Advertisement

Physical activity in leisure time and at work and risk of dementia: A prospective cohort study of 117,616 individuals

  • Author Footnotes
    1 These authors contributed equally to the manuscript.
    Ida Juul Rasmussen
    Footnotes
    1 These authors contributed equally to the manuscript.
    Affiliations
    Department of Clinical Biochemistry, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark

    The Copenhagen General Population Study, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark
    Search for articles by this author
  • Author Footnotes
    1 These authors contributed equally to the manuscript.
    Katrine Laura Rasmussen
    Footnotes
    1 These authors contributed equally to the manuscript.
    Affiliations
    Department of Clinical Biochemistry, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark

    The Copenhagen General Population Study, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark
    Search for articles by this author
  • Jesper Q. Thomassen
    Affiliations
    Department of Clinical Biochemistry, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark

    The Copenhagen General Population Study, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark
    Search for articles by this author
  • Børge G. Nordestgaard
    Affiliations
    The Copenhagen General Population Study, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark

    The Copenhagen City Heart Study, Frederiksberg Hospital, Nordre Fasanvej 57, DK-2000, Frederiksberg, Denmark

    Department of Clinical Medicine, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen, Denmark

    Department of Clinical Biochemistry, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark
    Search for articles by this author
  • Peter Schnohr
    Affiliations
    The Copenhagen City Heart Study, Frederiksberg Hospital, Nordre Fasanvej 57, DK-2000, Frederiksberg, Denmark
    Search for articles by this author
  • Anne Tybjærg-Hansen
    Affiliations
    Department of Clinical Biochemistry, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark

    The Copenhagen General Population Study, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark

    The Copenhagen City Heart Study, Frederiksberg Hospital, Nordre Fasanvej 57, DK-2000, Frederiksberg, Denmark

    Department of Clinical Medicine, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen, Denmark
    Search for articles by this author
  • Ruth Frikke-Schmidt
    Correspondence
    Corresponding author. Department of Clinical Biochemistry KB 3011 Copenhagen, University Hospital, RigshospitaletBlegdamsvej 9, DK-2100, Copenhagen Ø, Denmark.
    Affiliations
    Department of Clinical Biochemistry, Rigshospitalet, Blegdamsvej 9, DK-2100, Copenhagen, Denmark

    The Copenhagen General Population Study, Herlev and Gentofte Hospital, Borgmester Ib Juuls Vej 1, DK-2730, Herlev, Denmark

    Department of Clinical Medicine, University of Copenhagen, Blegdamsvej 3, DK-2200, Copenhagen, Denmark
    Search for articles by this author
  • Author Footnotes
    1 These authors contributed equally to the manuscript.
Open AccessPublished:August 17, 2022DOI:https://doi.org/10.1016/j.atherosclerosis.2022.08.004

      Highlights

      • Prospective study including 117,616 individuals from the general Danish population.
      • Up to 43 years of follow-up with no losses to follow-up.
      • Low leisure physical activity associates with high risk of non-Alzheimer's dementia.
      • Leisure versus work physical activity is superior in predicting risk.
      • Public health advice on physical activity may prevent non-Alzheimer's dementia.

      Abstract

      Background and aims

      Up to 40% of all dementia cases may be preventable, primarily by treating or acting on well-established cardiovascular risk factors such as diabetes, hypertension, smoking, and physical inactivity. Whether physical inactivity is associated with risk of non-Alzheimer's dementia – a disease influenced by cardiovascular risk factors – and whether a given association differs for physical activity in leisure time and at work remains unknown.

      Methods

      We conducted a prospective cohort study including 117,616 individuals from the Copenhagen General Population Study and the Copenhagen City Heart Study with up to 43 years of follow-up.

      Results

      Multifactorially adjusted hazard ratios for low versus high physical activity at leisure time was 1.60 (95% confidence interval 1.40–1.83) for non-Alzheimer's dementia and 0.94 (0.80–1.11) for Alzheimer's disease. Corresponding values for non-Alzheimer's dementia after additional adjustment for physical activity at work or apolipoprotein E (APOE) genotype were 1.60 (1.40–1.83) and 1.82 (1.34–2.15). Multifactorially and APOE adjusted hazard ratios for high versus low physical activity at work were 1.50 (1.10–2.05) for non-Alzheimer's dementia and 1.62 (1.14–2.31) for Alzheimer's disease. When combining the two types of physical activity, physical activity in leisure time had the strongest relationship with risk of non-Alzheimer's dementia.

      Conclusions

      Physical inactivity in leisure time was associated with increased risk of non-Alzheimer's dementia, independent of modifiable risk factors and physical activity at work. The present study thus provides evidence for public health advice on physical activity in leisure time for the vascular part of dementia.

