Advertisement

Relation Between Self-Reported Physical Activity Level, Fitness, and Cardiometabolic Risk

Published:November 25, 2013DOI:https://doi.org/10.1016/j.amjcard.2013.11.010
      Physical activity and cardiorespiratory fitness are associated with improved cardiovascular health and reduced all-cause mortality. The relation between self-reported physical activity, objective physical fitness, and the association of each with cardiometabolic risk has not been fully described. We studied 2,800 healthy Brazilian subjects referred for an employer-sponsored health screening. Physical activity level was determined as “low,” “moderate,” or “high” with the International Physical Activity Questionnaire: Short Form (IPAQ-SF). Fitness was measured as METs achieved on a maximal, symptom-limited, treadmill stress test. Using multivariate linear regression analysis, we calculated age, gender, and smoking-adjusted correlation coefficients among IPAQ-SF, fitness, and cardiometabolic risk factors. Mean age of study participants was 43 ± 9 years; 81% were men, and 43% were highly active. Mean METs achieved was 12 ± 2. IPAQ-SF category and fitness were moderately correlated (r = 0.377). Compared with IPAQ-SF category, fitness was better correlated with cardiometabolic risk factors including anthropomorphic measurements, blood pressure, fasting blood glucose, dyslipidemia, high-sensitivity C-reactive protein, and hepatic steatosis (all p <0.01). Among these, anthropomorphic measurements, blood pressure, high-sensitivity C-reactive protein, and hepatic steatosis had the largest discrepancies in correlation, whereas lipid factors had the least discrepant correlation. When IPAQ-SF and fitness were discordant, poor fitness drove associations with elevated cardiometabolic risk. In conclusion, self-reported physical activity level and directly measured fitness are moderately correlated, and the latter is more strongly associated with a protective cardiovascular risk profile.
      In this study, we analyzed a population of relatively healthy adult Brazilian men and women to describe (1) the correlation between self-reported physical activity and directly measured fitness, (2) the relative association of self-reported physical activity and measured fitness with modern cardiometabolic risk factors, (3) the characteristics of subjects with discordant self-reported physical activity and measured fitness, and (4) the implication of this discordance on cardiometabolic risk.

      Methods

      The study population consisted of 2,800 Brazilian men and women (age range 19 to 78 years) free of baseline cardiovascular disease who were evaluated as part of an employer-sponsored health examination at the Preventive Medicine Center of the Hospital Israelita Albert Einstein in São Paulo, Brazil, from November 2008 to July 2010. The clinical examination consisted of a medical history questionnaire and comprehensive physical examination by a physician including anthropomorphic measurements of obesity, fasting blood laboratory analysis, and hepatic ultrasound for determination of liver fat. Patients were screened for excessive alcohol consumption using the Alcohol Use Disorders Identification Test (AUDIT).
      • Babor T.F.
      • Higgins-Biddle J.C.
      • Saunders J.B.
      • Monteiro M.G.
      AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Health Care.
      The AUDIT test was developed and validated by the World Health Organization among men and women of different ages and nationalities. A total score of ≥8 is generally accepted as an indicator of hazardous and/or harmful alcohol use.
      Blood specimens were collected after an overnight fast. Laboratory analysis included a standard lipid panel, fasting glucose, aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transpeptidase, and creatinine, all of which were analyzed using a Vitros platform automated laboratory system (Johnson & Johnson Clinical Diagnostics, New Brunswick, New Jersey). High-sensitivity C-reactive protein (hs-CRP) was determined by immunonephelemetry (Dade-Behrin GmbH, Mannheim, Germany). Hepatic steatosis was assessed by ultrasonography after a minimum 6-hour fast using an Acuson XP-10 device (Mountain View, California). Images were read by 2 board-certified radiologists blinded to laboratory test results. The diagnosis of hepatic steatosis was identified as a bright liver with contrast between hepatic and renal parenchyma.
      • Joseph A.E.
      • Saverymuttu S.H.
      • al-Sam S.
      • Cook M.G.
      • Maxwell J.D.
      Comparison of liver histology with ultrasonography in assessing diffuse parenchymal liver disease.
      • Bellentani S.
      • Saccoccio G.
      • Masutti F.
      • Crocè L.S.
      • Brandi G.
      • Sasso F.
      • Cristanini G.
      • Tiribelli C.
      Prevalence of and risk factors for hepatic steatosis in Northern Italy.
      Metabolic syndrome was defined at the time of the baseline examination using the International Diabetes Foundation criteria.
      • Alberti K.G.M.M.
      • Zimmet P.
      • Shaw J.
      Metabolic syndrome-a new world-wide definition. A Consensus Statement from the International Diabetes Federation.
      • Expert Panel on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults (U.S.), National Heart, Lung, and Blood Institute, National Institute of Diabetes and Digestive and Kidney Diseases (U.S.)
      Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: the Evidence Report.
      Dyslipidemia was defined as triglycerides (TGs) ≥150 mg/dl, high-density lipoprotein (HDL) <40 mg/dl for men, and HDL <50 mg/dl for women.
      • Jellinger P.S.
      • Smith D.A.
      • Mehta A.E.
      • Ganda O.
      • Handelsman Y.
      • Rodbard H.W.
      • Shepherd M.D.
      • Seibel J.A.
      American Association of Clinical Endocrinologists' guidelines for management of dyslipidemia and prevention of atherosclerosis.
      Physical activity level was assessed by a physician using the International Physical Activity Questionnaire: Short Form (IPAQ-SF), which was previously validated in a similar patient population.
      • Craig C.L.
      • Marshall A.L.
      • Sjöström M.
      • Bauman A.E.
      • Booth M.L.
      • Ainsworth B.E.
      • Pratt M.
      • Ekelund U.
      • Yngve A.
      • Sallis J.F.
      • Oja P.
      International physical activity questionnaire: 12-country reliability and validity.
      • Hallal P.C.
      • Gomez L.F.
      • Parra D.C.
      • Lobelo F.
      • Mosquera J.
      • Florindo A.A.
      • Reis R.S.
      • Pratt M.
      • Sarmiento O.L.
      Lessons learned after 10 years of IPAQ use in Brazil and Colombia.
      • Sharma S.
      Assessing diet and lifestyle in the Canadian Arctic Inuit and Inuvialuit to inform a nutrition and physical activity intervention programme.
      The IPAQ-SF traditionally categorizes subjects into 1 of 3 distinct qualitative categories of physical activity: high, moderate, and low.

