The association between venous thromboembolism and physical inactivity in everyday life

BMJ 2011; 343 doi: http://dx.doi.org/10.1136/bmj.d3865 (Published 04 July 2011) Cite this as: BMJ 2011;343:d3865
  1. James D Douketis, MD1,
  2. Alfonso Iorio, MD2
  1. 1Vascular Medicine, McMaster University, Hamilton, ON, Canada, L8N 4A6
  2. 2Health Information Research Unit, Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, ON, Canada
  1. jdouket{at}mcmaster.ca

Seems to be small and slightly higher than that for oral contraceptive use

Observational studies have shown that several lifestyle choices and habits, such as eating too much refined sugar or drinking more than one glass of wine a day, may have adverse health effects.1 2 The linked prospective cohort study by Kabrhel and colleagues (doi:10.1136/bmj.d3867) adds inactivity to this list of sins.3 The study followed 69 950 female nurses for an average of 18 years. Those women who were the most inactive, defined by the number of hours of sitting a day (>41 hours a week outside of work), were two to three times more likely to develop otherwise unprovoked venous thromboembolism (VTE), which manifested as pulmonary embolism, than women who spent the least amount of time sitting (<10 hours a week outside of work). The association remained robust after controlling for other risk factors for VTE such as increasing age, body mass index, and concomitant disease, and was not mitigated by periods of physical activity and exercise.

If the findings are valid they may have major public health ramifications. The study also showed that physical inactivity correlated with coronary heart disease (spanning from 1.2% to 5.1% across fifths of physical inactivity) and hypertension (from 18% to 25%). Prolonged periods of physical inactivity could be one of the hidden mechanisms that link arterial disease and venous disease.4

Before raising alarms about the implications of the results, several questions about the study need to be explored. Firstly, is the association between inactivity and VTE valid? Secondly, if valid, can anything be done to mitigate the risk of VTE? This question is important because VTE is often clinically silent and its initial manifestation may be life threatening pulmonary embolism. Thirdly, if the answer to both of these questions is yes, are public health initiatives needed to increase activity levels outside of work?

The validity of an association between a putative exposure (inactivity) and an outcome (VTE) rests on several factors. The first is whether the association is biologically plausible; in this case, increased sitting may promote venous stasis and coagulation activation, as shown by an increase in D-dimer concentrations or other markers, which may lead to VTE.5 Secondly, the exposure must predate the outcome: one potential confounding factor is that the authors assumed that baseline inactivity levels (in 1988 and 1990) applied over the 18 year observation period. Physical inactivity is neither a one-time exposure, as with radiation exposure after a nuclear reactor accident, nor an ongoing exposure, as with air pollution among city dwellers.6 Several factors, such as emerging disease or worsening of existing disease, might have modified inactivity levels over time. Thirdly, there should be a graded response between the amount of exposure and the likelihood of an outcome, which the study showed, because the risk of VTE increased with the degree of inactivity. Finally, it is crucial to account for potential confounders because superimposed risk factors for VTE that are linked to both the exposure and outcome may weaken the association between inactivity and VTE. For example, varicose veins and treatment with oestrogen, which are risk factors for VTE in women,7 8 were not accounted for but are unlikely to be linked with inactivity. There may be unmeasured confounders—for example, were women with VTE more likely to have lipid disorders or raised inflammatory markers, which recently have been identified as potential determinants of VTE?9 10 Indeed, recent studies have looked at whether conventional cardiovascular risk factors, such as the metabolic syndrome, contribute to VTE as they do with arterial vascular disease.4 10

Furthermore, even with the most robust multivariable analysis, it may be unclear which variables are causal effectors and which are epiphenomena. For example, at baseline inactive women were more often affected by coronary artery disease, so was physical inactivity the trigger or the consequence? On balance, however, the study findings seem to favour an association between physical inactivity and VTE.

On a practical level, is the risk of VTE with physical inactivity modifiable, and if so, is it worth modifying? Changing the level of inactivity (or sitting), especially outside of the workplace, is possible but not easily done and may be regarded with benign neglect, especially among nurses, many of whom are on their feet for several hours a day. As to whether it is worth modifying, consider other common exposures such as oral contraceptive use or long haul airplane travel. Both of these exposures (like inactivity) increase the risk of VTE two to four times.8 11 However, the size of the increase in risk with these exposures is so small that use of oral contraceptives is not discouraged (except in certain high risk people), and there is certainly no inclination to curtail overseas air travel.

In Kabrhel and colleagues’ study the incidence of pulmonary embolism was 0.039% a year in women who were most active (41/104 720 person years) and 0.11% a year (16/14 565 person years) in women who were most inactive. This is an absolute risk increase of about 0.07% a year or seven additional cases per 10 000 person years, which is slightly higher than the three to five additional cases of VTE a year seen in users of oral contraceptives. Furthermore, those at highest risk sat, on average, for six hours a day outside of work, which seems excessive, and only about 5% of nurses studied fell into this risk group. However, it is important to note that if deep vein thrombosis had been documented the incidence of VTE associated with inactivity would have been higher.

Overall, the study reinforces the notion that prolonged inactivity increases the risk of VTE, and it shows how this occurs in everyday life. The findings also indirectly support the use of preventive interventions for at risk people with prolonged immobility, typically patients in hospital, in whom anticoagulants to prevent VTE remain underused.12 For otherwise healthy people, the take home message may be to apply the ancient Greek proverb of “métron áriston” or “moderation is best” to both our activity and inactivity.


Cite this as: BMJ 2011;343:d3865


  • Research, doi:10.1136/bmj.d3867
  • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.

  • Provenance and peer review: Commissioned; not externally peer reviewed.