May 19, 2024


International Student Club UK

Evidence from the UK Biobank cohort study


Methods and findings

We obtained accelerometer-measured moderate-intensity and vigorous-intensity physical activities and total volume of PA, over a 7-day period in 2013–2015, for 90,211 participants without prior or concurrent CVD in the UK Biobank cohort. Participants in the lowest category of total PA smoked more, had higher body mass index and C-reactive protein, and were diagnosed with hypertension. PA was associated with 3,617 incident CVD cases during 440,004 person-years of follow-up (median (interquartile range [IQR]): 5.2 (1.2) years) using Cox regression models. We found a linear dose–response relationship for PA, whether measured as moderate-intensity, vigorous-intensity, or as total volume, with risk of incident of CVD. Hazard ratios (HRs) and 95% confidence intervals for increasing quarters of the PA distribution relative to the lowest fourth were for moderate-intensity PA: 0.71 (0.65, 0.77), 0.59 (0.54, 0.65), and 0.46 (0.41, 0.51); for vigorous-intensity PA: 0.70 (0.64, 0.77), 0.54 (0.49,0.59), and 0.41 (0.37,0.46); and for total volume of PA: 0.73 (0.67, 0.79), 0.63 (0.57, 0.69), and 0.47 (0.43, 0.52). We took account of potential confounders but unmeasured confounding remains a possibility, and while removal of early deaths did not affect the estimated HRs, we cannot completely dismiss the likelihood that reverse causality has contributed to the findings. Another possible limitation of this work is the quantification of PA intensity-levels based on methods validated in relatively small studies.


Higher levels of physical activity (PA) have been found to be inversely associated with the risk of cardiovascular disease (CVD) [1–5]. However, most of this evidence has relied on questionnaires that measure PA relatively imprecisely and with different validity in population subgroups defined by age, sex, and socioeconomic status [6]. Further, questionnaire-based methods are not well suited to capture incidental activity that occurs throughout the day and therefore do not validly measure all movement that occurs in a specified time period [7]. This could, therefore, lead to uncertainty in the estimation of the strength of the association between PA and CVD and in estimating its shape, i.e., are increasing levels of activity associated with additional cardiovascular benefit?

Objective measures, such as wrist-worn accelerometers, are able to incorporate all components of PA—frequency, intensity, and duration—as a continuous score and thus validly capture all PA undertaken [7]. This is likely to be important since total energy expenditure is conceivably the primary pathway through which PA reduces risk of disease [8]. While an increasing number of studies have examined the association of objectively measured PA and mortality, few have assessed its association with incident CVD [9–12]. We investigated the association of moderate-intensity and vigorous-intensity PAs and total volume of PA, measured objectively by accelerometer, with incident CVD in 90,211 UK Biobank participants.


The UK Biobank, a large population-based cohort study, was established to enable research into genetic and nongenetic risk factors for diseases of middle and old age through longitudinal follow-up of participants throughout the UK. The UK Biobank recruited over 500,000 people aged between 40 and 69 years in 2006 to 2010 from across the UK. These participants provided blood, urine, and saliva samples for future analysis and detailed information about themselves and agreed to be followed for multiple health-related outcomes [13–15]. Ethical approval was obtained by the UK Biobank from the North West Multicentre Research Ethics Committee, the National Information Governance Board for Health and Social Care in England and Wales, and the Community Health Index Advisory Group in Scotland. All participants provided written informed consent.

Assessment of physical activity

Data from a subsample of 103,687 participants who wore an Axivity AX3 triaxial accelerometer on their dominant wrist were collected over a 7-day period in 2013 to 2015. It has been found that wrist-worn accelerometers can explain about 44% to 47% of the variation in PA energy expenditure as measured by doubly labelled water [16]. We only included participants whose accelerometer data could be successfully calibrated, meaning that different devices provide comparable data outputs [14]. We excluded participants with >1% clipped values, which occur when the sensor’s dynamic range of ±8 g is exceeded before or after calibration [14]. In addition, we excluded participants with implausibly high activity values, defined as average vector magnitude scores of >100 mg. We also excluded participants with insufficient wear-time, defined as the unavailability of at least 72 hours of data or who lacked data for every 1-hour period of the 24-hour cycle (scattered over multiple days) [14]. This resulted in 96,675 participants with accelerometer data for the analysis (see S1 Fig for flowchart of study participants).

