Associations of Fitness, Fatness Indices and Fit–Fat Index Variants With Cardiovascular and All‐Cause Mortality in Men

ABSTRACT

Background

Fit–Fat Index (FFI), defined as the ratio of cardiorespiratory fitness (CRF) to waist‐to‐height ratio (WHtR), combines measures of fitness and fatness, and may offer a more accurate assessment of cardiometabolic risk than either component alone. The prospective associations of CRF, fatness indices, and mortality were examined.

Methods

CRF and fatness indices (body mass index [BMI], waist‐to‐hip ratio [WHR], and WHtR) were assessed in 1089 men aged 42–61 years. FFI variants (FFI_BMI, FFI_WHR, and FFI_WHtR) were calculated by dividing CRF by each corresponding fatness measure. Hazard ratios (HRs) with 95% confidence intervals (CIs) and risk prediction metrics were estimated.

Results

Over a median follow‐up of 31.2 years, 288 CVD deaths and 695 all‐cause deaths were recorded. After multivariable adjustment, the HR (95% CI) for CVD death per 1 SD increase was 0.81 (0.70–0.93) for CRF, 1.09 (0.96–1.23) for BMI, 1.13 (1.02–1.25) for WHR, 1.13 (0.99–1.30) for WHtR, 0.82 (0.71–0.95) for FFI_BMI, 0.78 (0.68–0.91) for FFI_WHR, and 0.80 (0.69–0.93) for FFI_WHtR. Findings were similar for all‐cause mortality. Addition of CRF and FFI variants to a model containing established risk factors significantly improved CVD mortality risk prediction and reclassification (p‐value for all < 0.05).

Conclusions

Combining aerobic fitness and fatness into FFI measures may provide better risk stratification than either aerobic fitness or fatness alone.

1. Introduction

Cardiovascular disease (CVD) remains a leading cause of death globally, accounting for over 17 million deaths annually [1]. Major behavioral risk factors for CVD include physical inactivity, unhealthy diet, and harmful use of alcohol and tobacco [1]. The effects of these risk factors may manifest in individuals as high blood pressure, high blood glucose, overweight and obesity. Among modifiable risk factors, physical inactivity and obesity are particularly notable for their strong associations with CVD and all‐cause mortality [2, 3, 4, 5]. Early identification of individuals or populations at an increased risk of CVD is critical to the implementation of timely preventive or therapeutic strategies to reduce premature deaths and alleviate associated health and economic burdens.

Cardiorespiratory fitness (CRF), a physiological indicator of physical activity, is a well‐established marker of cardiovascular and pulmonary function and is considered the gold standard for assessing aerobic capacity [6, 7, 8]. Commonly used measures of fatness to assess overweight and obesity include body mass index (BMI), waist‐to‐height ratio (WHtR), and waist‐hip ratio (WHR). Both measures of aerobic fitness and fatness have been independently associated with the risk of CVD death and all‐cause mortality [9, 10, 11, 12]. Whereas CRF is a strong predictor of these outcomes [13], measures of fatness, such as higher WHR, have been linked to an increased risk [11, 12]. Therefore, integrating objective assessments of both CRF and fatness may provide a more comprehensive approach to identifying individuals at increased risk of cardiovascular and all‐cause mortality.

The concept of Fit‐Fat index (FFI), defined as the ratio of CRF to WHtR, was first introduced in 2016 as a potential predictor of incident diabetes risk [14]. Given that diabetes is a major risk factor for CVD [15], the utility of FFI may extend to predicting CVD and mortality risk. To our knowledge, only one study based on data from the National Health and Nutrition Examination Survey (NHANES) in the United States [16], has evaluated the predictive ability of FFI for all‐cause and CVD‐specific mortality. However, this study estimated CRF using a non‐exercise‐based method and considered only WHtR as the fatness measure. Few other studies have also explored the association of FFI with CVD risk factors, such as heart rate variability (HRV) [17], C‐reactive protein (CRP) [18], and recently sudden cardiac death [19], and demonstrated FFI to be a stronger predictor than measures of CRF or fatness alone.