      IPAQ Research Committee. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ)—Short and Long Forms. 2005:1–15.

      According to IPAQ-SF scoring guidelines, high physical activity is equivalent to >1 hour of moderate-intensity activity over and above basal activity or >30 minutes of vigorous-intensity activity above basal levels daily. Moderate activity is defined as 30 minutes of at least moderate-intensity activity on most days of the week, and low activity describes all subjects not meeting the aforementioned criteria.

      IPAQ Research Committee. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ)—Short and Long Forms. 2005:1–15.

      For purposes of continuity with the previous literature, participants were stratified into these 3 categories of physical activity (low, moderate, and high) for the primary analysis of our data.
      Fitness was quantified by the total time achieved on a maximal, symptom-limited, exercise treadmill stress test using either an Ellestad or a standard Bruce protocol.
      • Ellestad M.H.
      • Allen W.
      • Wan M.
      • Kemp G.L.
      Maximal treadmill stress testing for cardiovascular evaluation.
      • McDonough J.R.
      • Bruce R.A.
      Maximal exercise testing in assessing cardiovascular function.
      Stress test duration was converted into METs using previously validated methods of conversion.
      For an Ellestad protocol exercise stress test, the following conversion formula was used for men and women
      • Pollock M.L.
      • Bohannon R.L.
      • Cooper K.H.
      • Ayres J.J.
      • Ward A.
      • White S.R.
      • Linnerud A.C.
      A comparative analysis of four protocols for maximal treadmill stress testing.
      :
      VO2max=(3.933×totaltimeonthetreadmill+4.46)


      For a standard Bruce protocol exercise stress test, the following gender-specific formulas were used to convert total treadmill time achieved to Vo2max
      • Vitola J.V.
      • Kormann O.J.
      • Stier A.L.
      • Chalela W.A.
      • Mastrocolla L.E.
      • Delbeke D.
      Stress modalities to evaluate myocardial perfusion.
      :
      MalesVO2max=(2.9×totaltimeonthetreadmill)+8.33FemalesVO2max=(2.74×totaltimeonthetreadmill)+8.03


      The following formula was used to convert Vo2max to METs for both exercise protocols:
      METs=VO2max/3.515


      Baseline characteristics, including baseline distributions of IPAQ-SF physical activity and fitness, were analyzed by gender. Unadjusted correlation between IPAQ-SF activity level and fitness (both total time achieved and calculated METs) was determined using a Pearson correlation coefficient. Adjusted correlation coefficients were derived from multivariate linear regression models adjusting for age, gender, and smoking status. Additional subgroup analyses were carried out according to baseline gender and obesity status.
      To assess the association of physical activity and fitness with cardiometabolic risk factors, age-, gender-, and smoking-adjusted correlation coefficients were calculated from multivariate linear regression models. Risk factors were considered as continuous variables when possible. Additional analyses were carried out according to gender. Differences in the association between IPAQ-SF and cardiometabolic risk and fitness and cardiometabolic risk were formally tested using standard Fisher r-to-z transformation.
      We then conducted an analysis of study subjects with discordant levels of fitness and self-reported physical activity. A fit/inactive group was defined by an IPAQ-SF category of moderate or low and having achieved ≥11 METs on a treadmill stress test. An unfit/active group was defined by an IPAQ-SF category of high and having achieved <11 METs on a treadmill stress test. A cut point of 11 METs was decided on as this represented the median METs achieved for the study population.
      The prevalence of cardiometabolic risk factors in the 2 discordant groups was compared using a standard chi-square analysis. Multivariate logistic regression models adjusting for age, gender, and smoking status were used to determine odds ratios (ORs) for various cardiometabolic risk factors among discordant groups.

