of clusters of trajectory should be made clear for each analysis e.g. prevalent cardiometabolic events, Kaplan-Meïer analysis and incident cardiometabolic events. This is not clear in the methods.

Results, Study sample, Page 8: Authors may also give antihypertensive consumption in the analysed population during the exams, along with decreasing mean levels of blood pressure.

Results, Characterisation of longitunal trajectory clusters, Page 9: a better description of the clusters would be nice to have with a charaterisation of each (what is it exactly?) instead of only interpreting values comparatively. The high risk cluster seem to have a different behavior for DBP, BMI, HBA1c, Total and LDL cholesterol both in emn and women and this should be mentionned. Is it also possible to validate the cluster against established and published severity values of the different risk factors?

Results: I as not able to find results of the Kaplan Meïer except sensitivity analyses in appendix.

Reviewer #2: In this manuscript, the authors explored the effects of joint longitudinal trajectories of multiple risk factors in cardiometabolic events from the population-based cohort. The single time-point measure of metabolic parameters provides limit evidences in predicting cardiovascular events, while the assessment of cumulative effects of metabolic disorders are prominent in predicting the clinical outcomes. This manuscript is well written and the conclusion is convincing. While there are some concerns needed to be solved before the publication.

1. The “nine cardiometabolic risk factors” in abstract should be described.

2. The population were divided into three cluster according to Calinski-Harabasz criterion. Therefore, the clinical characteristics should be provided and compared among these clusters.

3. The data were analyzed in the sex-specific trajectory clusters, while the difference between the cardiometabolic events were solely attributed to sex hormones in the Discussion.

4. The discussion should be more precise.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Michel Vaillant, PhD

Reviewer #2: No

**********

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2024 Mar 28;19(3):e0300966. doi: 10.1371/journal.pone.0300966.r002

Author response to Decision Letter 0

21 Feb 2024

Reviewer 1

This is a very interesting manuscript dealing with a topic of importance in the field of cardiology.

Answer: We thank the reviewer for the positive evaluation of our work and for the detailed comments. We have addressed each comment and made the respective changes in the manuscript and feel the manuscript has improved.

Comment: 1) Methods, Study sample, last paragraph, Page 5: the authors should state the population that was accounted for the analysis e.g. participants with the 3 exams and endpoint at follow-up, and from there indicating the excluded participants.

Answer: We thank the reviewer for raising this point. We rephrased the last paragraph in the methods section “study sample“ and now start with the sample soze of the final analytical sample.

Comment: 2) Methods, association of trajectory clusters with cardiometabolic events, Page 8: the authors do not state the population on which they performed kaplan meïer methods which may be different than participants with 3 exams.

Answer: We thank the reviewer for this observation. For the Kaplan-Meier analyses, we used the same sample as for the cluster analysis, except for the exclusion of n=72 prevalent cases at baseline. We have now clarified this in the respective paragraph in the Methods, section “Association of trajectory clusters with cardiometabolic events”. This sample comprised all participants with complete data at all three exams. We did not consider mortality as an event between Exam 1 and Exam 3, as described in the Methods.

Comment: 3) Methods, association of trajectory clusters with cardiometabolic events, Page 8: in a more general aspect definition of clusters of trajectory should be made clear for each analysis e.g. prevalent cardiometabolic events, Kaplan-Meïer analysis and incident cardiometabolic events. This is not clear in the methods.

Answer: We apologize for the lack of clarity. We have now included a diagram showing the structure and sequence of the analyses for cluster derivation, KM curves, and the regression models in Supplementary Figure S2, and added an overview paragraph to the Statistical Analysis section. In addition, we expanded the legend of Table 2 and Figures 4, S4 and S6 to include information on the sample size used. Briefly, we used the final sample of n = 976 men and n = 1,004 women to derive clusters. For the visualization of event-free survival by KM curves and for the quantification of prevalent events at Exam 3 by logistic regression models, we excluded n = 72 participants with prevalent events at Exam 1, resulting in a sample of n = 926 men and n = 982 women. For quantification of incident events at Exam 3 by logistic regression models, we excluded n = 292 participants with prevalent events at Exam 3, resulting in a sample of n = 806 men and n = 892 women. We have added a description of this in the Methods section, “Association of trajectory clusters with cardiometabolic events”.