Given the limited prior research on FFI in relation to CVD and all‐cause mortality, particularly in European settings, the present analysis examined these associations in a large prospective European cohort. In this study, CRF was measured objectively, and multiple indicators of fatness (BMI, WHR, and WHtR) were used to calculate FFI, resulting in the development of several FFI variants.

2. Methods

2.1. Study Design and Participants

The Kuopio Ischemic Heart Disease risk factor study (KIHD) is a population‐based prospective cohort study that was designed to investigate the risk factors for developing atherosclerotic CVDs and related outcomes in a representative sample of people living within Kuopio and its surrounding communities in Eastern Finland. The baseline examination was carried out among 2682 men (82.9% of the 3235 randomly selected and invited participants) between 1984 and 1989 [20, 21]. For the current analysis, participants with a prevalent history of CVD were excluded, leaving 1089 men with complete information on the exposure variables, relevant covariates, and outcomes. The KIHD research protocol was approved by the Research Ethics Committee of the University of Eastern Finland (ref. number 143/97), and the study protocol complied with the ethical guidelines of the Declaration of Helsinki. All participants gave informed consent.

2.2. Assessment of Exposures and Covariates

The assessment of clinical characteristics, risk markers and physical examination has been described previously [22, 23]. Briefly, weight, height, waist and hip circumferences (WC and HC) measured at baseline were used to calculate BMI, WHR, and WHtR. BMI was calculated by dividing the weight (kg) by the square of the height (m²); WHR was the ratio between WC and HC; and WHtR was the ratio between the WC and height. Cardiorespiratory fitness, measured by peak oxygen uptake (VO2peak in mL/kg/minute), and converted to METS ((1 MET = 3.5 mL/kg/min)), was assessed using respiratory gas exchange analyzers (Medical Graphics, MCG, St. Paul, Minnesota) during progressive cycle ergometer exercise testing to voluntary fatigue. The FFI variants were generated from the following ratios: FFI_BMI = CRF/BMI; FFI_WHR = CRF/WHR; and FFI_WHtR = CRF/WHtR. Blood pressure measurements were taken in the morning after a supine rest of 5 min using a random‐zero mercury sphygmomanometer (Hawskley, UK). A detailed questionnaire was used to assess the smoking status, alcohol intake, physical activity, socioeconomic status (SES) and history of chronic diseases [24]. A validated KIHD 12‐month leisure‐time physical activity questionnaire was used to assess the energy expenditure of physical activity. Adult SES was done using self‐reported questionnaires based on combined measures of income, occupational prestige, education, material standard of living, and housing conditions. Blood samples were taken between 8 and 10 in the morning after 12 h of fasting and 3 days of abstinence from alcohol and smoking. The cholesterol contents of serum lipoprotein fractions and triglycerides were measured enzymatically (CHOD‐PAP, Boehringer, Mannheim, Germany). Fasting plasma glucose (FPG) levels were measured using fresh samples using the glucose dehydrogenase method (Merck, Darmstadt, Germany) after protein precipitation with trichloroacetic acid. Prevalent type 2 diabetes (T2D) was identified based on WHO guideline FPG levels (≥ 7.0 mmol/L) [25] or a clinical diagnosis of T2D requiring dietary, oral, or insulin treatment. A history of coronary heart disease (CHD) was defined by previous myocardial infarction, angina pectoris, regular nitro‐glycerine use for chest discomfort at least once a week, or reported chest pain.

2.3. Ascertainment of Follow‐Up Events

All CVD and all‐cause deaths that occurred from baseline through 2022 were included. All KIHD participants are under continuous annual monitoring using Finnish personal identification codes, and data for incident outcomes, including CVD deaths and all‐cause mortality, are collected through the Finnish cause‐of death registry (the Statistics Finland Dnro TK/782/07.03.00/2021). CVD deaths were coded according to cause of death related to the ICD‐9 (International Classification of Diseases, Ninth Revision, Code numbers 390–459) or the ICD‐10 (code numbers I00–I99). Censoring was carried out on the date from the baseline visit to CVD death, death, or the end of the observation period (31 December 2022).

