Journal Logo

EPIDEMIOLOGY

Physical Activity and Residual-Specific Mortality among Adults in the United States

LOPRINZI, PAUL D.; SNG, EVELEEN; ADDOH, OVUOKERIE

Author Information
Medicine & Science in Sports & Exercise: September 2016 - Volume 48 - Issue 9 - p 1730-1736
doi: 10.1249/MSS.0000000000000952
  • Free

Abstract

The National Center for Health Statistics (NCHS) codes causes of death in 10 broad categories using comparable International Classification of Diseases, 10th Revision (ICD-10)-based groups. These causes include 1) diseases of the heart; 2) malignant neoplasms; 3) chronic lower respiratory diseases; 4) accidents (unintentional injuries); 5) cerebrovascular diseases; 6) Alzheimer’s disease; 7) diabetes mellitus; 8) influenza and pneumonia; 9) nephritis, nephrotic syndrome, and nephrosis; and 10) all other causes (residual). This classification mirrors the leading causes of death in the United States. Examples of residual-specific mortalities include diseases of the blood and immune system, as well as mental health conditions (additional examples are noted in the Discussion section).

Regular participation in physical activity is inversely associated with all-cause mortality risk (8,9,16–18,25). There is also an inverse association between physical activity and cause-specific mortality for the majority of these aforementioned cause-specific deaths, including diseases of the heart (21,33), malignant neoplasms (2,4,25), chronic lower respiratory diseases (4), cerebrovascular diseases (33), and diabetes mellitus (33). In addition, physical activity is inversely associated with mortality risk among those with Alzheimer’s disease (29) and kidney disease (5). Collectively, these findings indicate that physical activity may have a favorable effect on reducing all-cause and cause-specific mortality risk. What has yet to be explored, however, is if physical activity is associated with reduced residual-specific mortality. This is plausible because, for example, physical activity is favorably associated with blood-related conditions (e.g., abnormal red blood cell distribution width) (22) and mental health outcomes (11,19). Thus, the purpose of this study was to examine the association between physical activity and residual-specific mortality among a national sample of adults in the United States. Examining the effect of physical activity on residual-specific mortality is noteworthy of investigation given the relatively large proportion of deaths not attributed to the first nine major causes of death defined in ICD-10.

METHODS

Study design with assessment of mortality status

The National Health and Nutrition Examination Survey (NHANES) is an ongoing survey conducted by the Centers for Disease Control and Prevention designed to evaluate the health status of US adults through a complex, multistage, stratified clustered probability design. Participants are interviewed in their homes and then subsequently examined in a mobile examination center. Further information on NHANES methodology and data collection is available on the NHANES website (http://www.cdc.gov/nchs/nhanes.htm). Procedures were approved by the NCHS review board. Consent was obtained from all participants before data collection.

Participant data from the 1999–2006 National Health and Nutrition Examination Survey were used. Data from participants in these cycles were linked to death certificate data from the National Death Index via a probabilistic algorithm. Person-months of follow-up were calculated from the date of the interview until the date of death or censoring on December 31, 2011, whichever came first. Analyses are based on data from 16,329 adults (20–85 yr) who provided complete data for the study variables; a flow description of the participant sample size after exclusions is described in the Results section.

Physical activity

As described elsewhere (17), participants were asked open-ended questions about participation in leisure time physical activity over the past 30 d. Data were coded into 48 activities, including 16 sports-related activities, 14 exercise-related activities, and 18 recreational-related activities.

For each of the 48 activities where participants reported moderate or vigorous intensity for the respective activity, they were asked to report the number of times they engaged in that activity over the past 30 d and the average duration they engaged in that activity. For each activity, moderate to vigorous physical activity (MVPA) metabolic equivalent of task (MET)-minute-month was calculated by multiplying the number of days by the mean duration and by the respective MET level (MVPA MET-minute-month = days × duration × MET level). The MVPA MET levels for each activity are provided elsewhere (1). Consistent with the government physical activity guidelines (i.e., 500 MVPA MET·min·wk) and identical to other work (17), participants were defined as “active” based on self-reporting ≥2000 MVPA MET·min·month. As described elsewhere, this physical activity assessment has demonstrated evidence of convergent validity by positively associating with accelerometer-assessed physical activity (17).