      Results

      Most study subjects were men (81%). Mean age of men was 44 ± 9 years. Mean age of women was 41 ± 9 years. Baseline characteristics of the population are listed according to gender in Table 1.
      Table 1Study population characteristics by gender
      CharacteristicMen, 2,269 (81%)Women, 531 (19%)
      Age (yrs)44 ± 941 ± 9
      Smoker (%)98
      AUDIT alcohol consumption score5 ± 43 ± 2
      Body mass index (kg/m2)27 ± 424 ± 4
      Waist circumference (cm)95 ± 1079 ± 10
      Obesity
      BMI >30 kg/m2 or waist circumference >102 cm in men or >88 cm in women when BMI was 25 to 30 kg/m2.
      (%)
      2415
      Metabolic syndrome
      Defined according to International Diabetes Foundation criteria.4
      (%)
      236
      Systolic blood pressure (mm Hg)121 ± 11110 ± 11
      Diastolic blood pressure (mm Hg)78 ± 771 ± 7
      Hypertension
      Systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥85 mm Hg.
      (%)
      145
      Antihypertensive use (%)135
      Fasting glucose (mg/dl)90 ± 1084 ± 8
      Impaired fasting glucose + diabetes (%)153
      Total cholesterol (mg/dl)207 ± 37197 ± 33
      Low-density lipoprotein cholesterol (mg/dl)133 ± 33117 ± 30
      HDL cholesterol (mg/dl)45 ± 1159 ± 14
      Non-HDL cholesterol (mg/dl)162 ± 37137 ± 33
      TGs (mg/dl)122 (89–172)88 (67–122)
      Statin use (%)103
      Hs-CRP (mg/L)1.2 (0.6–2.1)1.2 (0.7–3.1)
      Hepatic steatosis (%)4210
      Aspartate aminotransferase (units/L)32 ± 1225 ± 6
      Alanine aminotransferase (units/L)41 ± 2224 ± 9
      Gamma-glutamyl transpeptidase (units/L)44 ± 4724 ± 18
      Microalbuminuria (%)51
      10-yr Framingham risk (%)5 ± 51 ± 1
      BMI >30 kg/m2 or waist circumference >102 cm in men or >88 cm in women when BMI was 25 to 30 kg/m2.
      Defined according to International Diabetes Foundation criteria.
      • Alberti K.G.M.M.
      • Zimmet P.
      • Shaw J.
      Metabolic syndrome-a new world-wide definition. A Consensus Statement from the International Diabetes Federation.
      Systolic blood pressure ≥130 mm Hg or diastolic blood pressure ≥85 mm Hg.
      Overall, 23% of participants reported low, 34% moderate, and 43% high physical activity levels according to IPAQ-SF. Similar distributions were noted among men and women (p = 0.130; Figure 1). Men had higher mean levels of fitness (p <0.001; Figure 1). Fitness increased significantly with increasing level of self-reported physical activity for men and women (p <0.001; Table 2).
      Figure thumbnail gr1
      Figure 1(A) Distribution of IPAQ-SF–assessed physical activity by gender. p = 0.130 indicates no significant difference in distribution across physical activity categories. (B) Distribution of fitness by gender. Significant difference in mean METs between men and women (p <0.001).
      Table 2Distribution of fitness stratified by International Physical Activity Questionnaire: Short Form (IPAQ-SF) and gender
      Variablen (%)Mean Stress Test Duration, minMETs Achievedp Value
      Men and women
       Low643 (23)8.9 ± 1.611.1 ± 1.8<0.001
       Moderate960 (34)9.7 ± 1.312.1 ± 2.2
       High1,197 (43)10.8 ± 2.113.3 ± 2.4
      Men
       Low513 (23)9.1 ± 1.611.4 ± 1.8<0.001
       Moderate779 (34)10.1 ± 1.812.5 ± 2.1
       High977 (43)11.1 ± 2.113.6 ± 2.4
      Women
       Low130 (25)8.1 ± 1.410.1 ± 1.6<0.001
       Moderate181 (34)8.3 ± 1.510.3 ± 1.8
       High220 (41)9.3 ± 1.711.6 ± 2.0
      Increasing IPAQ-SF activity level correlated with statistically significant increases in fitness (measured by maximal time achieved on treadmill stress test and calculated METs).
      Correlation of IPAQ-SF activity level with fitness was moderate (unadjusted model: r = 0.356 and 0.360 using stress test duration and calculated METs, respectively, p = 0.865). Adjusting for age, gender, and smoking status, correlations were similar (r = 0.370 and 0.377, respectively, p = 0.764). The correlation of IPAQ-SF with fitness was similar among men and women (for men: r = 0.402 and 0.408 using stress test duration and calculated METs, respectively; for women: r = 0.333 and 0.351, respectively; p = 0.168). Correlation coefficients for obese and nonobese subjects were not statistically different (obese: r = 0.311 and 0.322 using stress test duration and calculated METs, respectively; nonobese r = 0.354 and 0.360, respectively, p = 0.159; Table 3). All individual correlations were statistically significant (p <0.01).
      Table 3Correlation of International Physical Activity Questionnaire: Short Form (IPAQ-SF) category and physical fitness
      Variabler (IPAQ-SF, Stress Test Duration)r (IPAQ-SF, METs)
      Total population (n = 2,800)
       Unadjusted0.3560.360
       Adjusted
      Correlation coefficients adjusted for age, gender, and smoking.
      0.3700.377
      Gender stratified
       Women
      Correlation coefficients adjusted for age and smoking.
      (n = 530)
      0.3330.351
       Men
      Correlation coefficients adjusted for age and smoking.
      (n = 2,265)
      0.4020.408
      Obese patients
      Correlation coefficients adjusted for age, gender, and smoking.