We extracted total volume of PA, measured as the average vector magnitude in milli-gravity (mg) units. The metric, average vector magnitude, has demonstrated good face validity in the UK Biobank (activity: 7.5% lower per decade of age) [14] and has been validated against doubly labelled water which is a gold-standard measure for energy expenditure [16]. We estimated minutes of moderate and vigorous PA per week from percentage of time spent in 100 mg to 400 mg and above 400 mg, respectively [17,18]. The total volume, and each intensity of PA, were categorised into the equal quarters of their distributions in the analytic sample.

Statistical analysis

This study did not have a prespecified analysis plan, and no data-driven analyses were included.

We excluded participants who had been diagnosed with CVD from follow-up hospital records, before the end of their accelerometer wear (prevalent cases) resulting in 91,040 participants (S1 Fig). Follow-up time was calculated as person-time in months for each participant from the final date of accelerometer wear to the first occurrence of CVD or the end of study (31 March 2020, 29 February 2016, and 31 October 2016 for participants from England, Wales, and Scotland, respectively). The analytic sample consisted of 90,211 participants who had complete data for PA, age, sex, ethnicity, age completed full time education, Townsend Deprivation Index, smoking, and alcohol consumption.

We computed descriptive statistics—mean (standard deviation)/median (interquartile range (IQR)) for continuous measures and percentage for categorical variables—by categories of total volume of PA. We used multivariable-adjusted Cox proportional hazards regression models to estimate hazard ratios (HRs) for the association between total volume, moderate- and vigorous PAs, and risk of CVD. Cox regression was also used for subgroup and sensitivity analyses. Analyses were adjusted for age (stratified by 5-year age-at-risk intervals to satisfy the proportional hazards assumption within each age-at-risk group), sex, ethnicity (white and nonwhite), age completed full time education (years), area-based social deprivation using the Townsend Deprivation Index [19] (categorised into quarters), smoking (never, former, and current), and alcohol (never, less than 3 times/week, and ≥3 times/week) which were collected at baseline (2006 to 2010). The proportional hazards assumption was assessed using log (-log) survival plots and covariate-by-(log) time interaction terms [20], and none of the analysed variables violated this assumption. We conducted competing risk analysis using cause-specific and Fine and Gray methods [21]. There was no material difference between the estimates from these models and the models that did not account for competing risks. Therefore, we have presented the results from the latter.

We assessed the shape of the relationship between moderate-intensity and vigorous-intensity, total volume of PA, and incident CVD using a restricted cubic spline model. For this purpose, we trimmed observations less than 5% and greater than 95% of the distribution. We specified the knots at the 25th, 50th, and 75th centiles that were used for the categorisation of the variables for total volume, moderate-intensity, and vigorous-intensity of PA.

To address reverse causation, we repeated the Cox regression analyses after removing CVD events that occurred within the first year and then 2 years of follow-up. We conducted sensitivity analysis by adjusting for 4 groups of comorbid conditions that may affect participants’ ability to engage in PA—cancer (ICD codes: C01-C26, C30-C58, C60-C97, and D00-D48), diabetes mellitus (ICD codes: E10-E14), hypertension (ICD code: I10), and chronic lower respiratory disease (ICD codes: J43 and J44.9). We conducted prespecified subgroup analysis by sex. For subgroup analyses by sex, we categorised total volume and each intensity of PA into equal quarters of their distributions that were sex-specific. Furthermore, we examined the change in HR for each level of PA and CVD after sequential adjustment of variables not included in the main Cox regression analyses. These variables were hypertension (ICD code: I10); self-rated health (good/excellent versus fair/poor); body mass index (BMI) (categorised as underweight (<18.5 kg/m2), normal weight (18.5 to 24.9 kg/m2), overweight (25.0 to 29.9 kg/m2), and obese (≥30.0 kg/m2)); total cholesterol (mmol/L), high-density lipoprotein (HDL, mmol/L); low-density lipoprotein (LDL, mmol/L); triglycerides (mmol/L); log-transformed C-reactive protein (mg/L); glycated haemoglobin (HbA1c) (mmol/mol) categorised into <42 mmol/mol, 42 to 47 mmol/mol, and ≥48 mmol/mol; red and processed meat (times/week); fresh fruit (serving/day); and cooked vegetable (serving/day).

Data were analysed using SAS 9.4 (SAS Institute, Cary, North Carolina, United States of America) and STATA 14.0 (StataCorp. 2015, College Station, Texas, USA), and statistical testing was conducted at a 2-tailed alpha level of 0.05.

This study was reported as per the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines [22] (S1 STROBE Checklist).