Comment: 4) Results, Study sample, Page 8: Authors may also give antihypertensive consumption in the analysed population during the exams, along with decreasing mean levels of blood pressure.

Answer: We thank the reviewer for this observation. We have added this information to the corresponding description of the blood pressure reductions in the section Results, “Study sample”.

Comment: 5) Results, Characterisation of longitunal trajectory clusters, Page 9: a better description of the clusters would be nice to have with a charaterisation of each (what is it exactly?) instead of only interpreting values comparatively. The high risk cluster seem to have a different behavior for DBP, BMI, HBA1c, Total and LDL cholesterol both in emn and women and this should be mentionned. Is it also possible to validate the cluster against established and published severity values of the different risk factors?

Answer: The reviewer mentions an important point, also observed by Reviewer #2, Comment 2). We have now substantially rewritten the results section on “Characterization of longitudinal trajectory clusters” to better describe the individual clusters for men and women. We now also include a more through description of sex-specific risk factor patterns. Furthermore, we have added two Supplementary Tables (S5 and S6 Table) showing mean risk factor values according to cluster membership, corresponding to Figures 1 and 2 (line charts). Furthermore, we have incorporated the relation with established thresholds for disease diagnoses, e.g., hypertension, obesity, dyslipidemia in the section Results, “Characterization of longitudinal trajectory clusters”.

Comment: 6) Results: I as not able to find results of the Kaplan Meïer except sensitivity analyses in appendix.

Answer: The KM curves were supposed to be Figure 4 in the main manuscript. We apologize that this was not visible in the first submission, we hope this issue is now fixed.

Reviewer 2

Comment: In this manuscript, the authors explored the effects of joint longitudinal trajectories of multiple risk factors in cardiometabolic events from the population-based cohort. The single time-point measure of metabolic parameters provides limit evidences in predicting cardiovascular events, while the assessment of cumulative effects of metabolic disorders are prominent in predicting the clinical outcomes. This manuscript is well written and the conclusion is convincing. While there are some concerns needed to be solved before the publication.

Comment: 1. The “nine cardiometabolic risk factors” in abstract should be described.

Answer: We have added the respective information to the abstract.

Comment: 2. The population were divided into three cluster according to Calinski-Harabasz criterion. Therefore, the clinical characteristics should be provided and compared among these clusters.

Answer: The reviewer mentions an important point, also observed by Reviewer #1, Comment 5). We have now substantially re-worked the results section on “Characterization of longitudinal trajectory clusters” to better describe the individual clusters for men and women. Furthermore, we have added two Supplementary Tables (S5 and S6 Table) with the mean risk factor values according to cluster membership, corresponding to Figures 1 and 2 (line charts). We related the mean risk factor levels to established thresholds for disease diagnoses, e.g., hypertension, obesity, dyslipidemia.

Comment: 3. The data were analyzed in the sex-specific trajectory clusters, while the difference between the cardiometabolic events were solely attributed to sex hormones in the Discussion.

Answer: We thank the reviewer for raising this point. We have now added a more thorough discussion of potential reasons for sex-specific differences, including differences in treatment, therapy adherence, underrepresentation in clinical trials, etc., in the respective discussion section.

Comment: 4. The discussion should be more precise.

Answer: We have substantially re-worked the discussion section. Besides incorporating the points mentioned above, we have rephrased, summarized, and shortened parts of the discussion to make it more precise.