      (n = 625)
      0.3110.322
      Nonobese patients
      Correlation coefficients adjusted for age, gender, and smoking.
      (n = 2,175)
      0.3540.360
      All individual correlation coefficients were significant at p <0.001. Obesity was defined as BMI ≥30 kg/m2 or waist circumference >102 cm in men and >88 cm in women with a BMI ≥25 kg/m.
      Correlation coefficients adjusted for age, gender, and smoking.
      Correlation coefficients adjusted for age and smoking.
      Compared with IPAQ-SF, fitness was more strongly correlated with a healthier profile of cardiovascular risk factors by several traditional and nontraditional measures. The largest absolute difference in correlation coefficients was noted among anthropomorphic risk factors (Table 4). Relatively large differences in correlation coefficients were also noted for hs-CRP and hepatic steatosis. Blood pressure showed large relative differences in correlation, although absolute correlations were very weak.
      Table 4Treadmill fitness and International Physical Activity Questionnaire: Short Form (IPAQ-SF) correlations with cardiometabolic risk factors
      Total PopulationCorrelation Coefficient, (R) Adjusted For Age, Gender, And SmokingMenCorrelation Coefficient (R) Adjusted For Age And SmokingWomenCorrelation Coefficient (R) Adjusted For Age And Smoking
      METsIPAQ-SFp ValueMETsIPAQ-SFp ValueMETsIPAQSFp Value
      AnthropomorphicsAnthropomorphicsAnthropomorphics
       BMI−0.438−0.135<0.001 BMI−0.452−0.149<0.001 BMI−0.392−0.113<0.001
       Waist circumference−0.422−0.184<0.001 Waist circumference−0.519−0.227<0.001 Waist circumference−0.363−0.164<0.001
      Blood pressureBlood pressureBlood pressure
       Systolic blood pressure−0.202−0.058<0.001 Systolic blood pressure−0.211−0.065<0.001 Systolic blood pressure−0.181−0.050<0.001
       Diastolic blood pressure−0.211−0.051<0.001 Diastolic blood pressure−0.221−0.063<0.001 Diastolic blood pressure−0.186−0.018<0.001
      GlycemiaGlycemiaGlycemia
       Log fasting glucose−0.146−0.058<0.001 Log fasting glucose−0.154−0.0710.002 Log fasting glucose−0.091−0.003<0.001
      LipidsLipidsLipids
       Total cholesterol−0.080−0.0810.968 Total cholesterol−0.083−0.0820.968 Total cholesterol−0.078−0.0860.764
       LDL cholesterol−0.044−0.0600.549 LDL cholesterol−0.044−0.0530.734 LDL cholesterol−0.083−0.1070.363
       HDL cholesterol0.1920.1120.002 HDL cholesterol0.2360.1550.002 HDL cholesterol0.1000.0300.009
       Log TGs−0.297−0.171<0.001 Log TGs−0.313−0.209<0.001 Log TGs−0.173−0.023<0.001
       Non-HDL cholesterol−0.142−0.1160.322 Non-HDL cholesterol−0.150−0.1260.363 Non-HDL cholesterol−0.122−0.1000.407
       TC:HDL−0.231−0.1630.008 TC:HDL−0.252−0.1950.025 TC:HDL−0.162−0.072<0.001
       TG:HDL−0.232−0.1590.002 TG:HDL−0.235−0.1820.039 TG:HDL−0.162−0.031<0.001
      Novel risk factorsNovel risk factorsNovel risk factors
       Log hs-CRP (n = 2,337)−0.323−0.130<0.001 Log hs-CRP (n = 1,875)−0.300−0.135<0.001 Log hs-CRP (n = 462)−0.323−0.109<0.001
       Hepatic steatosis−0.345−0.131<0.001 Hepatic steatosis−0.356−0.157<0.001 Hepatic steatosis−0.2270.005<0.001
       Framingham risk score−0.115−0.0610.041 Framingham risk score−0.107−0.0680.142 Framingham risk score−0.0680.0290.144
      Correlation of cardiometabolic risk factors with calculated METs and IPAQ-SF. Comparison of correlation coefficients is reflected in the p value.
      Glycemic and lipid risk factors showed smaller differences in correlation (Table 4). There was no statistical difference in correlation coefficients of total cholesterol, low-density lipoprotein cholesterol, and non-HDL cholesterol. Similar observations regarding differences in correlation were noted among both men and women when analyzed separately.
      Among study participants, 36% were classified as fit but inactive (fit/inactive) and 7% as unfit but active (unfit/active). Compared with those who were fit/inactive, subjects who were unfit/active were older and had higher average systolic blood pressure, diastolic blood pressure, body mass index (BMI), median hs-CRP, and TG:HDL cholesterol. The unfit/active group also had less prevalence of men, a greater prevalence of diabetes, impaired fasting glucose + diabetes, and obesity.
      Multivariate logistic regression between discordant groups demonstrated greater odds of hypertension (OR 2.79, 95% confidence interval [CI] 1.75 to 4.43, p <0.001), elevated hs-CRP (OR 1.90, 95% CI 1.19 to 3.05, p = 0.007), metabolic syndrome (OR 1.76, 95% CI 1.13 to 2.75, p = 0.012), hepatic steatosis (OR 2.04, 95% CI 1.33 to 3.07, p = 0.001), and obesity (OR 2.39, 95% CI 1.56 to 3.68, p <0.001) for unfit/active subjects compared with fit/inactive (Figure 2). In cases of discordant physical activity and fitness, there were no statistically significant differences in the odds of dyslipidemia or impaired fasting glucose + diabetes.
      Figure thumbnail gr2
      Figure 2Adjusted ORs for risk factor conditions in unfit/active versus fit/inactive groups. ORs adjusted for age, gender, and smoking. Reference group is fit/inactive.