2.4. Statistical Analysis

Baseline characteristics were presented as means (standard deviation, SD) or median (interquartile range, IQR) for continuous variables and counts (percentages) for categorical variables. Hazard ratios (HRs) with 95% confidence intervals (CIs) for cardiovascular and all‐cause mortality were estimated using Cox proportional hazard models after confirming no major violations of the assumptions of proportionality of hazards using Schoenfeld residuals [26]. Two models were used for covariate adjustment: (Model 1) age and (Model 2) Model 1 plus smoking status, history of T2D, total cholesterol, high‐density lipoprotein cholesterol (HDL‐C), systolic blood pressure (SBP), alcohol consumption, SES, and total physical activity. These covariates were chosen based on their established roles as risk factors for adverse cardiovascular outcomes, previous associations with these outcomes in the KIHD study [27, 28, 29], and their potential as confounders given known associations with these outcomes and observed associations with the exposures using available data [30].

To compare the predictive abilities of the various measures of fitness, fatness and FFI for CVD mortality, measures of discrimination for censored time‐to‐event data (Harrell's C‐index) were calculated [31] and reclassified [32, 33]. To investigate the change in C‐index on the addition of each measure, two CVD mortality risk prediction models were fitted: one based on a model containing established risk factors (i.e., age, smoking status, history of T2D, total cholesterol, HDL‐C, SBP, alcohol consumption, SES, and total physical activity) and the second model with these risk factors plus each measure separately. Changes in the C‐index for models including and not including information on each measure were conducted according to the methodology of DeLong et al. [34] and with the Stata command “somersd”. The 95% CIs for C‐indices and their changes were derived from jackknife standard error. Second, the continuous net reclassification improvement (NRI) was computed [33]. Additionally, the integrated discrimination improvement (IDI) was calculated, which integrates the NRI over all possible cut‐offs [32].

Given that Harrell's C‐index is based on ranks rather than on continuous data, the measure can be insensitive in detecting differences [35, 36]. To avoid discarding potential biomarkers relevant for risk prediction, the use of more sensitive risk discrimination methods, such as the −2 log likelihood test, has been recommended [35, 36]. Therefore, in addition to Harrel's C‐index, differences in the −2 log likelihood of prediction models with and without inclusion of each exposure were tested. All statistical analyses were conducted using Stata version MP 18 (Stata Corp, College Station, Texas).

3. Results

3.1. Baseline Characteristics

Table 1 shows baseline characteristics of the 1089 study participants. The overall mean (SD) age was 51 (6) years. The mean (SD) CRF, BMI, WHR, WHtR, FFI_BMI, FFI_WHR, and FFI_WHtR was 9.39 (2.2) METs, 26.7 (3.4) kg/m², 0.94 (0.06), 0.52 (0.06), 0.36 (0.11) METs/kg/m², 10.04 (2.69) METs, and 18.39 (5.44) METs, respectively.

TABLE 1.

Baseline characteristics.

Variable	Overall (N = 1089) Mean (SD) or median (IQR)
Exposures
CRF (METs)	9.39 (2.22)
Body mass index (kg/m2)	26.7 (3.4)
Waist‐to‐hip ratio	0.94 (0.06)
Waist‐to‐height ratio	0.52 (0.06)
FFIBMI (METs/kg/m2)	0.36 (0.11)
FFIWHR (METs)	10.04 (2.69)
FFIWHtR (METs)	18.39 (5.44)
Questionnaire/prevalent conditions
Age (years)	51 (6)
Socioeconomic status	11.1 (4.9)
Alcohol (g/week)	35.6 (6.9, 93.1)
Physical activity (kj/day)	1231 (668, 1947)
Current smokers, n (%)	333 (30.6%)
History of T2D, n (%)	42 (3.9%)
Physical measurements
Systolic blood pressure (mmHg)	133 (16)
Diastolic blood pressure (mmHg)	88 (10)
Blood biomarkers
Total cholesterol (mmol/L)	5.79 (1.00)
HDL cholesterol (mmol/L)	1.31 (0.30)

Abbreviations: CRF, cardiorespiratory fitness; FFI_BMI, Fit‐Fat Index calculated body mass index; FFI_WHR, Fit‐Fat Index calculated waist‐to‐hip ratio; FFI_WHtR, Fit‐Fat Index calculated waist‐to‐height ratio; HDL, high‐density lipoprotein; T2D, type 2 diabetes.