Attachment

Submitted filename: ResponseLetter_Niedermayer_PLOSOne_Revision.docx

pone.0300966.s002.docx^{(24.1KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0300966.r003

Decision Letter 1

Peng Gao

7 Mar 2024

Clusters of longitudinal risk profile trajectories are associated with cardiometabolic diseases: results from the population-based KORA cohort

PONE-D-23-38330R1

Dear Dr. Rospleszcz,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice will be generated when your article is formally accepted. Please note, if your institution has a publishing partnership with PLOS and your article meets the relevant criteria, all or part of your publication costs will be covered. Please make sure your user information is up-to-date by logging into Editorial Manager at http://www.editorialmanager.com/pone/ and clicking the ‘Update My Information' link at the top of the page. If you have any questions relating to publication charges, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Peng Gao, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

Reviewer #1: No

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Reviewer #1: All previous comments were addressed. The added precisions and supplementary materials are sufficient for the reader to fully understand and follow the work in the article. There is no further comment.

Reviewer #2: The maniscript is well-written and the results are further interpreted. The concerns have been clearly addressed and i am statisficed with this revised version.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Michel Vaillant

Reviewer #2: No

**********

PLoS One. doi: 10.1371/journal.pone.0300966.r004

Acceptance letter

Peng Gao

15 Mar 2024

PONE-D-23-38330R1

PLOS ONE

Dear Dr. Rospleszcz,

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now being handed over to our production team.

At this stage, our production department will prepare your paper for publication. This includes ensuring the following:

* All references, tables, and figures are properly cited

* All relevant supporting information is included in the manuscript submission,

* There are no issues that prevent the paper from being properly typeset

If revisions are needed, the production department will contact you directly to resolve them. If no revisions are needed, you will receive an email when the publication date has been set. At this time, we do not offer pre-publication proofs to authors during production of the accepted work. Please keep in mind that we are working through a large volume of accepted articles, so please give us a few weeks to review your paper and let you know the next and final steps.

Lastly, if your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

If we can help with anything else, please email us at customercare@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Professor Peng Gao

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 File

(DOCX)