For the 90,211 participants included in the current study, 3,617 were diagnosed with CVD (3,305 nonfatal and 312 fatal) during 440,004 person-years of follow-up (median (IQR): 5.2 (1.2) years). Out of 6,614 participants without prevalent CVD who were excluded due to poor wear time, 262 (4.0%) were diagnosed with incident CVD during a median follow-up of 62.6 months (compared with 4.0% cases during a median follow-up of 61.9 months for the study participants). Participants in the highest category of total PA engaged in more moderate and vigorous accelerometry-measured PA. Compared with participants in the highest category of total PA, participants in the lowest category also had a higher BMI, smoked more, drank slightly more alcohol, were diagnosed with hypertension, and had higher levels of C-reactive protein and HbA1c (Table 1).

We found a linear dose–response relationship between moderate and vigorous PAs and risk of incident CVD (Figs 1, 2B and 2C). The findings for total PA were very similar (Figs 1 and 2A). Compared with the lowest category of moderate-intensity PA, the HRs and 95% CIs for increasing quarters were 0.71 (0.65, 0.77), 0.59 (0.54, 0.65), and 0.46 (0.41, 0.51) (Fig 1), and the corresponding values for vigorous activity were 0.70 (0.64, 0.77), 0.54 (0.49, 0.59), and 0.41 (0.37, 0.46). We found similar trends in HRs for total PA and incident CVD (Fig 1).

Subgroup, secondary, and sensitivity analyses

During the study period, 2,220 men and 1,397 women developed CVD. There were minimal differences between the sexes in risk of CVD for moderate and total volume of PAs. However, for vigorous PA, compared with men, a stronger inverse association was observed for women (Fig 3) (Pinteraction < 0.001) especially for second quarter of vigorous PA compared with the first.

For analyses by type of CVD, there were 2,773 participants who were diagnosed with IHD and 844 with cerebrovascular disease. The HRs by categories of moderate, vigorous, and total PAs did not differ substantially from the overall CVD estimates (Fig 4).

During the first year of follow-up, 607 participants were diagnosed with CVD. After excluding participants during the first year of follow-up, there were minimal changes to the HRs (S1 Table). Furthermore, 1,244 participants were diagnosed with CVD within first 2 years of follow-up. After removal of participants within the first 2 years of follow-up, the strong association of total volume of PA with CVD incidence persisted (S1 Table) across both moderate- intensity and vigorous-intensity and total volume of PAs.

The analyses that excluded participants with comorbid conditions such as cancer, diabetes mellitus, hypertension, or chronic lower respiratory disease did not substantially alter the findings (S2 Fig). Furthermore, sensitivity analyses conducted to examine the inclusion of additional covariates (hypertension, self-rated health, BMI, total cholesterol, HDL, LDL, triglycerides, C-reactive protein, HbA1c, red and processed meat, fresh fruit, and cooked vegetables) to the multivariable Cox models utilised in the main analyses attenuated the HRs by just 6.6% to 18.0% (S2–S4 Tables).


In this large population-based cohort study in adults, we found a linear inverse dose–response association between accelerometer-measured moderate and vigorous PA, as well as total PA and incident CVD, with no threshold of effect at low or high levels. We found broadly similar associations within each subcomponent of CVD, and the associations were comparable for men and women, with the exception of vigorous activity where there was a 23% lower risk for incident CVD among men compared with 36% lower risk among women for the second quarter compared to the first.

The inverse association that we found for accelerometer-measured PA and incident CVD is much stronger than that reported from questionnaire-based studies [1,2,23,24]. Notably, we saw no evidence of a higher risk of CVD, particularly stroke, in those engaging in high levels of PA, including vigorous, in contrast to an observation of greater risk in such individuals in 1 large British cohort [25]. In our study, there were minimal differences in the HRs for total, moderate, and vigorous PAs. In contrast to these findings, the protective effect reported by questionnaire-based studies is about 26% for total PA, 20% to 25% for moderate volume/intensity of PA, and 30% to 35% for high amounts/intensity of PA [1,2]. However, the results of the current study concur with the findings from a harmonised meta-analysis with mortality as the outcome that found the protective effect of accelerometer-measured PA to be much greater than that reported in the literature for questionnaire-based studies [10].