      Discussion

      In the present study, we found a moderate correlation between self-reported physical activity determined by IPAQ-SF and objectively measured fitness with exercise treadmill stress testing. Furthermore, fitness correlated significantly better than IPAQ-SF for most cardiometabolic risk factors. When analyzing 2 discordant groups of unfit/active and fit/inactive subjects, we found that fitness correlated better with cardiometabolic risk than did self-reported physical activity.
      Adjusted for age, gender, and smoking status, the correlation of IPAQ-SF with fitness in our study was moderate (r = 0.377). This result concurs with data from previous studies that report mild to moderate correlations between IPAQ-SF and fitness assessed with actometer, accelerometer, pedometer, 6-minute walk test, doubly labeled water, standard Bruce maximal treadmill test, and Vo2max.
      • Hallal P.C.
      • Gomez L.F.
      • Parra D.C.
      • Lobelo F.
      • Mosquera J.
      • Florindo A.A.
      • Reis R.S.
      • Pratt M.
      • Sarmiento O.L.
      Lessons learned after 10 years of IPAQ use in Brazil and Colombia.
      • Lee P.H.
      • Macfarlane D.J.
      • Lam T.H.
      • Stewart S.M.
      Validity of the international physical activity questionnaire short form (IPAQ-SF): a systematic review.
      • Papathanasiou G.
      • Georgoudis G.
      • Georgakopoulos D.
      • Katsouras C.
      • Kalfakakou V.
      • Evangelou A.
      Criterion-related validity of the short International Physical Activity Questionnaire against exercise capacity in young adults.
      Correlations in these studies have ranged from 0.09 to 0.39.
      • Lee P.H.
      • Macfarlane D.J.
      • Lam T.H.
      • Stewart S.M.
      Validity of the international physical activity questionnaire short form (IPAQ-SF): a systematic review.
      The reason for modest correlations between self-reported physical activity and objectively measured fitness is poorly understood, although several factors may play a role including inaccurate reporting of physical activity and varying genotypes conferring adaptive fitness responsiveness to physical activity.
      Given the relatively moderate correlation of fitness with IPAQ-SF, a clinically important question is the relative correlation of each with traditional and nontraditional cardiometabolic risk factors. Our study showed that fitness was more strongly inversely correlated with all cardiometabolic risk factors analyzed, except for total cholesterol, low-density lipoprotein cholesterol, and non-HDL cholesterol. To parse out the impact of obesity on these correlations, we performed a post hoc secondary analysis additionally adjusting for BMI as a continuous variable. This adjustment attenuated the correlation between both self-reported physical activity and METs with cardiometabolic risk factors up to 50%; however, METs remained more highly correlated with cardiometabolic risk factors than self-reported physical activity. Therefore, although obesity attenuates individual correlations, objectively measured fitness still is more strongly inversely correlated with most cardiometabolic risk factors measured.
      The risk factors most strongly correlated with fitness were BMI, waist circumference, hs-CRP, and hepatic steatosis. Risk factors that correlated weakly with fitness included systolic blood pressure, diastolic blood pressure, HDL cholesterol, TGs, TC:HDL, and TG:HDL. These findings are consistent with previous descriptions of an inverse association between fitness and cardiometabolic risk factors (including novel risk factors such as hs-CRP and hepatic steatosis), cardiovascular morbidity, and mortality.
      • Grundy S.M.
      • Barlow C.E.
      • Farrell S.W.
      • Vega G.L.
      • Haskell W.L.
      Cardiorespiratory fitness and metabolic risk.
      • Lee D.C.
      • Sui X.
      • Artero E.G.
      • Lee I.
      • Church T.S.
      • McAuley P.A.
      • Stanford F.C.
      • Kohl H.W.
      • Blair S.N.
      Long-term effects of changes in cardiorespiratory fitness and body mass index on all-cause and cardiovascular disease mortality in men: The Aerobics Center Longitudinal Study.
      • Peterson P.N.
      • Magid D.J.
      • Ross C.
      • Ho P.M.
      • Rumsfeld J.S.
      • Lauer M.S.
      • Lyons E.E.
      • Smith S.S.
      • Masoudi F.A.
      Association of exercise capacity on treadmill with future cardiac events in patients referred for exercise testing.
      • Aronson D.
      • Sella R.
      • Sheikh-Ahmad M.
      • Kerner A.
      • Avizohar O.
      • Rispler S.
      • Bartha P.
      • Markiewicz W.
      • Levy Y.
      • Brook G.J.
      The association between cardiorespiratory fitness and C-reactive protein in subjects with the metabolic syndrome.
      • Vranian M.N.
      • Keenan T.
      • Blaha M.J.
      • Silverman M.G.
      • Michos E.D.
      • Minder C.M.
      • Blumenthal R.S.
      • Nasir K.
      • Meneghelo R.S.
      • Santos R.D.
      Impact of fitness versus obesity on routinely measured cardiometabolic risk in young, healthy adults.