Further sensitivity analyses evaluated a subsample of NHANES participants from the 2003–2006 cycles who had objectively measured physical activity data. In this sample, we explored the association between accelerometer-assessed MVPA and residual-specific mortality in a sample of 5536 adult participants who provided valid accelerometry data (i.e., wore the monitor for at least four out of the seven monitoring days for at least 10 h·d−1). The MVPA cut-point of 2020 counts per minute was applied to define MVPA (30), with a cut-point of 100–2019 counts per minute used to define light-intensity physical activity (24). Nonwear time was identified as ≥60 consecutive minutes of zero activity counts, with allowance for 1–2 min of activity counts between 0 and 100 (30).

Statistical analysis and covariates

Statistical analyses were performed via procedures from survey data using Stata (v. 12). Analyses accounted for the complex survey design employed in NHANES by using sample weights, primary sampling units, and strata via the Taylor series (linearization) method. Sample weights were reweighted to account for the use of combined NHANES cycles. Briefly, each person in the NHANES dataset is assigned a sample weight. This sample weight is created using three steps: first, the base weight is calculated for each person, which takes into consideration the probability of the participant or his or her county, city block, and household being selected; second, the sample weight is adjusted for nonresponse (i.e., whether they were a nonrespondent to the interview portion and/or the examination portion) and noncoverage (i.e., not sampled in the NHANES population); and third, poststratification adjustment is made to the sample weights to match the 2000 US Census population. Briefly, and before any analyses, the following Stata command is used to define the survey design: svyset [w = weight, psu (psu variable) strata (strata variable)]. Then, “svy” commands are used for each analysis to ensure that the complex survey design of NHANES is accounted for when determining variance estimates.

Weighted multivariable Cox proportional hazards models were used to examine the association between physical activity and residual-specific mortality. This was performed by first using the “stset” command followed by the “stcox” command. Schoenfeld’s residuals, using the “estat phtest” command, were used to verify the proportional hazards assumption. The Harrell’s C concordance statistic for the Cox proportional hazards model was estimated using the “estat concordance” command. In addition to examining the hazards for the different physical activity classification groups, population attributable fraction (PAF) and its corresponding 95% confidence interval (CI) were calculated for each group to assess the proportion of the residual-specific mortality burden attributable to physical inactivity. To assess PAF, an unweighted multivariable Cox proportional hazards model was computed using the “stcox” command along with at-specifications (“at”), “eform,” and “vce” options (e.g., punafcc, at[groups==1] eform vce[unconditional]). Further details are provided elsewhere (27).

As noted in the Results section, various sensitivity analyses were computed (e.g., Cox proportional hazards model among those who did not die within the first year of follow-up to address any potential reverse causality concerns). In these sensitivity analyses, we also evaluated potential multiplicative interaction associations of physical activity and the covariates (described below) with residual-specific mortality. For these, a cross-product term of physical activity and the covariate was created and entered into the Cox proportional model, along with their main effect and the other covariates.

For all Cox proportional hazards models, and unless stated otherwise, covariates included age (yr, continuous), weight status, gender, race–ethnicity (Mexican American, non-Hispanic white, non-Hispanic black, and others), mean arterial pressure (MAP) (mm Hg, continuous), HDL cholesterol (mg·dL−1, continuous), total cholesterol (mg·dL−1, continuous), C-reactive protein (CRP) (mg·dL−1, continuous), previous year’s changes in physical activity (categorical; activity increased, decreased, or stayed the same), congestive heart failure (yes/no), CAD (yes/no), heart attack (yes/no), emphysema (yes/no), chronic bronchitis (yes/no), stroke (yes/no), diabetes (yes/no), and smoking status (categorical; smokes every day, smokes some days, former smoker, or never smoker). These covariates were selected based on previous research demonstrating their association with physical activity and mortality (31).