3.2. Associations of Measures of Fit, Fatness and Fit–Fat Index With Outcomes

During a median (IQR) follow‐up of 31.2 (22.6–34.9) years, there were 288 CVD deaths and 695 all‐cause deaths. Estimates of the associations of measures of fit, fatness and FFI with cardiovascular and all‐cause mortality are reported in Table 2. In analysis adjusted for age, the HR (95% CI) for CVD mortality per 1 SD increase in each exposure was 0.71 (0.62–0.81) for CRF, 1.20 (1.07–1.34) for BMI, 1.20 (1.11–1.31) for WHR, 1.29 (1.14–1.45) for WHtR, 0.71 (0.62–0.82) for FFI_BMI, 0.68 (0.59–0.78) for FFI_WHR, and 0.69 (0.60–0.79) for FFI_WHtR. On further adjustment for smoking status, history of T2D, total cholesterol, HDL‐C, SBP, alcohol consumption, SES, and total physical activity, the HRs (95% CI) were attenuated to 0.81 (0.70–0.93) for CRF, 1.09 (0.96–1.23) for BMI, 1.13 (1.02–1.25) for WHR, 1.13 (0.99–1.30) for WHtR, 0.82 (0.71–0.95) for FFI_BMI, 0.78 (0.68–0.91) for FFI_WHR, and 0.80 (0.69–0.93) for FFI_WHtR.

TABLE 2.

Associations of measures of fit, fatness and fit–fat Index with CVD and all‐cause mortality.