pone.0300966.s001.docx^{(1.2MB, docx)}

Attachment

Submitted filename: ResponseLetter_Niedermayer_PLOSOne_Revision.docx

pone.0300966.s002.docx^{(24.1KB, docx)}

Data Availability Statement

[pone.0300966.ref001] 1.Khan MAB, Hashim MJ, King JK, Govender RD, Mustafa H, Al Kaabi J. Epidemiology of Type 2 Diabetes—Global Burden of Disease and Forecasted Trends. J Epidemiol Glob Health. 2020;10(1):107–11. doi: 10.2991/jegh.k.191028.001 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref002] 2.Roth GA, Mensah GA, Fuster V. The Global Burden of Cardiovascular Diseases and Risks: A Compass for Global Action. J Am Coll Cardiol. 2020;76(25):2980–1. doi: 10.1016/j.jacc.2020.11.021 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref003] 3.Gerdts E, Regitz-Zagrosek V. Sex differences in cardiometabolic disorders. Nat Med. 2019;25(11):1657–66. doi: 10.1038/s41591-019-0643-8 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref004] 4.Mosca L, Barrett-Connor E, Wenger NK. Sex/gender differences in cardiovascular disease prevention: what a difference a decade makes. Circulation. 2011;124(19):2145–54. doi: 10.1161/CIRCULATIONAHA.110.968792 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref005] 5.D’Agostino RB, Sr., Vasan RS, Pencina MJ, Wolf PA, Cobain M, Massaro JM, et al. General cardiovascular risk profile for use in primary care: the Framingham Heart Study. Circulation. 2008;117(6):743–53. doi: 10.1161/CIRCULATIONAHA.107.699579 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref006] 6.Lindstrom J, Tuomilehto J. The diabetes risk score: a practical tool to predict type 2 diabetes risk. Diabetes Care. 2003;26(3):725–31. doi: 10.2337/diacare.26.3.725 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref007] 7.Paprott R, Muhlenbruch K, Mensink GB, Thiele S, Schulze MB, Scheidt-Nave C, et al. Validation of the German Diabetes Risk Score among the general adult population: findings from the German Health Interview and Examination Surveys. BMJ Open Diabetes Res Care. 2016;4(1):e000280. doi: 10.1136/bmjdrc-2016-000280 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref008] 8.Guo F, Garvey WT. Trends in Cardiovascular Health Metrics in Obese Adults: National Health and Nutrition Examination Survey (NHANES), 1988–2014. J Am Heart Assoc. 2016;5(7). doi: 10.1161/JAHA.116.003619 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref009] 9.Hardoon SL, Whincup PH, Lennon LT, Wannamethee SG, Capewell S, Morris RW. How much of the recent decline in the incidence of myocardial infarction in British men can be explained by changes in cardiovascular risk factors? Evidence from a prospective population-based study. Circulation. 2008;117(5):598–604. doi: 10.1161/CIRCULATIONAHA.107.705947 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref010] 10.Allen N, Berry JD, Ning H, Van Horn L, Dyer A, Lloyd-Jones DM. Impact of blood pressure and blood pressure change during middle age on the remaining lifetime risk for cardiovascular disease: the cardiovascular lifetime risk pooling project. Circulation. 2012;125(1):37–44. doi: 10.1161/CIRCULATIONAHA.110.002774 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref011] 11.Reges O, Ning H, Wilkins JT, Wu CO, Tian X, Domanski MJ, et al. Association of Cumulative Systolic Blood Pressure With Long-Term Risk of Cardiovascular Disease and Healthy Longevity: Findings From the Lifetime Risk Pooling Project Cohorts. Hypertension. 2021;77(2):347–56. doi: 10.1161/HYPERTENSIONAHA.120.15650 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref012] 12.Tsao CW, Preis SR, Peloso GM, Hwang SJ, Kathiresan S, Fox CS, et al. Relations of long-term and contemporary lipid levels and lipid genetic risk scores with coronary artery calcium in the framingham heart study. J Am Coll Cardiol. 2012;60(23):2364–71. doi: 10.1016/j.jacc.2012.09.007 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref013] 13.Warren Liao T. Clustering of time series data—a survey. Pattern Recognition. 2005;38(11):1857–74. doi: 10.1016/j.patcog.2005.01.025 [DOI] [Google Scholar]

[pone.0300966.ref014] 14.Nagin DS, Tremblay RE. Developmental trajectory groups: fact or a useful statistical fiction? Criminology. 2005;43(4). doi: 10.1111/j.1745-9125.2005.00026.x [DOI] [Google Scholar]

[pone.0300966.ref015] 15.Govender P, Sivakumar V. Application of k-means and hierarchical clustering techniques for analysis of air pollution: A review (1980–2019). Atmos Pollut Res. 2020;11(1):40–56. doi: 10.1016/j.apr.2019.09.009 [DOI] [Google Scholar]