Age, gender, race–ethnicity, changes in physical activity, and smoking status were assessed via a self-report questionnaire. With regard to changes in physical activity, participants were asked, “How does the amount of activity that you reported for the past 30 d compare with your physical activity for the past 12 months? (response options: more active, less active, or about the same.” Regarding the chronic diseases (congestive heart failure, CAD, heart attack, emphysema, chronic bronchitis, stroke, and diabetes), participants were asked if they had ever been told by a physician or other health professionals that they had this disease. In addition to diabetes being assessed via physician diagnosis, here we also defined individuals as having diabetes if they had a fasting plasma glucose ≥126 mg·dL−1 or an A1C ≥6.5%. Using the average of up to four manually assessed blood pressure measurements, we calculated MAP using the following formula: [(diastolic blood pressure × 2) + systolic blood pressure]/3. HDL cholesterol and total cholesterol were assessed enzymatically in serum or plasma via a blood sample. High sensitivity CRP concentration was quantified using latex-enhanced nephelometry. Lastly, using measured height and weight to calculate body mass index (BMI), we defined weighted status as normal weight (18.5–24.9 kg·m−2), overweight (25.0–29.9 kg·m−2), or obese (30+ kg·m−2); notably, those with a BMI <18.5 were excluded from analysis.

RESULTS

In the 1999–2000, 2001–2002, 2003–2004, and 2005–2006 cycles, the respective NHANES response rates for adults (20+ yr) in the mobile examination center examination were 68.3%, 71.6%, 68.1%, and 69.8%.

In the 1999–2006 NHANES, 20,311 adults 20–85 yr old were enrolled. Among these, 3697 had missing physical activity or covariate data (Nresultant = 16,614). Among these 16,614 participants, a further 267 were excluded because of having a BMI <18.5 kg·m−2 (Nresultant = 16,347). Among these, a further 18 were excluded because of missing mortality status data (Nresultant = 16,329). These 16,329 adult participants constituted the analytical sample.

Among the analyzed sample of 16,329 participants, 2000 deaths accrued over the follow-up period. Among these 2000 deaths, 388 were due to diseases of the heart; 458 due to malignant neoplasms; 105 due to chronic lower respiratory diseases; 74 due to accidents (unintentional injuries); 125 due to cerebrovascular diseases; 63 due to Alzheimer’s disease; 77 due to diabetes; 49 due to influenza or pneumonia; 47 due to nephritis, nephrotic syndrome, or nephrosis; and 603 due to all other causes (residual); 11 participants had an unidentifiable cause of death. The outcome of interest in this study was the 603 residual-specific mortalities. Notably, the exact causes of these residual mortalities are not available in the publically accessible linked mortality data.

Table 1 displays the weighted characteristics of the study variables with considerations by activity and mortality status. A greater number of deaths (450 vs 153) were observable among those who were inactive at baseline. With regard to activity status, inactive individuals, compared with active individuals, were older, more likely to be female; had a higher MAP, total cholesterol, CRP, and BMI; had a lower HDL cholesterol level; and had a higher prevalence of each of the physician-diagnosed chronic conditions. In relation, those who died during the follow-up period, compared with those alive, were older, had a higher MAP at baseline, had higher total cholesterol and CRP, were less active, and had a higher prevalence of each of the physician-diagnosed chronic conditions.

TABLE 1
TABLE 1:
Weighted characteristics (mean/proportion (SE)) of the study variables with considerations by activity and mortality status, 1999–2006 NHANES (N = 16,329).

In an unadjusted Cox proportional hazards model, participants who were active (vs inactive) had a 56% reduced hazard of residual-specific mortality (HR = 0.44; 95% CI, 0.35–0.54; P < 0.001). After adjusting for age, gender, and race–ethnicity, results were attenuated but still statistically significant (HR = 0.59; 95% CI, 0.47–0.74; P < 0.001). After adding weight status as a covariate in this model, results were unchanged (HR = 0.59; 95% CI, 0.46–0.74; P < 0.001). After adjusting for all covariates (listed in the Methods section), results were attenuated but still statistically significant (HR = 0.67; 95% CI, 0.53–0.85; P = 0.001). In this fully adjusted model, the Harrell’s C concordance statistic was 0.85, and the test of proportional hazards assumption was not violated (P = 0.11). The PAF was 20.4% (95% CI, 8.3–30.9; P = 0.002).

Results were also similar when MVPA was expressed as a continuous variable instead of dichotomized at the cut-point of 2000 MET·min·month. After complete adjustment, and for a 1 MVPA MET·min·month increase, participants had a 1% reduced hazard for residual-specific mortality (HR = 0.99997; 95% CI, 0.99995–0.99999; P = 0.04). When expressed as a larger interval change, for every 2000 MET·min·month increase (equivalent to 30 min·d−1 of MVPA), participants had a 5% reduced hazard for residual-specific mortality (HR = 0.95; 95% CI, 0.90–0.99; P = 0.04).

Notably, as mentioned above, underweight participants (BMI <18.5 kg·m−2, n = 267) were excluded from the analyses, which we chose to do so because underweight status may have been a result of an unidentified pathology that could have confounded our associations. However, when analyses were recomputed with the inclusion of these underweight participants, being active was still inversely associated with residual-specific mortality (HR = 0.68; 95% CI, 0.54–0.86; P = 0.002).

In an effort to address any potential reverse causality, various sensitivity analyses were computed that examined the association between physical activity and residual-specific mortality among those who did not die during the early part of the follow-up and those who did not have various chronic diseases. There was little evidence to suggest any potential reverse causality. After complete adjustment, MVPA remained significantly associated with residual-specific mortality among those who did not die during the first 12 months of the follow-up (HR = 0.69; 95% CI, 0.54–0.87; P = 0.003; N = 16,191). Results were similar among those who did not die during the first 24 months of the follow-up (HR = 0.71; 95% CI, 0.55–0.93; P = 0.01; N = 16,005), 36 months of the follow-up (HR = 0.69; 95% CI, 0.53–0.90; P = 0.007; N = 15,807), or 48 months of the follow-up (HR = 0.70; 95% CI, 0.52–0.95; P = 0.02; N = 15,583).

With regard to excluding those with chronic diseases, and in a fully adjusted model, MVPA remained significantly associated with residual-specific mortality among those who did not have congestive heart failure (HR = 0.68; 95% CI, 0.52–0.88; P = 0.005; N = 15,834), CAD (HR = 0.67; 95% CI, 0.51–0.89; P = 0.006; N = 15,637), heart attack (HR = 0.71; 95% CI, 0.54–0.92; P = 0.01; N = 15,628), stroke (HR = 0.69; 95% CI, 0.53–0.89; P = 0.005; N = 15,802), emphysema (HR = 0.68; 95% CI, 0.53–0.87; P = 0.004; N = 16,036), chronic bronchitis (HR = 0.65; 95% CI, 0.52–0.82; P < 0.001; N = 15,360), diabetes (HR = 0.64; 95% CI, 0.50–0.82; P = 0.001; N = 14,233), or cancer (HR = 0.70; 95% CI, 0.55–0.88; P = 0.003; N = 14,976). Similarly, MVPA remained significantly associated with residual-specific mortality among those who did not have any of these aforementioned chronic diseases (HR = 0.64; 95% CI, 0.46–0.87; P = 0.006; N = 11,576).

With regard to multiplicative interaction effects, there was a significant multiplicative interaction effect for activity status and weight status on residual-specific mortality (HRinteraction = 1.33; 95% CI, 1.01–1.73; P = 0.04). When stratified by weight status, MVPA was associated with residual-specific mortality among those with a normal BMI (HR = 0.44; 95% CI, 0.29–0.65; P < 0.001; N = 4958), but not among those who were overweight (HR = 0.86; 95% CI, 0.61–1.22; P = 0.41; N = 5967) or obese (HR = 0.83; 95% CI, 0.54–1.28; P = 0.40; N = 5404). There was no multiplicative interaction effect for any of the other parameters, including age (P = 0.40), gender (P = 0.55), race–ethnicity (P = 0.97), MAP (P = 0.99), HDL cholesterol (P = 0.96), total cholesterol (P = 0.25), CRP (P = 0.08), change in physical activity (P = 0.24), congestive heart failure (P = 0.64), CAD (P = 0.96), heart attack (P = 0.15), stroke (P = 0.53), emphysema (P = 0.75), chronic bronchitis (P = 0.30), diabetes (P = 0.50), smoking status (P = 0.35), or cancer (P = 0.46).

With regard to the subsample with accelerometer-assessed MVPA data, and after complete adjustment, and for every 30 min·d−1 increase in MVPA, participants had a 66% reduced hazard for residual-specific mortality (HR = 0.34; 95% CI, 0.15–0.76; P = 0.01; N = 5536); the Harrell’s C concordance statistic for this model was 0.85, and the proportional hazards assumption was not violated (P = 0.45). In this model, there was no evidence of a multiplicative interaction effect of accelerometer-assessed MVPA and weight status on residual-specific mortality (HRinteraction = 0.74; 95% CI, 0.28–1.85; P = 0.48). When light-intensity physical activity was entered into the MVPA model, MVPA remained significantly associated with residual-specific mortality (HR = 0.38; 95% CI, 0.17–0.84; P = 0.01), but light-intensity physical activity was not significant (HR = 0.998; 95% CI, 0.994–1.001). Notably, a separate model evaluated the association of total physical activity (MVPA + light; expressed as a 1 min·d−1 increase in total physical activity) with residual-specific mortality, and it was just outside the statistical significance threshold (HR = 0.996; 95% CI, 0.993–1.000; P = 0.07).

Given that there are a considerable number of NHANES participants with invalid accelerometry data (<4 d of 10+ h·d−1 of monitoring data), coupled with the bias this introduces (20,23), we recomputed these accelerometer analyses that used multiple imputation to impute the invalid accelerometry data. Among the 2003–2006 NHANES participants with complete data for the variables used in the self-reported physical activity model, 7339 participants completed the accelerometer monitoring protocol. As stated above, among these participants, 5536 participants provided valid accelerometer data. In these analyses, we used multiple imputation to impute missing values for the 1803 participants with invalid accelerometry data. The following parameters were included in the imputation model, with 30 imputations created: the failure indicator (32), the Nelson–Aalen estimate of the baseline cumulative hazard (32), age, gender, and race–ethnicity. Among this multiply imputed multivariable Cox proportion hazards model of 7339 participants, for every 30 min·d−1 increase in MVPA, participants had a 53% reduced hazard for residual-specific mortality (HR = 0.47; 95% CI, 0.26–0.84; P = 0.01; N = 7739). In this model, there was no evidence of a multiplicative interaction effect of accelerometer-assessed MVPA and weight status on residual-specific mortality (P = 0.53). With employing multiple imputation, total physical activity (expressed as a 1-min·d−1 increase) was significantly inversely associated with residual-specific mortality (HR = 0.997; 95% CI, 0.994–0.999; P = 0.04). When expressed as a 60 min·d−1 increase in total physical activity, there was a 17% reduced hazard for residual-specific mortality (HR = 0.83; 95% CI, 0.69–0.94; P = 0.04).

DISCUSSION

Previous research demonstrates protective effects of physical activity on all-cause and cause-specific mortality, including the major causes of death as identified by the ICD-10. To our knowledge, however, what has yet to be explored in the literature is the association between physical activity and residual-specific mortality. This is important to consider; as identified in the third paragraph of our Results section, the greatest number of deaths in this national sample was not of the nine major causes of death identified by the ICD-10, i.e., they were residual-specific mortalities. The main finding of this study was that physical activity was inversely associated with residual-specific mortality risk. This observation was consistent across unadjusted, minimally adjusted, and fully adjusted models; was consistent among those with and without various chronic diseases; and was significant for both subjectively and objectively measured physical activity. With regard to the latter, associations between MVPA and residual-specific mortality were stronger for objectively measured MVPA when compared with self-reported MVPA. This greater strength of association for objectively measured physical activity aligns with other research (3). Further, the PAF was 20.4%, which suggests that 20.4% of the residual-specific mortality observed is attributable to participants not being physically active. Thus, about one out of five residual-specific deaths, in theory, could have been averted if inactivity was removed.

These findings, coupled with previous work demonstrating a protective effect of physical activity on all-cause and cause-specific mortality, underscores the powerful effect of physical activity on health. Given that the specific origins of the residual mortalities were not identifiable, we can only speculate as to the explanation behind our observed inverse association between physical activity and residual-specific mortality. Juxtaposing the NCHS codes for mortality and ICD-10 codes, we found that residual-specific mortality may include the following causes of death: 1) infectious and parasitic diseases; 2) in situ neoplasms, benign neoplasms, and neoplasms of unknown behavior of specified and unspecified sites; 3) diseases of the blood and blood-forming organs and certain disorders involving the immune system; 4) endocrine, nutritional, and metabolic diseases (other than diabetes mellitus); 5) mental and behavioral disorders; 6) diseases of the nervous system (other than Alzheimer’s disease); 7) diseases of the eye and adnexa; 8) diseases of the ear and mastoid process; 9) diseases of the circulatory system (other than heart disease); 10) diseases of the respiratory system (other than chronic lower respiratory diseases, pneumonia, and influenza); 11) diseases of the digestive system; 12) diseases of the skin and subcutaneous tissue; 13) diseases of the musculoskeletal and connective tissue; 14) diseases of the genitourinary system (other than nephritis, nephrotic syndrome, and nephrosis); 15) pregnancy, childbirth, and the puerperium; 16) certain conditions originating in the prenatal period; 17) congenital malformations, deformations, and chromosomal abnormalities; 18) symptoms, signs, and abnormal clinical and laboratory findings, not elsewhere defined; 19) external causes of mortality (other than accidents/unintentional injuries); and 20) factors influencing health status and contact with health services.

The underlying mechanisms for these residual deaths may include (but is not limited to) infection, immune dysfunction, systemic inflammation, and poor mental well-being. Physical activity is associated with a decrease in the risk of systemic inflammation (15), improved immune function (28), attenuation in risk of infection with moderate exercise (28), as well as significant improvement in mental well-being (14). These provide a plausible explanation for the attenuation in residual-specific deaths. In addition, studies have shown beneficial effects of physical activity in adults with HIV (13), Parkinson’s disease (7), essential hypertension (12), and also among those at risk of suicide (10). Further studies are, however, needed to substantiate this “herd” beneficial effect of MVPA on residual-specific mortality as seen in our study as well as the cause-specific effects of MVPA on the individual diseases that constitute residual mortality.

Future work would also benefit by evaluating the effects of cardiorespiratory fitness on residual mortality. This is important to consider because recent research has demonstrated that cardiorespiratory fitness may be a stronger indicator of health, including mortality, as compared with physical activity (6,26,34).

Limitations of this study include the unidentifiable residual-specific mortalities; however, the purpose of this study was to examine the association between MVPA and mortalities not attributed to the major causes of death, as opposed to exploring the association between MVPA and cause-specific mortality from nonmajor causes of death. Like most observational epidemiological studies, a limitation of this study was only having exposure (MVPA) data for a single period (baseline). Future work would benefit by looking at how changes in MVPA influence residual-specific mortality risk. Strengths of this study include the national sample employed, reasonable follow-up period (over 100 months, Table 1), comprehensive analytical assessment, and use of both subjectively and objectively measured MVPA.

In conclusion, the major finding of this study was that MVPA was inversely associated with residual-specific mortality risk. This finding persisted with different methods (self-report and accelerometry) for assessing MVPA and even among those without the evaluated chronic diseases. There was also some evidence of an inverse association between total physical activity (MVPA + light) and residual-specific mortality.

The authors are not aware of any affiliations, financial support, or memberships that may influence this manuscript.

The results of the present study do not constitute endorsement by the American College of Sports Medicine.

REFERENCES

1. Ainsworth BE, Haskell WL, Whitt MC, et al. Compendium of physical activities: an update of activity codes and MET intensities. Med Sci Sports Exerc. 2000;32(9 Suppl):S498–504.
2. Arem H, Moore SC, Park Y, et al. Physical activity and cancer-specific mortality in the NIH-AARP Diet and Health Study cohort. Int J Cancer. 2014;135(2):423–31.
3. Atienza AA, Moser RP, Perna F, et al. Self-reported and objectively measured activity related to biomarkers using NHANES. Med Sci Sports Exerc. 2011;43(5):815–21.
4. Batty GD, Shipley MJ, Marmot M, Smith GD. Physical activity and cause-specific mortality in men: further evidence from the Whitehall study. Eur J Epidemiol. 2001;17(9):863–9.
5. Beddhu S, Baird BC, Zitterkoph J, Neilson J, Greene T. Physical activity and mortality in chronic kidney disease (NHANES III). Clin J Am Soc Nephrol. 2009;4(12):1901–6.
6. Blair SN, Cheng Y, Holder JS. Is physical activity or physical fitness more important in defining health benefits? Med Sci Sports Exerc. 2001;33(6 Suppl):S379–99 discussion S419–320.
7. Crizzle AM, Newhouse IJ. Is physical exercise beneficial for persons with Parkinson’s disease? Clin J Sport Med. 2006;16(5):422–5.
8. Dankel SJ, Loenneke JP, Loprinzi PD. Determining the importance of meeting muscle-strengthening activity guidelines: is the behavior or the outcome of the behavior (strength) a more important determinant of all-cause mortality? Mayo Clin Proc. 2016;91(2):166–74.
9. Dankel SJ, Loenneke JP, Loprinzi PD. Does the fat-but-fit paradigm hold true for all-cause mortality when considering the duration of overweight/obesity? Analyzing the WATCH (Weight, Activity and Time Contributes to Health) paradigm. Prev Med. 2016;83:37–40.
10. Davidson CL, Babson KA, Bonn-Miller MO, Souter T, Vannoy S. The impact of exercise on suicide risk: examining pathways through depression, PTSD, and sleep in an inpatient sample of veterans. Suicide Life Threat Behav. 2013;43(3):279–89.
11. Dunn AL, Trivedi MH, O’Neal HA. Physical activity dose–response effects on outcomes of depression and anxiety. Med Sci Sports Exerc. 2001;33(6 Suppl):S587–97 discussion 609–510.
12. Farahani AV, Mansournia MA, Asheri H, et al. The effects of a 10-week water aerobic exercise on the resting blood pressure in patients with essential hypertension. Asian J Sports Med. 2010;1(3):159–67.
13. Garcia A, Fraga GA, Vieira RC Jr, et al. Effects of combined exercise training on immunological, physical and biochemical parameters in individuals with HIV/AIDS. J Sports Sci. 2014;32(8):785–92.
14. Lindegård A, Jonsdottir IH, Börjesson M, Lindwall M, Gerber M. Changes in mental health in compliers and non-compliers with physical activity recommendations in patients with stress-related exhaustion. BMC Psychiatry. 2015;15:272.
15. Loprinzi P, Cardinal B, Crespo C, et al. Objectively measured physical activity and C-reactive protein: National Health and Nutrition Examination Survey 2003–2004. Scand J Med Sci Sports. 2013;23(2):164–70.
16. Loprinzi PD. Accelerometer-determined physical activity and mortality in a national prospective cohort study of adults at high risk of a first atherosclerotic cardiovascular disease event. Int J Cardiol. 2016;202:417–8.
17. Loprinzi PD. Dose–response association of moderate-to-vigorous physical activity with cardiovascular biomarkers and all-cause mortality: considerations by individual sports, exercise and recreational physical activities. Prev Med. 2015;81:73–7.
18. Loprinzi PD. The effects of objectively-measured, free-living daily ambulatory movement on mortality in a national sample of adults with diabetes. Physiol Behav. 2016;154:126–8.
19. Loprinzi PD, Cardinal BJ. Interrelationships among physical activity, depression, homocysteine, and metabolic syndrome with special considerations by sex. Prev Med. 2012;54(6):388–92.
20. Loprinzi PD, Cardinal BJ, Crespo CJ, Brodowicz GR, Andersen RE, Smit E. Differences in demographic, behavioral, and biological variables between those with valid and invalid accelerometry data: implications for generalizability. J Phys Act Health. 2013;10(1):79–84.
21. Loprinzi PD, Davis RE. Effects of individual, combined, and isolated physical activity behaviors on all-cause mortality and CVD-specific mortality: Prospective cohort study among U.S. adults. Physiol Behav. 2015;151:355–9.
22. Loprinzi PD, Hall ME. Physical activity and dietary behavior with red blood cell distribution width. Physiol Behav. 2015;149:35–8.
23. Loprinzi PD, Smit E, Cardinal BJ, Crespo C, Brodowicz G, Andersen R. Valid and invalid accelerometry data among children and adolescents: comparison across demographic, behavioral, and biological variables. Am J Health Promot. 2014;28(3):155–8.
24. Loprinzi PD, Walker JF, Lee H. Association between physical activity and inflammatory markers among U.S. adults with chronic obstructive pulmonary disease. Am J Health Promot. 2014;29(2):81–8.
25. Matthews CE, Cohen SS, Fowke JH, et al. Physical activity, sedentary behavior, and cause-specific mortality in black and white adults in the Southern Community Cohort Study. Am J Epidemiol. 2014;180(4):394–405.
26. Myers J, Kaykha A, George S, et al. Fitness versus physical activity patterns in predicting mortality in men. Am J Med. 2004;117(12):912–8.
27. Newson RB. Attributable and unattributable risks and fractions and other scenario comparisons. Stata J. 2013;13(4):672–98.
28. Nieman DC. Exercise, infection, and immunity. Int J Sports Med. 1994;15(3 Suppl):S131–41.
29. Scarmeas N, Luchsinger JA, Brickman AM, et al. Physical activity and Alzheimer disease course. Am J Geriatr Psychiatry. 2011;19(5):471–81.
30. Troiano RP, Berrigan D, Dodd KW, Masse LC, Tilert T, McDowell M. Physical activity in the United States measured by accelerometer. Med Sci Sports Exerc. 2008;40(1):181–8.
31. Vischer UM, Safar ME, Safar H, et al. Cardiometabolic determinants of mortality in a geriatric population: is there a “reverse metabolic syndrome”? Diabetes Metab. 2009;35(2):108–14.
32. White IR, Royston P. Imputing missing covariate values for the Cox model. Stat Med. 2009;28(15):1982–98.
33. Williams PT. Dose–response relationship of physical activity to premature and total all-cause and cardiovascular disease mortality in walkers. PLoS One. 2013;8(11):e78777.
34. Williams PT. Physical fitness and activity as separate heart disease risk factors: a meta-analysis. Med Sci Sports Exerc. 2001;33(5):754–61.
Keywords:

EPIDEMIOLOGY; EXERCISE; NHANES; SURVIVAL

© 2016 American College of Sports Medicine