Exposures	CVD mortality					All‐cause mortality
	Events/total	Model 1		Model 2		Events/total	Model 1		Model 2
		HR (95% CI)	p‐value	HR (95% CI)	p‐value		HR (95% CI)	p‐value	HR (95% CI)	p‐value
CRF (METs)
Per 1 SD increase	288/1089	0.71 (0.62–0.81)	< 0.001	0.81 (0.70–0.93)	0.003	695/1089	0.71 (0.65–0.78)	< 0.001	0.77 (0.71–0.85)	< 0.001
Tertile 1 (3.0–8.0)	130/392	ref		ref		308/392	ref		ref
Tertile 2 (8.1–10.0)	109/382	0.83 (0.64–1.07)	0.15	0.96 (0.74–1.25)	0.75	249/382	0.78 (0.66–0.93)	0.004	0.86 (0.72–1.02)	0.092
Tertile 3 (≥ 10.1)	49/315	0.42 (0.30–0.59)	< 0.001	0.54 (0.37–0.77)	0.001	138/315	0.48 (0.39–0.60)	< 0.001	0.57 (0.46–0.71)	< 0.001
BMI (kg/m2)
Per 1 SD increase	288/1089	1.20 (1.07–1.34)	0.002	1.09 (0.96–1.23)	0.19	695/1089	1.13 (1.05–1.21)	0.001	1.08 (0.99–1.17)	0.073
Tertile 1 (19.0–24.9)	84/363	ref		ref		210/363	ref		ref
Tertile 2 (25.0–27.6)	96/363	1.08 (0.81–1.45)	0.59	1.02 (0.76–1.37)	0.91	227/363	1.03 (0.86–1.25)	0.73	1.03 (0.85–1.25)	0.77
Tertile 3 (≥ 27.7)	108/363	1.39 (1.04–1.85)	0.025	1.09 (0.80–1.49)	0.57	258/363	1.33 (1.11–1.60)	0.002	1.21 (1.00–1.48)	0.053
WHR
Per 1 SD increase	288/1089	1.20 (1.11–1.31)	< 0.001	1.13 (1.02–1.25)	0.017	695/1089	1.14 (1.07–1.21)	< 0.001	1.08 (1.01–1.16)	0.023
Tertile 1 (0.71–0.91)	70/363	ref		ref		199/363	ref		ref
Tertile 2 (0.92–0.96)	97/363	1.34 (0.99–1.82)	0.062	1.15 (0.84–1.58)	0.37	229/363	1.12 (0.93–1.35)	0.25	1.03 (0.85–1.25)	0.77
Tertile 3 (≥ 0.97)	121/363	1.97 (1.47–2.65)	< 0.001	1.49 (1.09–2.05)	0.013	267/363	1.52 (1.27–1.83)	< 0.001	1.29 (1.06–1.57)	0.012
WHtR
Per 1 SD increase	288/1089	1.29 (1.14–1.45)	< 0.001	1.13 (0.99–1.30)	0.079	695/1089	1.19 (1.10–1.29)	< 0.001	1.11 (1.01–1.21)	0.027
Tertile 1 (0.38–0.48)	73/363	ref		ref		195/363	ref		ref
Tertile 2 (0.49–0.53)	101/364	1.22 (0.90–1.65)	0.20	1.05 (0.77–1.43)	0.78	233/364	1.07 (0.88–1.29)	0.50	1.00 (0.82–1.22)	0.99
Tertile 3 (≥ 0.54)	114/362	1.61 (1.20–2.16)	0.002	1.19 (0.86–1.65)	0.28	267/362	1.43 (1.18–1.72)	< 0.001	1.23 (1.01–1.51)	0.043
FFIBMI (MET/kg/m2)
Per 1 SD increase	288/1089	0.71 (0.62–0.82)	< 0.001	0.82 (0.71–0.95)	0.007	695/1089	0.73 (0.67–0.79)	< 0.001	0.79 (0.72–0.86)	< 0.001
Tertile 1 (0.08–0.30)	121/363	ref		ref		282/363	ref		ref
Tertile 2 (0.31–0.39)	104/363	0.78 (0.60–1.01)	0.060	0.93 (0.72–1.22)	0.61	242/363	0.77 (0.65–0.92)	0.003	0.86 (0.73–1.03)	0.10
Tertile 3 (≥ 0.40)	63/363	0.45 (0.33–0.62)	< 0.001	0.58 (0.41–0.80)	0.001	171/363	0.52 (0.42–0.63)	< 0.001	0.59 (0.48–0.73)	< 0.001
FFIWHR (METs)
Per 1 SD increase	288/1089	0.68 (0.59–0.78)	< 0.001	0.78 (0.68–0.91)	0.001	695/1089	0.70 (0.64–0.77)	< 0.001	0.77 (0.70–0.84)	< 0.001
Tertile 1 (2.86–8.72)	120/363	ref		ref		288/363	ref		ref
Tertile 2 (8.73–10.92)	113/363	0.86 (0.66–1.11)	0.25	1.03 (0.79–1.34)	0.82	239/363	0.75 (0.63–0.89)	0.001	0.84 (0.70–1.00)	0.048
Tertile 3 (≥ 10.93)	55/363	0.40 (0.29–0.56)	< 0.001	0.53 (0.37–0.75)	< 0.001	168/363	0.50 (0.41–0.61)	< 0.001	0.61 (0.49–0.75)	< 0.001
FFIWHtR (METs)
Per 1 SD increase	288/1089	0.69 (0.60–0.79)	< 0.001	0.80 (0.69–0.93)	0.003	695/1089	0.72 (0.66–0.78)	< 0.001	0.78 (0.71–0.86)	< 0.001
Tertile 1 (4.43–15.73)	121/363	ref		ref		287/363	ref		ref
Tertile 2 (15.74–20.16)	114/363	0.88 (0.68–1.14)	0.32	1.04 (0.80–1.35)	0.78	245/363	0.79 (0.66–0.93)	0.006	0.87 (0.73–1.03)	0.11
Tertile 3 (≥ 20.17)	53/363	0.39 (0.28–0.54)	< 0.001	0.51 (0.36–0.73)	< 0.001	163/363	0.49 (0.40–0.60)	< 0.001	0.59 (0.47–0.72)	< 0.001

Note: Model 1: Adjusted for age. Model 2: Model 1 plus smoking status, history of type 2 diabetes, total cholesterol, high‐density lipoprotein cholesterol, systolic blood pressure, alcohol consumption, socioeconomic status, and total physical activity.