[pone.0300966.ref016] 16.Duncan MS, Vasan RS, Xanthakis V. Trajectories of Blood Lipid Concentrations Over the Adult Life Course and Risk of Cardiovascular Disease and All-Cause Mortality: Observations From the Framingham Study Over 35 Years. J Am Heart Assoc. 2019;8(11):e011433. doi: 10.1161/JAHA.118.011433 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref017] 17.Rahman F, Yin X, Larson MG, Ellinor PT, Lubitz SA, Vasan RS, et al. Trajectories of Risk Factors and Risk of New-Onset Atrial Fibrillation in the Framingham Heart Study. Hypertension. 2016;68(3):597–605. doi: 10.1161/HYPERTENSIONAHA.116.07683 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref018] 18.Yacaman-Mendez D, Trolle-Lagerros Y, Zhou M, Monteiro Ponce de Leon A, Gudjonsdottir H, Tynelius P, et al. Life-course trajectories of weight and their impact on the incidence of type 2 diabetes. Sci Rep. 2021;11(1):12494. doi: 10.1038/s41598-021-91910-z [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref019] 19.Ogata S, Watanabe M, Kokubo Y, Higashiyama A, Nakao YM, Takegami M, et al. Longitudinal Trajectories of Fasting Plasma Glucose and Risks of Cardiovascular Diseases in Middle Age to Elderly People Within the General Japanese Population: The Suita Study. J Am Heart Assoc. 2019;8(3):e010628. doi: 10.1161/JAHA.118.010628 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref020] 20.Jin C, Chen S, Vaidya A, Wu Y, Wu Z, Hu FB, et al. Longitudinal Change in Fasting Blood Glucose and Myocardial Infarction Risk in a Population Without Diabetes. Diabetes Care. 2017;40(11):1565–72. doi: 10.2337/dc17-0610 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref021] 21.Niiranen TJ, Enserro DM, Larson MG, Vasan RS. Multisystem Trajectories Over the Adult Life Course and Relations to Cardiovascular Disease and Death. J Gerontol A Biol Sci Med Sci. 2019;74(11):1778–85. doi: 10.1093/gerona/gly249 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref022] 22.Kessler EL, Rivaud MR, Vos MA, van Veen TAB. Sex-specific influence on cardiac structural remodeling and therapy in cardiovascular disease. Biol Sex Differ. 2019;10(1):7. doi: 10.1186/s13293-019-0223-0 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref023] 23.Dayimu A, Wang C, Li J, Fan B, Ji X, Zhang T, et al. Trajectories of Lipids Profile and Incident Cardiovascular Disease Risk: A Longitudinal Cohort Study. J Am Heart Assoc. 2019;8(21):e013479. doi: 10.1161/JAHA.119.013479 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref024] 24.Holle R, Happich M, Lowel H, Wichmann HE, Group MKS. KORA—a research platform for population based health research. Gesundheitswesen. 2005;67 Suppl 1:S19–25. doi: 10.1055/s-2005-858235 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref025] 25.Laxy M, Knoll G, Schunk M, Meisinger C, Huth C, Holle R. Quality of Diabetes Care in Germany Improved from 2000 to 2007 to 2014, but Improvements Diminished since 2007. Evidence from the Population-Based KORA Studies. PLoS One. 2016;11(10):e0164704. doi: 10.1371/journal.pone.0164704 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref026] 26.Lorbeer R, Rospleszcz S, Schlett CL, Heber SD, Machann J, Thorand B, et al. Correlation of MRI-derived adipose tissue measurements and anthropometric markers with prevalent hypertension in the community. J Hypertens. 2018;36(7):1555–62. doi: 10.1097/HJH.0000000000001741 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref027] 27.Rospleszcz S, Lorbeer R, Storz C, Schlett CL, Meisinger C, Thorand B, et al. Association of longitudinal risk profile trajectory clusters with adipose tissue depots measured by magnetic resonance imaging. Sci Rep. 2019;9(1):16972. doi: 10.1038/s41598-019-53546-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref028] 28.Ruckert IM, Heier M, Rathmann W, Baumeister SE, Doring A, Meisinger C. Association between markers of fatty liver disease and impaired glucose regulation in men and women from the general population: the KORA-F4-study. PLoS One. 2011;6(8):e22932. doi: 10.1371/journal.pone.0022932 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref029] 29.Sujana C, Seissler J, Jordan J, Rathmann W, Koenig W, Roden M, et al. Associations of cardiac stress biomarkers with incident type 2 diabetes and changes in glucose metabolism: KORA F4/FF4 study. Cardiovasc Diabetol. 2020;19(1):178. doi: 10.1186/s12933-020-01117-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref030] 30.Meisinger C, Lowel H, Thorand B, Doring A. Leisure time physical activity and the risk of type 2 diabetes in men and women from the general population. The MONICA/KORA Augsburg Cohort Study. Diabetologia. 2005;48(1):27–34. doi: 10.1007/s00125-004-1604-3 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref031] 31.Genolini C, Falissard B. KmL: k-means for longitudinal data. Computational Statistics. 2009;25(2):317–28. doi: 10.1007/s00180-009-0178-4 [DOI] [Google Scholar]