We found a linear dose–response association between accelerometer-measured PA and incident CVD with no threshold effect. This is in contrast to the curvilinear dose–response associations observed in studies based on leisure time PA measured via questionnaires [26,27], but similar to those observed in accelerometer-based studies [12,28]. It is noteworthy that Shiroma and Lee [2] found the curvilinear dose–response relationship between self-reported PA and CVD to be present among men whereas in women the shape of the curve appears to be more linear. In contrast, our study found no threshold effect in either men or women.

In our study we found that, using objective measurement, the magnitude of association for total volume of PA was similar to that for moderate and vigorous PAs, which is important for clinical and public health practice. One explanation for this finding is that the participants in the highest level of total volume of PA engaged in higher amounts of moderate-intensity and vigorous-intensity PAs. In support of the contention that the total PA might have the central role in lower risk of CVD, a recent individual-level meta-analysis of 8 studies of accelerometer-assessed PA found that higher levels of total volume of PA irrespective of intensity had an inverse association with mortality from all causes [10]. Future research may need to concentrate on the components of total volume of PA using validated measures of activity intensity [29] in a large sample to unravel the contribution of each.

There is insufficient evidence in the literature to demonstrate that PA confers differential benefits for men and women for CVD. The similarity of findings between men and women for total volume of PA is in accordance with a study from the China Kadoorie Biobank study that used questionnaire data [23]. However, Sattelmair and colleagues found the protective effect to be greater among women than men (36% versus 21%) [1]. Similarly, a review by Shiroma and Lee [2] reported a median risk reduction of 40% and 30%, respectively, for coronary heart disease/CVD among women compared with men, when comparing highest versus lowest total PA. This could conceivably be explained by a different balance of the components of total PA across the different studies. In other words, if moderate and vigorous PA did have different effects on risk, then differences in observed associations of total PA with risk across studies could be explained by a disparity in the contribution of vigorous versus moderate PA to total PA across study samples.

Strengths and limitations

The major strengths of this study are its large sample size and prospectively collected data on CVD events. In addition, PA was measured objectively by an accelerometer which can capture leisure and non-leisure PAs across multiple domains [28] and minimises the potential for recall bias associated with PA assessed with questionnaires. The UK Biobank cohort consists of mostly white participants living in less socioeconomically deprived areas. Even though the characteristics of the participants may not be generalisable to other populations, it can be used to provide valid estimates of exposure–disease relationships due to its large sample size and multitude of exposures [30,31].

There is currently a lack of robust evidence on whether a 7-day measurement is representative of habitual PA. A previous validation study showed a 7-day measure is strongly associated with PA over a period of up to 3.7 years [9]. In contrast, a recent study found that moderate-to-vigorous PA measured by hip-worn accelerometers is moderately stable over time, but there is considerable within-subject variability [32]. It is therefore possible that if the measurement error is random, our findings are attenuated and have underestimated the true association between PA and CVD.

The mean levels of moderate PA in this study, 743.2 min/week, are much higher than the recommended 150 minutes/week of moderate-to-vigorous activity [33]. However, it is important to note that current guidelines are based on self-reported data on ‘sustained’ participation in moderate-to-vigorous intensity PA undertaken in bouts of 10 minutes or more. It is suggested that approximately 750 to 1,000 minutes/week of moderate PA is to be expected when using devices that can capture all forms of incidental activity and removing the arbitrary bout criteria [34,35]. In addition, the accurate quantification of levels of moderate PA from accelerometers remains an active challenge, as we have had to rely on methods validated in relatively small validation studies [17,36]. There is a future need for large-scale validation studies in both healthy and diseased individuals [37]. In light of these considerations, we have pragmatically measured mean participant levels of moderate PA in this study.

Given that this is observational data, we also cannot rule out concerns around reverse causality [38] where incipient CVD, not yet detected clinically, might lead to reduced PA because it makes it more difficult for an individual to undertake PA. To address this possibility we removed the first 1 and 2 years following measurement by accelerometry in sensitivity analyses, and the inverse associations for all 3 types of PA persisted. Our adjustment for potentially important confounders (age, sex, ethnicity, education, smoking, alcohol consumption, and deprivation) only had a modest impact on the associations, although we recognise that residual confounding cannot be ruled out entirely.