Abbreviations: CI, confidence interval; CRF, cardiorespiratory fitness; FFI_BMI, Fit‐Fat Index calculated body mass index; FFI_WHR, Fit‐Fat Index calculated waist‐to‐hip ratio; FFI_WHtR, Fit‐Fat Index calculated waist‐to‐height ratio; HR, hazard ratio; ref, reference; SD, standard deviation.

The age‐adjusted HR (95% CI) for all‐cause mortality per 1 SD increase in each exposure was 0.71 (0.65–0.78) for CRF, 1.13 (1.05–1.21) for BMI, 1.14 (1.07–1.21) for WHR, 1.19 (1.10–1.29) for WHtR, 0.73 (0.67–0.79) for FFI_BMI, 0.70 (0.64–0.77) for FFI_WHR, and 0.72 (0.66–0.78) for FFI_WHtR. Following additional adjustment for smoking status, history of T2D, total cholesterol, HDL‐C, SBP, alcohol consumption, SES, and total physical activity, the HRs (95% CI) were attenuated to 0.77 (0.71–0.85) for CRF, 1.08 (0.99–1.17) for BMI, 1.08 (1.01–1.16) for WHR, 1.11 (1.01–1.21) for WHtR, 0.79 (0.72–0.86) for FFI_BMI, 0.77 (0.70–0.84) for FFI_WHR, and 0.78 (0.71–0.86) for FFI_WHtR.

The associations were qualitatively similar when the exposures were modeled as tertiles (Table 2).

3.3. Fit, Fatness and Fit–Fat Index and CVD Mortality Risk Prediction

Results of risk prediction analyses are presented in Table 3. The results are reported for CRF, WHR, FFI_BMI, FFI_WHR, and FFI_WHtR because of their independent associations with CVD mortality. The CVD mortality risk prediction model containing established risk factors yielded a C‐index of 0.7354 (95% CI, 0.7057, 0.7651). There was a non‐significant increase in C‐index on addition of information on the exposures: 0.0046 (95% CI, −0.0013, 0.0105) for CRF, 0.0019 (95% CI, −0.0018, 0.0057) for WHR, 0.0036 (95% CI, −0.0020, 0.0091) for FFI_BMI, 0.0057 (95% CI, −0.0010, 0.0125) for FFI_WHR, and 0.0047 (95% CI, −0.0014, 0.0109) for FFI_WHtR. The −2 log likelihood was significantly improved on addition of each measure (p for comparison for all < 0.05). Except for WHR, the continuous NRI and IDI were improved for all the other measures (p‐value for all < 0.05).

TABLE 3.

Measures of risk discrimination and reclassification upon addition of CRF, WHR and FFI variants to a CVD mortality risk model containing conventional risk factors.

	CRF	WHR	FFIBMI	FFIWHR	FFIWHtR
Discrimination
C‐index (95% CI): Conventional risk factors	0.7354 (0.7057, 0.7651)	0.7354 (0.7057, 0.7651)	0.7354 (0.7057, 0.7651)	0.7354 (0.7057, 0.7651)	0.7354 (0.7057, 0.7651)
C‐index (95% CI): Conventional risk factors plus exposure	0.7400 (0.7107, 0.7692)	0.7373 (0.7077, 0.7669)	0.7389 (0.706, 0.7683)	0.7411 (0.7119, 0.7703)	0.7401 (0.7108, 0.7694)
C‐index change (95% CI)	0.0046 (−0.0013, 0.0105)	0.0019 (−0.0018, 0.0057)	0.0036 (−0.0020, 0.0091)	0.0057 (−0.0010, 0.0125)	0.0047 (−0.0014, 0.0109)
p‐value	0.13	0.31	0.21	0.095	0.13
p‐value for difference in −2 log likelihood	0.003	0.025	0.006	< 0.001	0.003
Reclassification
Continuous net reclassification index (95% CI)	41.04% (6.82, 75.26)	2.76% (−22.13, 27.65)	47.03% (14.65, 79.41)	47.78% (15.69, 79.87)	56.21% (26.42, 86.00)
p‐value	0.019	0.83	0.004	0.004	< 0.001
Integrated discrimination index (95% CI)	0.0072 (0.0017, 0.0128)	−0.0053 (−0.0093, −0.0013)	0.0065 (0.0012, 0.0118)	0.0079 (0.0018, 0.0140)	0.0070 (0.0015, 0.0124)
p‐value	0.011	0.009	0.016	0.012	0.012