      There was very weak to virtually no correlation between IPAQ-SF and all cardiometabolic risk factors included in our study. These findings suggest enhanced value for using clinical measures of fitness as opposed to self-reported physical activity level for determining cardiometabolic risk. Although the present study measured cardiometabolic risk factors and not hard clinical end points, our results corroborate with previous investigations that found fitness was superior to self-reported physical activity in predicting risk of cardiovascular disease, metabolic syndrome, and mortality.
      • Myers J.
      • Kaykha A.
      • George S.
      • Abella J.
      • Zaheer N.
      • Lear S.
      • Yamazaki T.
      • Froelicher V.
      Fitness versus physical activity patterns in predicting mortality in men.
      • Williams P.T.
      Physical fitness and activity as separate heart disease risk factors: a meta-analysis.
      We found that a large majority (∼83%) of subjects with discordant measures of fitness and physical activity were in fact more fit than their IPAQ-SF physical activity category may suggest. This discrepancy in fit/inactive and unfit/active subjects was unexpected because previous validation studies of IPAQ-SF noted significant overreporting of physical activity up to 173% beyond objectively measured values.
      • Lee P.H.
      • Macfarlane D.J.
      • Lam T.H.
      • Stewart S.M.
      Validity of the international physical activity questionnaire short form (IPAQ-SF): a systematic review.
      • Brown W.
      • Bauman A.
      • Chey T.
      • Trost S.
      • Mummery K.
      Method: comparison of surveys used to measure physical activity.
      The large proportion of fit/inactive subjects in our study may be partially explained by the demographics of our patient population, which was predominantly young, healthy, and largely men. All 3 of these attributes have been associated with increased fitness.
      • American College of Sports Medicine
      ACSM's Guidelines for Exercise Testing and Prescription.
      Accordingly, the fit/inactive group was significantly younger and had a greater proportion of men compared with the unfit/active group.
      In cases when subjects had discordant levels of physical activity and fitness, our analysis suggests that directly measured fitness may have greater cardioprotective effects than self-reported physical activity, as evidenced by increased odds of hypertension, elevated hs-CRP, metabolic syndrome, hepatic steatosis, and obesity in the unfit/active group relative to the fit/inactive group. However, additional adjustment of our discordant analysis for BMI attenuated these observed effects, making all ORs nonstatistically significant, except for hypertension. Although this further analysis is highly limited by power, obesity clearly plays an important role in the discordance comparison.
      Although fitness appeared to trump physical activity level in relation to most cardiovascular risk factors, we did note higher levels of HDL-C in the unfit/active group. This observation contrasts with findings from previous studies demonstrating a positive correlation between fitness level and HDL-C concentration.
      • Ahmed H.M.
      • Blaha M.J.
      • Nasir K.
      • Rivera J.J.
      • Blumenthal R.S.
      Effects of physical activity on cardiovascular disease.
      The higher levels of HDL-C in the unfit/active cohort were likely due in part to the greater proportion of women who are known to have higher baseline HDL-C relative to men.
      • Jellinger P.S.
      • Smith D.A.
      • Mehta A.E.
      • Ganda O.
      • Handelsman Y.
      • Rodbard H.W.
      • Shepherd M.D.
      • Seibel J.A.
      American Association of Clinical Endocrinologists' guidelines for management of dyslipidemia and prevention of atherosclerosis.
      Furthermore, HDL-C levels may have been impacted by chronic alcohol consumption, especially among men (mean AUDIT score of 5 ± 4) or the use of medications to increase HDL-C, which was not controlled for in this study.
      Our study was conducted in healthy subjects, and it is unclear if similar correlations between fitness, physical activity, and cardiometabolic risk factors exist in elderly populations with greater chronic disease burden. Our study population also included a high proportion of fit individuals (median METs = 11 for the entire study population) and relatively few women. Because our population was recruited from an employer-sponsored health examination, most of our subjects are presumed to be gainfully employed and of higher socioeconomic background. Subjects were representative of a narrow geographical region (Brazil) and may not reflect patterns from other parts of the world. Furthermore, our study's cross-sectional design precludes inferences about causality or temporal trends. As a consequence, it is unclear if longitudinal changes in fitness result in corresponding changes in cardiovascular risk or vice versa. Furthermore, it is unclear if longitudinal changes in physical activity translate into corresponding changes in physical fitness.