  1. 1.
    Sattelmair J, Pertman J, Ding EL, Kohl 3rd HW, Haskell W, Lee IM. Dose response between physical activity and risk of coronary heart disease: a meta-analysis. Circulation. 2011;124(7):789–95. pmid:21810663
  2. 2.
    Shiroma Eric J, Lee IM. Physical activity and cardiovascular health. Circulation. 2010;122(7):743–52. Available from: pmid:20713909
  3. 3.
    Piercy KL, Troiano RP, Ballard RM, Carlson SA, Fulton JE, Galuska DA, et al. The Physical activity guidelines for Americans. JAMA. 2018;320(19):2020–8. Available from: pmid:30418471
  4. 4.
    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. pmid:18091006
  5. 5.
    Department of Health and Social Care Lywodraeth Cymru Welsh Government, Department of Health Northern Ireland and the Scottish Government. UK Chief Medical Officers’ Physical Activity Guidelines. 2019. Available from:
  6. 6.
    Shephard RJ. Limits to the measurement of habitual physical activity by questionnaires. Br J Sports Med. 2003;37(3):197. pmid:12782543
  7. 7.
    Bassett DR, Troiano RP, Mcclain JJ, Wolff DL. Accelerometer-based Physical activity: Total volume per day and standardized measures. Med Sci Sports Exerc. 2015;47(4):833–8. pmid:25102292
  8. 8.
    National Research Council (US) Committee on Diet and Health. 6. Calories: Total Macronutrient Intake, Energy Expenditure, and Net Energy Stores. In: Diet and Health: Implications for Reducing Chronic Disease Risk: Executive Summary. Washington, DC: The National Academies Press; 1989. Available from:
  9. 9.
    Dwyer T, Pezic A, Sun C, Cochrane J, Venn A, Srikanth V, et al. Objectively measured daily steps and subsequent long term all-cause mortality: The Tasped Prospective Cohort Study. PLoS One. 2015;10(11):e0141274. Available from: pmid:26536618
  10. 10.
    Ekelund U, Tarp J, Steene-Johannessen J, Hansen BH, Jefferis B, Fagerland MW, et al. Dose-response associations between accelerometry measured physical activity and sedentary time and all cause mortality: systematic review and harmonised meta-analysis. BMJ. 2019;366:l4570. pmid:31434697
  11. 11.
    Lee IM, Shiroma EJ, Evenson KR, Kamada M, LaCroix AZ, Buring JE. Accelerometer-measured physical activity and sedentary behavior in relation to all-cause mortality: The Women’s Health Study. Circulation. 2018;137(2):203–5. pmid:29109088
  12. 12.
    Jefferis BJ, Parsons TJ, Sartini C, Ash S, Lennon LT, Papacosta O, et al. Does total volume of physical activity matter more than pattern for onset of CVD? A prospective cohort study of older British men. Int J Cardiol. 2019;278:267–72. pmid:30578094
  13. 13.
    Bycroft C, Freeman C, Petkova D, Band G, Elliott LT, Sharp K, et al. The UK Biobank resource with deep phenotyping and genomic data. Nature. 2018;562(7726):203–9. Available from: pmid:30305743
  14. 14.
    Doherty A, Jackson D, Hammerla N, Plotz T, Olivier P, Granat MH, et al. Large scale population assessment of physical activity using wrist worn accelerometers: The UK Biobank Study. PLoS One. 2017;12(2):e0169649. pmid:28146576
  15. 15.
    Sudlow C, Gallacher J, Allen N, Beral V, Burton P, Danesh J, et al. UK Biobank: An open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 2015;12(3):e1001779. Available from: pmid:25826379
  16. 16.
    White T, Westgate K, Hollidge S, Venables M, Olivier P, Wareham N, et al. Estimating energy expenditure from wrist and thigh accelerometry in free-living adults: a doubly labelled water study. Int J Obes (Lond). 2019; Available from: pmid:30940917
  17. 17.
    Hildebrand M, VT VANH, Hansen BH, Ekelund U. Age group comparability of raw accelerometer output from wrist- and hip-worn monitors. Med Sci Sports Exerc. 2014;46(9):1816–24. pmid:24887173
  18. 18.
    Rowlands A V, Plekhanova T, Yates T, Mirkes EM, Davies M, Khunti K, et al. Providing a basis for harmonization of accelerometer-assessed physical activity outcomes across epidemiological datasets. J Meas Phys Behav. 2019;2(3).
  19. 19.
    Townsend P. Deprivation. J Soc Policy. 1987;16(2):125–46. Available from:
  20. 20.
    Bradburn MJ, Clark TG, Love SB, Altman DG. Survival analysis Part III: multivariate data analysis—choosing a model and assessing its adequacy and fit. Br J Cancer. 2003;89(4):605–11. pmid:12915864