Note: Conventional risk factors include age, smoking status, history of type 2 diabetes, total cholesterol, high‐density lipoprotein cholesterol, systolic blood pressure, alcohol consumption, socioeconomic status, and total physical activity.

Abbreviations: CI, confidence interval; CRF, cardiorespiratory fitness; FFI_BMI, Fit‐Fat Index calculated body mass index; FFI_WHR, Fit‐Fat Index calculated waist‐to‐hip ratio; FFI_WHtR, Fit‐Fat Index calculated waist‐to‐height ratio; WHR, waist‐to‐hip ratio.

4. Discussion

In this population‐based study, higher aerobic fitness and FFI measures (FFI_BMI, FFI_WHR, FFI_WHtR) were independently associated with lower risks of cardiovascular and all‐cause mortality. Furthermore, fatness measures, particularly WHR and WHtR, were positively associated with increased risk of all‐cause mortality; among fatness measures, WHR was the most strongly associated with CVD mortality. Moreover, the addition of CRF, WHR and FFI measures to a model containing established risk factors led to non‐significant increases in the C‐index; however, all measures (CRF, WHR and FFI measures) significantly improved the −2 log likelihood (a more sensitive marker of discrimination). The continuous NRI and IDI were improved for all the other measures except for WHR. The use of objective measures, in the current study, to generate FFI and the Finnish study population, extends FFI applicability in Europe, beyond Asian and American populations, for the prediction and risk stratification for CVD death and all‐cause mortality.

The findings of the present study are consistent with those reported in U.S.‐based cohorts, including the NHANES and the Aerobics Center Longitudinal Study (ACLS) [14, 16]. The NHANES study, which investigated similar outcomes to ours, found that the FFI was inversely associated with both CVD‐specific and all‐cause mortality [16]. This aligns with findings from the ACLS, which demonstrated that FFI was a stronger risk indicator of incident diabetes compared with CRF or fatness alone [14]. However, while NHAHES used a nonexercise method to estimate CRF, the ACLS used incident diabetes ‐ a known risk factor of CVD [15], which differs from the outcomes assessed in our study. FFI has also been explored in a limited number of studies assessing its relationship with cardiovascular risk factors, such as the HRV and CRP. The results on FFI and CVD risk factors support our current findings on mortality [17, 18], such that FFI has been reported to be stronger than CRF and fatness alone in its association with HRV [17], while an independent inverse association was demonstrated between FFI and CRP [18]. Heart rate variability and CRP‐ a marker of inflammation [37], are known to be strongly associated with the risk of CVD death and all‐cause mortality [38, 39, 40, 41]. In the Cardiovascular Health Study, individuals with abnormal HRV measures exhibited a higher risk of CVD mortality compared with those with normal HRV measures, irrespective of their risk category [38]. Similarly, lower values of the HRV parameter were associated with increased overall mortality [42] Furthermore, elevations in serum CRP levels were associated with a higher risk of incident CVD [40], as well as CVD‐specific all‐cause mortality [40, 41].