[pone.0300966.ref032] 32.Genolini C, Alacoque X, Sentenac M, Arnaud C. kml and kml3d: R Packages to Cluster Longitudinal Data. J Stat Softw. 2015;65(4):1–34. doi: 10.18637/jss.v065.i04 [DOI] [Google Scholar]

[pone.0300966.ref033] 33.WHO (World Health Organization). Waist Circumference and Waist–Hip Ratio: Report of a WHO Expert Consultation Geneva, 8–11 December 2008 Geneva: WHO; 2008. Available from: https://www.who.int/publications/i/item/9789241501491. [Google Scholar]

[pone.0300966.ref034] 34.Whelton PK, Carey RM, Aronow WS, Casey DE Jr., Collins KJ, Dennison Himmelfarb C, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. J Am Coll Cardiol. 2018;71(19):e127–e248. doi: 10.1016/j.jacc.2017.11.006 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref035] 35.Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001;285(19):2486–97. doi: 10.1001/jama.285.19.2486 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref036] 36.GBD Risk Factors Collaborators. Global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396(10258):1223–49. doi: 10.1016/S0140-6736(20)30752-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref037] 37.Lorbeer R, Rospleszcz S, Schlett CL, Rado SD, Thorand B, Meisinger C, et al. Association of antecedent cardiovascular risk factor levels and trajectories with cardiovascular magnetic resonance-derived cardiac function and structure. J Cardiovasc Magn Reson. 2021;23(1):2. doi: 10.1186/s12968-020-00698-w [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref038] 38.Allen NB, Siddique J, Wilkins JT, Shay C, Lewis CE, Goff DC, et al. Blood pressure trajectories in early adulthood and subclinical atherosclerosis in middle age. JAMA. 2014;311(5):490–7. doi: 10.1001/jama.2013.285122 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref039] 39.Tielemans SM, Geleijnse JM, Menotti A, Boshuizen HC, Soedamah-Muthu SS, Jacobs DR Jr., et al. Ten-year blood pressure trajectories, cardiovascular mortality, and life years lost in 2 extinction cohorts: the Minnesota Business and Professional Men Study and the Zutphen Study. J Am Heart Assoc. 2015;4(3):e001378. doi: 10.1161/JAHA.114.001378 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref040] 40.Schachter M. Changes in the usage of antihypertensive drugs: implications and prospects. Br J Clin Pharmacol. 2005;60(3):231–4. doi: 10.1111/j.1365-2125.2005.02421.x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref041] 41.Smitson CC, Scherzer R, Shlipak MG, Psaty BM, Newman AB, Sarnak MJ, et al. Association of Blood Pressure Trajectory With Mortality, Incident Cardiovascular Disease, and Heart Failure in the Cardiovascular Health Study. Am J Hypertens. 2017;30(6):587–93. doi: 10.1093/ajh/hpx028 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref042] 42.Ford ES, Capewell S. Trends in total and low-density lipoprotein cholesterol among U.S. adults: contributions of changes in dietary fat intake and use of cholesterol-lowering medications. PLoS One. 2013;8(5):e65228. doi: 10.1371/journal.pone.0065228 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref043] 43.Hulsegge G, Spijkerman AM, van der Schouw YT, Bakker SJ, Gansevoort RT, Smit HA, et al. Trajectories of Metabolic Risk Factors and Biochemical Markers prior to the Onset of Cardiovascular Disease—The Doetinchem Cohort Study. PLoS One. 2016;11(5):e0155978. doi: 10.1371/journal.pone.0155978 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref044] 44.Lee YC, Chien KL, Lee BC, Lin HJ, Hsu HC, Chen MF. High-Density Lipoprotein-Cholesterol Trajectory Pattern, Associated Lifestyle and Biochemical Factors Among Taiwanese. Circ J. 2009;73(10):1887–92. doi: 10.1253/circj.cj-09-0209 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref045] 45.Palmisano BT, Zhu L, Eckel RH, Stafford JM. Sex differences in lipid and lipoprotein metabolism. Mol Metab. 2018;15:45–55. doi: 10.1016/j.molmet.2018.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref046] 46.Verghese D, Muller L, Velamakanni S. Addressing Cardiovascular Risk Across the Arc of a Woman’s Life: Sex-Specific Prevention and Treatment. Curr Cardiol Rep. 2023;25(9):1053–64. doi: 10.1007/s11886-023-01923-5 [DOI] [PubMed] [Google Scholar]