      Disclosures

      The authors have no conflicts of interest to disclose.

      References

        • Babor T.F.
        • Higgins-Biddle J.C.
        • Saunders J.B.
        • Monteiro M.G.
        AUDIT: The Alcohol Use Disorders Identification Test: Guidelines for Use in Primary Health Care.
        World Health Organization, Department of Mental Health and Substance Dependence, Geneva2001
        • Joseph A.E.
        • Saverymuttu S.H.
        • al-Sam S.
        • Cook M.G.
        • Maxwell J.D.
        Comparison of liver histology with ultrasonography in assessing diffuse parenchymal liver disease.
        Clin Radiol. 1991; 43: 26-31
        • Bellentani S.
        • Saccoccio G.
        • Masutti F.
        • Crocè L.S.
        • Brandi G.
        • Sasso F.
        • Cristanini G.
        • Tiribelli C.
        Prevalence of and risk factors for hepatic steatosis in Northern Italy.
        Ann Intern Med. 2000; 132: 112-117
        • Alberti K.G.M.M.
        • Zimmet P.
        • Shaw J.
        Metabolic syndrome-a new world-wide definition. A Consensus Statement from the International Diabetes Federation.
        Diabet Med. 2006; 23: 469-480
        • Expert Panel on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults (U.S.), National Heart, Lung, and Blood Institute, National Institute of Diabetes and Digestive and Kidney Diseases (U.S.)
        Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults: the Evidence Report.
        National Institutes of Health, National Heart, Lung, and Blood Institute, 1998
        • Jellinger P.S.
        • Smith D.A.
        • Mehta A.E.
        • Ganda O.
        • Handelsman Y.
        • Rodbard H.W.
        • Shepherd M.D.
        • Seibel J.A.
        American Association of Clinical Endocrinologists' guidelines for management of dyslipidemia and prevention of atherosclerosis.
        Endocr Pract. 2012; 18: 1-78
        • Craig C.L.
        • Marshall A.L.
        • Sjöström M.
        • Bauman A.E.
        • Booth M.L.
        • Ainsworth B.E.
        • Pratt M.
        • Ekelund U.
        • Yngve A.
        • Sallis J.F.
        • Oja P.
        International physical activity questionnaire: 12-country reliability and validity.
        Med Sci Sports Exerc. 2003; 35: 1381-1395
        • Hallal P.C.
        • Gomez L.F.
        • Parra D.C.
        • Lobelo F.
        • Mosquera J.
        • Florindo A.A.
        • Reis R.S.
        • Pratt M.
        • Sarmiento O.L.
        Lessons learned after 10 years of IPAQ use in Brazil and Colombia.
        J Phys Act Health. 2010; 7: S259-S264
        • Sharma S.
        Assessing diet and lifestyle in the Canadian Arctic Inuit and Inuvialuit to inform a nutrition and physical activity intervention programme.
        J Hum Nutr Diet. 2010; 23: 5-17
      1. IPAQ Research Committee. Guidelines for Data Processing and Analysis of the International Physical Activity Questionnaire (IPAQ)—Short and Long Forms. 2005:1–15.