The pathways underlying the associations between FFI and adverse cardiovascular outcomes likely stem from its two key components: aerobic fitness and fatness, particularly overweight and obesity. Aerobic fitness exerts protective effects on arterial health, primarily through its anti‐atherosclerotic properties [43]. Regular physical activity induces recurrent laminar shear stress on the vascular endothelium, enhancing nitric oxide (NO) synthesis, which in turn promotes vasodilation ‐ a key defense mechanism against atherosclerosis and thrombosis in the coronary vessels [43]. Accordingly, higher levels of aerobic fitness are consistently associated with lower risks of both CVD and all‐cause mortality [9, 10, 44, 45]. In contrast, overweight and obesity may contribute to CVD and mortality risk through both direct and indirect mechanisms [46]. While traditionally under‐recognized, the direct effects of excess weight on CVDs are increasingly supported by evidence. Individuals with overweight or obesity often exhibit increased total blood volume and cardiac output, which can lead to structural and functional changes in the heart and vasculature, leading to both systolic and diastolic dysfunction [46]. Additionally, excess weight may impair mobility and physical activity levels, and exacerbate musculoskeletal conditions such as osteoarthritis, further promoting a cycle of inactivity and weight gain, and ultimately increasing cardiovascular risk [46]. Indirectly, obesity contributes to CVD risk through increased adiposity, which drives the development of key cardiometabolic risk factors such as insulin resistance (a hallmark of T2D), dyslipidemia, hypertension, endothelial dysfunction, and atherosclerosis [46, 47]. These pathways collectively support the growing body of evidence linking obesity to elevated risks of CVD‐related and all‐cause mortality [3, 48, 49].

The FFI, by combining objective measures of aerobic fitness and fatness, offers a simple yet powerful tool for identifying individuals at increased risk of CVD and all‐cause mortality. This integrated measure provides a clearer picture of cardiometabolic risk than fitness or fatness alone and may guide personalized intervention strategies. Given evidence that higher CRF can offset some of the risks associated with excess adiposity [42], the most effective approach is to promote both increased aerobic fitness and reduced fatness. Improving both components would raise FFI levels, which, as demonstrated in this study, is associated with lower CVD and all‐cause mortality risk. FFI could be easily implemented in clinical and public health settings to support risk stratification and monitor lifestyle interventions.

4.1. Strengths and Limitations

To our knowledge, this is the first study to evaluate the FFI combining objectively measured aerobic fitness and multiple fatness indicators (BMI, WHR, and WHtR) to assess its association with CVD and all‐cause mortality. The study is further strengthened by its prospective design, relatively large sample size, and long follow‐up period spanning over 3 decades. However, several limitations should be noted. The study population consisted exclusively of White Finnish men, which limits the generalizability of the findings to women, younger individuals, and other racial or ethnic groups. Additionally, the observational nature of the study introduces potential sources of bias, including residual confounding, reverse causality, and regression dilution, which may affect the interpretation of the associations between FFI and mortality outcomes. Although FFI variants are associated with mortality risk, CRF alone demonstrates comparable predictive value. Therefore, the incremental benefit of FFI may be limited and should be considered when interpreting the results.

5. Conclusion

Findings from the present study suggest that combining aerobic fitness and fatness into FFI measures may provide better risk stratification than either aerobic fitness or fatness alone. Thus, CRF and FFI indices should be considered as complementary tools in cardiovascular risk assessment. Further studies in diverse populations are warranted to evaluate FFI potential utility and added value in mortality risk assessment.

Author Contributions

Nzechukwu M. Isiozor: conceptualization, writing – original draft, methodology, project administration, writing – review and editing. Setor K. Kunutsor: writing – original draft, methodology, formal analysis, writing – review and editing. Sudhir Kurl: investigation, writing – review and editing. Kai Savonen: investigation, writing – review and editing. Jussi Kauhanen: investigation, writing – review and editing, project administration. Jari A. Laukkanen: conceptualization, writing – original draft, investigation, project administration, formal analysis, writing – review and editing.

Funding

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Isiozor, Nzechukwu M. , Kunutsor Setor K., Kurl Sudhir, Savonen Kai, Kauhanen Jussi, and Laukkanen Jari A.. 2026. “Associations of Fitness, Fatness Indices and Fit–Fat Index Variants With Cardiovascular and All‐Cause Mortality in Men,” Obesity Science & Practice: e70108. 10.1002/osp4.70108.