Benefit of excercise

Fox news reports that a new study shows that exercise is good for you.

Something to bear in mind. Remember correlation is not necessarily causation.

So the research shows that people who exercise vigorously had a reduced risk of dying regardless of bodyweight or chronic diseases.

75 minutes of vigorous exercise is all you need for the week

Some vigorous exercise gives some benefit on life span — reduced chance of early death.

Current recommendation is for 30 minutes of exercise per day.

If you can talk easily while exercising your are not doing it hard enough, Sweating is good. Heart should be really pumping, 7 or 8 out of 10 (maximum effort).

Any amount of exercise is better than none.

Unfortunately I don’t think Fox reported which study made these findings so I can’t see how accurately they reported on the findings of the paper.

Forbes (Alice G. Walton) report on this story (I presume) here:

… a large new study in JAMA Internal Medicine suggests that indeed vigorous exercise, regardless of body weight or chronic disease status, can reduce early mortality significantly…

The researchers from James Cook University and the University of Sydney looked at data tracking over 204,000 participants, 45 and older, for an average of six and a half years. They were divided into three groups: those who engaged in only moderate activity, like leisurely swimming, social tennis, or even household chores; and people whose activity was vigorous (jogging, aerobics, or competitive tennis) up to 30% of the time, or more than 30% of the time.

This news story is based on the paper “Effect of Moderate to Vigorous Physical Activity on All-Cause Mortality in Middle-aged and Older Australians” by
Klaus Gebel, Ding Ding, Tien Chey, Emmanuel Stamatakis, Wendy J. Brown, Adrian E. Bauman. JAMA Intern Med. Published online April 06, 2015. doi:10.1001/jamainternmed.2015.0541

These are quotes from the abstract, the full abstract can be seen on-line for free if you search for the paper.

… Objective To examine whether the proportion of total moderate to vigorous activity (MVPA) that is achieved through vigorous activity is associated with all-cause mortality independently of the total amount of MVPA…

… Design, Setting, and Participants We performed a prospective cohort study with activity data linked to all-cause mortality data from February 1, 2006, through June 15, 2014, in 204 542 adults aged 45 through 75 years from the 45 and Up population-based cohort study from New South Wales, Australia (mean [SD] follow-up, 6.52 [1.23] years). Associations between different contributions of vigorous activity to total MVPA and mortality were examined using Cox proportional hazards models, adjusted for total MVPA and sociodemographic and health covariates…

Results During 1 444 927 person-years of follow-up, 7435 deaths were registered. Compared with those who reported no MVPA (crude death rate, 8.34%), the adjusted hazard ratios for all-cause mortality were 0.66 (95% CI, 0.61-0.71; crude death rate, 4.81%), 0.53 (95% CI, 0.48-0.57; crude death rate, 3.17%), and 0.46 (95% CI, 0.43-0.49; crude death rate, 2.64%) for reporting 10 through 149, 150 through 299, and 300 min/wk or more of activity, respectively. Among those who reported any MVPA, the proportion of vigorous activity revealed an inverse dose-response relationship with all-cause mortality: compared with those reporting no vigorous activity (crude death rate, 3.84%) the fully adjusted hazard ratio was 0.91 (95% CI, 0.84-0.98; crude death rate, 2.35%) in those who reported some vigorous activity (but <30% of total activity) and 0.87 (95% CI, 0.81-0.93; crude death rate, 2.08%) among those who reported 30% or more of activity as vigorous. These associations were consistent in men and women, across categories of body mass index and volume of MVPA, and in those with and without existing cardiovascular disease or diabetes mellitus.

…Conclusions and Relevance Among people reporting any activity, there was an inverse dose-response relationship between proportion of vigorous activity and mortality. Our findings suggest that vigorous activities should be endorsed in clinical and public health activity guidelines to maximize the population benefits of physical activity…

So not a bad job of reporting by Fox news.. they told us that a study exists although they added a lot of caveats to the results given that it is in line with current advice that exercise if good.

Flu Pandemic, How worried should we be?

Just for a little fun, a model was made for the number of flu cases in the various countries which are reporting data. A neural network was trained which predicts the amount of flu cases given the number of cases the day before. GDP(PPP), size of population and population density were also included as inputs. Obviously living in the UK it is more important to consider the cases in UK first.

Confirmed Flu cases in UK

Confirmed Flu cases in UK

Confirmed cases of flu in Mexico

Confirmed cases of flu in Mexico

Confirmed Flu cases in USA

Confirmed Flu cases in USA

Confirmed flu cases in Canada

Confirmed flu cases in Canada

Here are the predictions for todays flu cases in the various countries, WHO will usually release these data at around 6pm-8pm UK time, so we should be able to see how good is the prediction in about 5 hours time.

Country Predicted Cases Uncertainity in prediction
NewZealand 6 1
Ireland 1 1
CostaRica 1 1
Denmark 1 1
ElSalvador 2 1
HongKong 1 1
Israel 7 1
Switzerland 2 1
Austria 1 1
Portugal 1 1
Netherlands 3 1
Canada 232 23
Colombia 3 1
Spain 101 5
Korea 4 1
Italy 7 1
UK 43 2
France 7 1
Germany 14 1
Mexico 1419 155
US 1207 544
Guatemala 3 2
Sweden 1 1
Poland 2 1

Here are the predictions using the same data, for tomorrow cases (using the number predicted above).

Country Predicted Cases Uncertainity in prediction
NewZealand 6 1
Ireland 2 1
CostaRica 0 1
Denmark 2 1
ElSalvador 2 2
HongKong 2 1
Israel 7 1
Switzerland 2 1
Austria 2 1
Portugal 1 1
Netherlands 4 1
Canada 253 61
Colombia 5 1
Spain 127 8
Korea 6 1
Italy 10 1
UK 57 2
France 10 1
Germany 20 1
Mexico 1910 748
US 1520 643
Guatemala 4 2
Sweden 2 1
Poland 3 1

The model should be improved as more data can be included, as the outbreak proceeds. It is planned to train using the days since the start of reporting, and also the level of Alert set by the WHO. The model shows the number of cases increasing approximately linearly in most cases.