        • Ellestad M.H.
        • Allen W.
        • Wan M.
        • Kemp G.L.
        Maximal treadmill stress testing for cardiovascular evaluation.
        Circulation. 1969; 39: 517-522
        • McDonough J.R.
        • Bruce R.A.
        Maximal exercise testing in assessing cardiovascular function.
        J S C Med Assoc. 1969; 65: 25-33
        • Pollock M.L.
        • Bohannon R.L.
        • Cooper K.H.
        • Ayres J.J.
        • Ward A.
        • White S.R.
        • Linnerud A.C.
        A comparative analysis of four protocols for maximal treadmill stress testing.
        Am Heart J. 1976; 92: 39-46
        • Vitola J.V.
        • Kormann O.J.
        • Stier A.L.
        • Chalela W.A.
        • Mastrocolla L.E.
        • Delbeke D.
        Stress modalities to evaluate myocardial perfusion.
        in: Vitola J.V. Delbeke D. Nuclear Cardiology and Correlative Imaging: A Teaching File. Springer, New York2004: 84-120
        • American College of Sports Medicine
        ACSM's Metabolic Calculations Handbook.
        Lippincott Williams & Wilkins, 2007: 25-35
        • Lee P.H.
        • Macfarlane D.J.
        • Lam T.H.
        • Stewart S.M.
        Validity of the international physical activity questionnaire short form (IPAQ-SF): a systematic review.
        Int J Behav Nutr Phys Act. 2011; 8: 115
        • Papathanasiou G.
        • Georgoudis G.
        • Georgakopoulos D.
        • Katsouras C.
        • Kalfakakou V.
        • Evangelou A.
        Criterion-related validity of the short International Physical Activity Questionnaire against exercise capacity in young adults.
        Eur J Cardiovasc Prev Rehabil. 2010; 17: 380-386
        • Grundy S.M.
        • Barlow C.E.
        • Farrell S.W.
        • Vega G.L.
        • Haskell W.L.
        Cardiorespiratory fitness and metabolic risk.
        Am J Cardiol. 2012; 109: 988-999
        • Lee D.C.
        • Sui X.
        • Artero E.G.
        • Lee I.
        • Church T.S.
        • McAuley P.A.
        • Stanford F.C.
        • Kohl H.W.
        • Blair S.N.
        Long-term effects of changes in cardiorespiratory fitness and body mass index on all-cause and cardiovascular disease mortality in men: The Aerobics Center Longitudinal Study.
        Circulation. 2011; 124: 2483-2490
        • Peterson P.N.
        • Magid D.J.
        • Ross C.
        • Ho P.M.
        • Rumsfeld J.S.
        • Lauer M.S.
        • Lyons E.E.
        • Smith S.S.
        • Masoudi F.A.
        Association of exercise capacity on treadmill with future cardiac events in patients referred for exercise testing.
        Arch Intern Med. 2008; 168: 174-179
        • Aronson D.
        • Sella R.
        • Sheikh-Ahmad M.
        • Kerner A.
        • Avizohar O.
        • Rispler S.
        • Bartha P.
        • Markiewicz W.
        • Levy Y.
        • Brook G.J.
        The association between cardiorespiratory fitness and C-reactive protein in subjects with the metabolic syndrome.
        J Am Coll Cardiol. 2004; 44: 2003-2007
        • Vranian M.N.
        • Keenan T.
        • Blaha M.J.
        • Silverman M.G.
        • Michos E.D.
        • Minder C.M.
        • Blumenthal R.S.
        • Nasir K.
        • Meneghelo R.S.
        • Santos R.D.
        Impact of fitness versus obesity on routinely measured cardiometabolic risk in young, healthy adults.
        Am J Cardiol. 2013; 111: 991-995
        • Myers J.
        • Kaykha A.
        • George S.
        • Abella J.
        • Zaheer N.
        • Lear S.
        • Yamazaki T.
        • Froelicher V.
        Fitness versus physical activity patterns in predicting mortality in men.
        Am J Med. 2004; 117: 912-918
        • Williams P.T.
        Physical fitness and activity as separate heart disease risk factors: a meta-analysis.
        Med Sci Sports Exerc. 2001; 33: 754-761
        • Brown W.
        • Bauman A.
        • Chey T.
        • Trost S.
        • Mummery K.
        Method: comparison of surveys used to measure physical activity.
        Aust N Z J Public Health. 2008; 28: 128-134
        • American College of Sports Medicine
        ACSM's Guidelines for Exercise Testing and Prescription.
        Lippincott Williams & Wilkins, 2006
        • Ahmed H.M.
        • Blaha M.J.
        • Nasir K.
        • Rivera J.J.
        • Blumenthal R.S.
        Effects of physical activity on cardiovascular disease.
        Am J Cardiol. 2012; 109: 288-295