  21. 21.
    Fine JP, Gray RJ. A Proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc. 1999 Jul 31;94(446):496–509.
  22. 22.
    Uhlig K, Menon V, Schmid CH. Recommendations for reporting of clinical research studies. Vol. 49, American Journal of Kidney Diseases: The Official Journal of the National Kidney Foundation. United States; 2007. p. 3–7.
  23. 23.
    Bennett DA, Du H, Clarke R, Guo Y, Yang L, Bian Z, et al. Association of physical activity with risk of major cardiovascular diseases in Chinese men and women. JAMA Cardiol. 2017;2(12):1349–58. Available from: pmid:29117341
  24. 24.
    Gunnell AS, Knuiman MW, Divitini ML, Cormie P. Leisure time physical activity and long-term cardiovascular and cancer outcomes: the Busselton Health Study. Eur J Epidemiol. 2014;29(11):851–7. pmid:25354993
  25. 25.
    Armstrong ME, Green J, Reeves GK, Beral V, Cairns BJ. Frequent physical activity may not reduce vascular disease risk as much as moderate activity: large prospective study of women in the United Kingdom. Circulation. 2015;131(8):721–9. pmid:25688148
  26. 26.
    Woodcock J, Franco OH, Orsini N, Roberts I. Non-vigorous physical activity and all-cause mortality: systematic review and meta-analysis of cohort studies. Int J Epidemiol. 2011;40(1):121–38. pmid:20630992
  27. 27.
    Arem H, Moore SC, Patel A, et al. Leisure time physical activity and mortality: A detailed pooled analysis of the dose-response relationship. JAMA Intern Med. 2015;175(6):959–67. Available from: pmid:25844730
  28. 28.
    Guo W, Key TJ, Reeves GK. Accelerometer compared with questionnaire measures of physical activity in relation to body size and composition: a large cross-sectional analysis of UK Biobank. BMJ Open. 2019;9(1):e024206. pmid:30700478
  29. 29.
    Doherty A, Smith-Byrne K, Ferreira T, Holmes M V, Holmes C, Pulit SL, et al. GWAS identifies 14 loci for device-measured physical activity and sleep duration. Nat Commun. 2018;9(1):5257. Available from: pmid:30531941
  30. 30.
    Fry A, Littlejohns TJ, Sudlow C, Doherty N, Adamska L, Sprosen T, et al. Comparison of sociodemographic and health-related characteristics of UK Biobank participants with those of the general population. Am J Epidemiol. 2017 Jun 21;186(9):1026–34. Available from: pmid:28641372
  31. 31.
    Batty GD, Gale CR, Kivimäki M, Deary IJ, Bell S. Comparison of risk factor associations in UK Biobank against representative, general population based studies with conventional response rates: prospective cohort study and individual participant meta-analysis. BMJ. 2020 Feb 12;368:m131. pmid:32051121
  32. 32.
    Saint-Maurice PF, Sampson JN, Keadle SK, Willis EA, Troiano RP, Matthews CE. Reproducibility of accelerometer and posture-derived measures of physical activity. Med Sci Sports Exerc. 2020 Apr;52(4):876–83. pmid:31688655
  33. 33.
    National Institute for Health and Care Excellence. Physical Activity Gudeline—Cardiovascular disease prevention. 2010. Available from:
  34. 34.
    Thompson D, Batterham AM, Peacock OJ, Western MJ, Booso R. Feedback from physical activity monitors is not compatible with current recommendations: A recalibration study. Prev Med. 2016 Oct;91:389–94. pmid:27330025
  35. 35.
    Shiroma EJ, Cook NR, Manson JE, Buring JE, Rimm EB, Lee I-M. Comparison of self-reported and accelerometer-assessed physical activity in older women. PLoS One. 2016 Dec 29;10(12):e0145950. Available from:
  36. 36.
    Rowlands A V, Mirkes EM, Yates T, Clemes S, Davies M, Khunti K, et al. Accelerometer-assessed physical activity in epidemiology: Are monitors equivalent? Med Sci Sports Exerc. 2018;50(2):257–65. pmid:28976493
  37. 37.
    Barker J, Smith Byrne K, Doherty A, Foster C, Rahimi K, Ramakrishnan R, et al. Response to: One size does not fit all—application of accelerometer thresholds in chronic disease. Int J Epidemiol. 2019 Aug 2;48(4):1381. Available from: pmid:31373623
  38. 38.
    Kivimäki M, Singh-Manoux A, Pentti J, Sabia S, Nyberg ST, Alfredsson L, et al. Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis. BMJ 2019;365:l1495. pmid:30995986