[pone.0300966.ref047] 47.Connelly PJ, Currie G, Delles C. Sex Differences in the Prevalence, Outcomes and Management of Hypertension. Curr Hypertens Rep. 2022;24(6):185–92. doi: 10.1007/s11906-022-01183-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref048] 48.Wilmes N, van Luik EM, Vaes EWP, Vesseur MAM, Laven S, Mohseni-Alsalhi Z, et al. Exploring Sex Differences of Beta-Blockers in the Treatment of Hypertension: A Systematic Review and Meta-Analysis. Biomedicines. 2023;11(5). doi: 10.3390/biomedicines11051494 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref049] 49.Nanna MG, Wang TY, Xiang Q, Goldberg AC, Robinson JG, Roger VL, et al. Sex Differences in the Use of Statins in Community Practice. Circ Cardiovasc Qual Outcomes. 2019;12(8):e005562. doi: 10.1161/CIRCOUTCOMES.118.005562 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref050] 50.Maas AH, Appelman YE. Gender differences in coronary heart disease. Neth Heart J. 2010;18(12):598–602. doi: 10.1007/s12471-010-0841-y [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref051] 51.Brennan P, Perola M, van Ommen GJ, Riboli E, European Cohort C. Chronic disease research in Europe and the need for integrated population cohorts. Eur J Epidemiol. 2017;32(9):741–9. doi: 10.1007/s10654-017-0315-2 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref052] 52.Kautzky-Willer A, Leutner M, Harreiter J. Sex differences in type 2 diabetes. Diabetologia. 2023;66(6):986–1002. doi: 10.1007/s00125-023-05891-x [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0300966.ref053] 53.Connelly PJ, Azizi Z, Alipour P, Delles C, Pilote L, Raparelli V. The Importance of Gender to Understand Sex Differences in Cardiovascular Disease. Can J Cardiol. 2021;37(5):699–710. doi: 10.1016/j.cjca.2021.02.005 [DOI] [PubMed] [Google Scholar]

PERMALINK

Clusters of longitudinal risk profile trajectories are associated with cardiometabolic diseases: Results from the population-based KORA cohort

Fiona Niedermayer

Gunther Schauberger

Wolfgang Rathmann

Stefanie J Klug

Barbara Thorand

Annette Peters

Susanne Rospleszcz

Roles

Abstract

Background

Methods

Results

Conclusions

Introduction

Materials and methods

Study sample

Risk factor assessment

Outcome assessment

Statistical analysis

Descriptive statistics

Identification of longitudinal trajectory clusters

Association of trajectory clusters with cardiometabolic events

Results

Study sample

Table 1. Characteristics of male and female participants at Exam 1, Exam 2 and Exam 3.

Characterization of longitudinal trajectory clusters

Fig 1. Risk factor trajectories according to cluster membership of men.

Fig 2. Risk factor trajectories according to cluster membership of women.

Association between trajectory clusters and cardiometabolic events

Fig 3. Distribution of prevalent and incident cardiometabolic cases according to cluster.

Fig 4. Kaplan-Meier curves showing cardiometabolic event-free survival stratified by cluster.

Table 2. Association between cluster membership and cardiometabolic events assessed by logistic regression models.

Discussion

Conclusions

Supporting information

Acknowledgments

List of abbreviations

Data Availability

Funding Statement

References

Decision Letter 0

Peng Gao

Roles

Author response to Decision Letter 0

Decision Letter 1

Peng Gao

Roles

Acceptance letter

Peng Gao

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases