Exercise: good for you, even if a weekend warrior

 A recent nationwide cohort study found that active physical activity, either as a "weekend warrior" or by regular activity, is associated with similarly decreased all-cause and cause-specific mortality (see exercise wkend warrior does well jamaintmed2022 in dropbox or doi:10.1001/jamainternmed.2022.2488).

 

Details:

-- 350,978 adults with self-reported physical activity in the US National Health Interview Survey, from 1997 to 2013. NHIS is a nationally representative survey of civilian, noninstitutionalized participants in the US

    -- data were linked to the National Death Index, through 31December2015

-- mean age 41 (57% 18-44yo/33% 45-64yo/10% 65-84yo), 51% women, 68% white/15% Latinx/12% Black, 53% >high school education/28% high school degree/15% less, 48% middle income/10% high/12% low, 64% married, BMI 18.5-25 in 38%/25-30 in 34%/>30 in 23%

-- cigarette smoking never in 60%/former 19%/current 21%, alcohol none 22%/former 12%/current 66%

-- psychological distress: none in 97% [not sure how they measured this, pretty hard to believe in the US....]

-- chronic conditions: none 71%/ 1 in 24%/more than 1 in 5%; self-related health excellent 35%/very good 35%/good 24%/fair-to-poor 6%

-- mobility difficulty: none or little 94%

-- participants excluded at baseline: those with cancer, chronic bronchitis, emphysema, heart disease, and stroke; those having limitations in activities of daily living; those unable to perform moderate or vigorous activity

-- categories of exercise:

    -- physically inactive: <150 min/week of moderate-to-vigorous activity (MVPA) [note: "physically inactive" includes those very close to being considered physically active]

    -- physically active: at least 150 min/week of MVPA or at least 75 min/week of vigorous activity

    -- physical activity subdivided into:

        -- 1-2 sessions/week (weekend warrior)

        -- regular active (at least 3 sessions/week)

-- median physical activity performed:

    -- weekend warriors: 240 min/week

    -- regularly active group: 420 min/week

-- main outcomes: all-cause mortality; and mortality from cardiovascular disease or cancer

-- median follow-up 10.4 years (3.6 million person-years)

 

Results:

-- Deaths on follow-up: 21,898 deaths overall, 4130 from cardiovascular disease, 6034 from cancer

-- results below were after multivariate adjustment for age, sex, race/ethnicity, education, income, marital status, smoking, alcohol intake, self-related health, psychological distress, number of comorbidities, mobility, BMI, and total volume of physical activity; the first 2 years of follow-up were excluded to decrease the possibility of reverse causation (ie, did those with little exercise do so because of health-related problems that would themselves increase early mortality?)

-- Regularly active individuals, compared to inactive ones:

    -- all-cause mortality: 15% decrease, HR 0.85 (0.83-0.88)

    -- CVD mortality: 23% decrease, HR 0.77 (0.71-0.84)

    -- cancer mortality: 12% decrease, HR 0.88 (0.83-0.94)

-- Weekend warriors, compared to inactive ones:

    -- all-cause mortality: 8% decrease, HR 0.92 (0.83-1.02)

    -- CVD mortality: 13% decrease, HR 0.87 (0.66-1.15)

    -- cancer mortality: 6% decrease, HR 0.94 (0.77-1.15)

-- duration of physical activity for regularly active individuals:

    -- sessions <20 min: HR 1.02 (0.86-1.21)

    -- sessions >60 min: HR 0.83 (0.79-0.87)

-- including VPA in addition to MVPA (ie, increasing the intensity of exercise): lowered mortality further

-- adjusting for the total amounts of MVPA: similar all-cause and cause-specific mortality rates [though these are by mathematical modeling: see limitations below]

 

-- comparing weekend warriors vs inactive ones: working out 2 vs 1 times/week, or intensity of that workout did not alter the results 

-- comparing weekend warrior to regularly active participants: similar rates of all-cause and disease-specific mortalities after adjusting for the amount of total MVPA

-- no difference in results if exclude the first 5 years of evaluation (further decreasing the possibility of reverse causation)

 

Discussion:

-- this study found a statistically significant mortality benefit (all-cause, cardiovascular, cancer) from regular exercise as per WHO and other guidelines, a statistical trend to benefit (but not statistically significant) for those with similar levels of exercise but only for 1-2 days ("weekend warriors"), no statistically significant difference between those doing regular vs weekend warrior activity if adjust for the quantity of exercise, and further decline of all-cause mortality if higher percentage of vigorous exercise (ie, more exercise intensity) or longer exercise sessions

-- there are several other studies quoted in the article that also found mortality benefit from weekend warrioring, though i had access to only one of them:

    -- Health Survey for England and Scotland (see doi:10.1001/jamainternmed.2016.8014):

        -- 63,591 adults with self-reported leisure-time activity, comparing those with recommended activity (at least 150 min/week moderate activity or 75 min/wk vigorous activity) but either as regularly active participants or as weekend warriors doing 1-2 sessions/week) vs those with less than recommended (similar definitions to above study)

        -- 8802 all-cause mortality (2780 for cardiovasc causes, 2526 from cancer) over 8.8 years of follow-up: 

            -- all-cause mortality:

                -- weekend warriors: decrease 30%, HR 0.70 (0.60-0.82)

                -- regularly active participants: decrease 35%, HR 0.65 (0.58-0.73)

            -- cardiovasc mortality: 

                -- weekend warriors: decrease 40%, HR 0.60 (0.45-0.82)

                -- regularly active participants: decrease 41%, HR 0.59 (0.48-0.73)

            -- cancer mortality: 

                -- weekend warriors: decrease 18%, HR 0.82 (0.63-1.06), almost statistically significant

                -- regularly active participants: decrease 21%, HR 0.79 ( 0.66-0.94) 

            -- all of these hazard ratios for weekend warriors and regularly active people have overlapping confidence intervals, so none reveal statistically significant differences between these types of exercisers

-- i have published many prior blogs on exercise (http://gmodestmedblogs.blogspot.com/search?q=exercise ), including:

    -- a pretty interesting evolutionary perspective suggesting the humans evolved into bipeds (through intermediaries) to improve their ability to run, as well as the ability to make weapons, and evolved into meat eaters (see http://gmodestmedblogs.blogspot.com/2019/01/exercise-new-evolutionary-model.html ), which also comments on a meta-analysis finding that increased steps/day was associated with decreased arterial stiffness, an independent predictor of cardiovascular disease

    -- exercise is beneficial for mental health: http://gmodestmedblogs.blogspot.com/2018/09/exercise-helps-mental-health.html

    -- an article from the AHA about importance of assessing cardiorespiratory fitness. ?should this be another vital sign: http://gmodestmedblogs.blogspot.com/2017/01/cardiovascular-fitness-new-vital-sign.html

    -- another article suggesting that exercise had a similar mortality benefit for secondary cardiovasc protection as statins, ACE-inhibitors, beta-blockers and antiplatelet meds: http://gmodestmedblogs.blogspot.com/2017/01/cardiovascular-fitness-new-vital-sign.html

    -- exercise leads to less deterioration of renal function in those with CKD: http://gmodestmedblogs.blogspot.com/2022/06/ckd-exercise-helps.html

    -- others blogs were on exercise decreasing recurrent strokes, development of diabetes, depression, heart failure, cancer risk and even benefit in the elderly performing low levels of exercise

 

-- another article hot off the press (ie internet) found increased exercise above guideline recommendations led to even more benefit (see exercise the more the better circ2022 in dropbox, or DOI: 10.1161/CIRCULATIONAHA.121.058162)

    -- 116,221 adults with long-term follow-up in the Nurses' Health Study and Health Professionals Follow-up Study from 1988-2018, which had self-reported leisure-time activity assessed up to 15 times [very useful: many studies just have baseline data; this one had important assessment over time, and therefore with more accurate quantitative assessment of exercise]

    -- 47,596 deaths over the 30 years of follow-up

    -- comparisons based on meeting guideline recommendations (vigorous physical activity VPA of 75-149 min/week; moderate physical activity MPA of 150-300 min/week): 

        -- VPA vs no VPA: 

            -- all-cause mortality: dec 19%, HR 0.81 (0.76-0.87)

            -- cardiovascular disease mortality: dec 31%, HR 0.69 (0.60-0.78)

            -- non-cardiovasc mortality: dec 15%, HR 0.85 (0.79-0.92)

        -- MPA vs no MPA (no MPA was between 0-19 min/week): 

            -- all-cause mortality: dec 20%, HR 0.80 (0.77-0.83)

            -- cardiovascular disease mortality: dec 22%, HR 0.78 (0.72-0.84)

            -- non-cardiovasc mortality: dec 19%, HR 0.81 (0.78-0.85)

        -- but, comparing higher levels of exercise:

            -- exercise 2-4 times that recommended by guidelines (ie VPA 150-299 min/wk; or MPA 300- 599 min/wk) vs guideline recommended:

                -- mortality for each mortality category: decreased an additional 2%-4% for VPA and 4%-13% for MPA

                    -- though no significant difference with further increases in exercise [but small numbers of people in this category]

 

Limitations:

-- the NHIS study at the top of this blog was predominantly of a white, educated, lean, healthy population (there were many exclusions for medical reasons, including such broad and not-defined categories such as "heart disease"), who were less likely to be smokers or drink alcohol, all of which limit generalizability to other groups

-- the categories assessed provided fixed demarcations for issues that involve continuous variables:

    -- eg those doing no exercise and those doing 149 minutes were all considered "physically inactive"; similarly, those doing 150 minutes/week are lumped in with those doing 3 hours/day

        -- and the reliability of mathematical modeling decreases when the averages of those doing regular physical activity (420 min/week) is so much higher than the weekend warriors (240 min/week)

    -- assessing by number of chronic conditions is also fraught: the total number of them may be less useful than which chronic conditions (those with severe heart disease may well have a worse prognosis than a bit of asthma), and within each condition there may be big differences (those who have a little vs a lot of kidney disease/asthma/copd/heart disease) ...

    -- and not sure how to interpret some of their broad categories: a yes-no answer for psychological distress?? likely pretty profound differences in how different people answer that vs their realities (eg those with rosier outlooks on life vs the pessimists), what people consider psychological distress, how it was asked, and if there is any relevance to a baseline assessment once to what happened for the full 10+ years of followup

        -- and the fact that 97% stated they had no significant psychological stress suggests either that exercise helps mood (unlikely to create no perceived stress to that extent) or more likely that there was a selection bias (those with a more positive attitude and less stressful lives are more likely to do more exercise, and this lack of stress by itself provides health benefits)

-- no breakdown by types of exercise: was weight-lifting as beneficial as more cardiovascular exercise?? that has been found in earlier studies with weight-lifting having similar health benefit, but would be useful to have assessed this in this study (it does seem intuitive that there would be more benefit to cardiovascular exercise in terms of cardiovascular mortality....)

-- there was no information about very important lifestyle issues that might affect these mortality outcomes (eg diet, occupational or environmental exposures, access to medical care, income inequality,...)

-- the associations between weekend warrior and the fatal events measured had confidence intervals that did cross the 1.0 line, rendering the results “not statistically significant”:

    -- this was likely due to the large difference in numbers of individuals (9,992 weekend warriors vs 150,906 regularly active), and the number of deaths (all-cause mortality of 89 vs 1080)

    -- there are strong conclusions that regular activity is associated inversely with all of the adverse fatal events measured, with a pretty clear dose-response curve showing that the longer the exercise sessions and the higher the exercise intensity, the lower the death rates

    -- their analysis did find that there was no significant difference in outcomes in weekend warriors vs those performing regular exercise, though this is also limited by the numbers of people and events in each group (still wide confidence intervals)

    -- the fact that their confidence ratios included 1 for weekend warriors still did find results with an overall skew to benefit (ie strong statistical trends for benefit in weekend warriors)

-- this study assessed "leisure time" activity. it seems pretty intuitive that the human body is not really able to differentiate classes of exercise (leisure time vs directly work-related or even sort-of work related such as bicycling to work) and there probably should be a combo of all exercises done. one way to do this would be to use an objective measure of work (eg accelerometers worn all day) with multiple samplings (eg every year, and not just at baseline) and assessing outcomes

    -- but this would not allow retrospectively assessing some of our historical databases as above (plentiful longterm data for the asking, easy-to-do questionnaires)

        -- these databases were not designed to assess this particular outcome of exercise types and fatal outcomes, do not have the granular data to have more rigorous conclusions, and self-reported activity may not be so reliable; all of this limits the rigor of their conclusions

-- this was an observational study, so cannot determine more than just associations (ie, not causality). But this was a huge study and had median 10-year follow-up (and, it might be a tad difficult to perform an actual RCT, randomizing lots of people to varying degrees of exercise and varying numbers of days of exercise/week, and including people with a variety of baseline underlying conditions, and following them 10 years...)

    -- they did (appropriately) try to mathematically compensate for dozens of potential confounders (eg age, sex, education level, smoking/drinking, BMI, comorbidities....), but this number of variables would increase confidence intervals and potentially undercut the statistical validity of their results

    -- and within these many variables (largely considered as binary buckets), we do not have sufficient psychosocial variables to understand the data more completely: eg we really do not know the specifics or severity of conditions that might influence the amount of exercise or improve/decrease the odds of exercise being beneficial (eg psychiatric conditions, social conditions, other unmeasured medical comorbidities, etc)

 

 

so, the implications from above:

-- exercise is important to health

-- it does not seem so important whether done frequently in smaller aliquots or similar amounts done in 1-2 days

    -- which all makes it easier on a population scale, since many people cannot include much exercise in their stressful workdays

-- there are many benefits to exercise above and beyond mortality, including decreasing chronic diseases (diabetes, hypertension, heart disease, respiratory disease, mental health, kidney disease, stress reduction.....)

-- and the added Nurses' Health Study and Health Professionals Follow-up Study noted above suggests that for exercise, there really seems to be increasing benefit to doing much more than the recommended levels (though orthopedic surgeons may be busier with more knee/shoulder/etc problems...)

 

geoff

 

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

 

For access to the dropbox, go to link: https://www.dropbox.com/sh/0bmvtita8mzms11/XDTwHySFFg

Then go to "clinic", then to "clinical stuff" for articles, or go to https://www.dropbox.com/sh/nyle22q1fn6lkpk/AAB9B2hBj5Kw4gtrJAkI-UF8a?dl=0 for the powerpoint presentations

 

to get access to all of the blogs (2 options):

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

 

or: go to https://www.bucommunitymedicine.org/ , a website from the Community Medicine section at Boston Medical Center.  This site does have a very searchable and accessible list of my blogs (though there have been a few that did not upload over the last year or two). but overall it is much easier to view blogs and displays more at a time.

 

 

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list