      Graphical abstract

      Keywords

      1. Introduction

      Worldwide dementia incidence numbers are projected to be higher than 130 million by 2050 and a third of old people now die with dementia [
      • Livingston G.
      • Huntley J.
      • Sommerlad A.
      • et al.
      Dementia prevention, intervention, and care: 2020 report of the Lancet Commission.
      ], making dementia a major challenge for health and social care in the 21st century. Recent estimates from the Lancet Commission suggest that up to 40% of all dementia cases may be preventable, primarily by treating or acting on well-established cardiovascular risk factors such as diabetes, hypertension, smoking, and physical inactivity. [
      • Livingston G.
      • Huntley J.
      • Sommerlad A.
      • et al.
      Dementia prevention, intervention, and care: 2020 report of the Lancet Commission.
      ].
      Several large meta-analyses of primarily prospective cohort studies report high physical activity to be associated with decreased risk of cognitive decline [
      • Sofi F.
      • Valecchi D.
      • Bacci D.
      • et al.
      Physical activity and risk of cognitive decline: a meta-analysis of prospective studies.
      ], all-cause dementia, and Alzheimer's disease. [
      • Hamer M.
      • Chida Y.
      Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence.
      ,
      • Groot C.
      • Hooghiemstra A.M.
      • Raijmakers P.G.H.M.
      • et al.
      The effect of physical activity on cognitive function in patients with dementia: a meta-analysis of randomized control trials.
      ,
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ,
      • Xu W.
      • Wang H.F.
      • Wan Y.
      • Tan C.C.
      • Yu J.T.
      • Tan L.
      Leisure time physical activity and dementia risk: a dose-response meta-analysis of prospective studies.
      ] However when assessing physical activity ≥ 10 years before dementia diagnosis, the associations disappear [
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ], which may indicate possible reverse causation – the fact that people might stop exercising due to prodromal phases of dementia. Further, randomized controlled trials only including physical activity interventions do not seem to improve cognition in healthy older adults. [
      • Groot C.
      • Hooghiemstra A.M.
      • Raijmakers P.G.H.M.
      • et al.
      The effect of physical activity on cognitive function in patients with dementia: a meta-analysis of randomized control trials.
      ,
      • Young J.
      • Angevaren M.
      • Rusted J.
      • Tabet N.
      Aerobic exercise to improve cognitive function in older people without known cognitive impairment.
      ,
      • Zotcheva E.
      • Bergh S.
      • Selbæk G.
      • et al.
      Midlife physical activity, psychological distress, and dementia risk: the HUNT study.
      ] The mechanism behind the association between physical inactivity and increased risk of dementia is primarily thought to be indirect through modification of other shared risk factors for dementia and cardiovascular disease, such as diabetes, obesity, and hypertension. [
      • Sofi F.
      • Valecchi D.
      • Bacci D.
      • et al.
      Physical activity and risk of cognitive decline: a meta-analysis of prospective studies.
      ,
      • Hamer M.
      • Chida Y.
      Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence.
      ,
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ,
      • Petersen C.B.
      • Grønbæk M.
      • Helge J.W.
      • Thygesen L.C.
      • Schnohr P.
      • Tolstrup J.S.
      Changes in physical activity in leisure time and the risk of myocardial infarction, ischemic heart disease, and all-cause mortality.
      ] Consequently, it is important to examine the association between physical inactivity and risk of non-Alzheimer's dementia – a disease influenced by cardiovascular risk factors – and to address whether potential associations are due to reverse causation or are independent of other genetic and cardiovascular risk factors including physical activity at work. Whether a given association differ for physical activity in leisure time and at work remains unknown – an important question that just recently has been resolved for cardiovascular disease. [
      • Holtermann A.
      • Schnohr P.
      • Nordestgaard B.G.
      • Marott J.L.
      The physical activity paradox in cardiovascular disease and all-cause mortality: the contemporary Copenhagen General Population Study with 104,046 adults.
      ].
      We therefore tested whether physical inactivity in leisure time is associated with increased risk of both non-Alzheimer's dementia and Alzheimer's disease. We adjusted for important risk factors including the major genetic risk factor for dementia, apolipoprotein E (APOE) genotype, and physical activity at work. To minimize possible reverse causation, we repeated the analyses after exclusion of individuals with follow-up periods within two, five, or ten years from baseline. Finally, we examined combined physical activity in leisure time and at work on risk of non-Alzheimer's dementia. The analyses were performed in two prospective cohorts of the general population, the Copenhagen General Population Study and the Copenhagen City Heart Study, with up to 43 years of follow-up totaling 117,616 individuals.

      2. Patients and methods

      The studies were approved by the institutional review boards and Danish ethical committees (no KF-100.2039/91 and no KF-01-144/01) and were conducted according to the Declaration of Helsinki. Written informed consent was obtained from all individuals.

      2.1 Participants

      We included individuals from the Copenhagen General Population Study (CGPS) and the Copenhagen City Heart Study (CCHS), two similar, prospective studies including individuals from the general population in Denmark. Individuals in both studies were white and of Danish descent, and none appeared in more than one study.
      CGPS was initiated in 2003–2015. [
      • Juul Rasmussen I.
      • Rasmussen K.L.
      • Nordestgaard B.G.
      • Tybjærg-Hansen A.
      • Frikke-Schmidt R.
      Impact of cardiovascular risk factors and genetics on 10-year absolute risk of dementia: risk charts for targeted prevention.
      ,
      • Rasmussen K.L.
      • Tybjærg-Hansen A.
      • Nordestgaard B.G.
      • Frikke-Schmidt R.
      Plasma levels of Apolipoprotein E, APOE genotype, and all-cause and cause-specific mortality in 105 949 individuals from a white general population cohort.
      ,
      • Jørgensen A.B.
      • Frikke-Schmidt R.
      • Nordestgaard B.G.
      • Tybjærg-Hansen A.
      Loss-of-function mutations in APOC3 and risk of ischemic vascular disease.
      ] Individuals were selected randomly based on the national Danish Civil Registration System to reflect the adult Danish population aged 20–100 years. Data were obtained from a self-administered questionnaire reviewed together with an investigator at the day of attendance, a physical examination, and from blood samples including DNA extraction. Information on physical activity in leisure time, vital status and disease status was available on 98,871 individuals. CCHS was initiated in 1976–78 with follow-up examinations in 1981–83, 1991–94 and 2001–03. [
      • Juul Rasmussen I.
      • Rasmussen K.L.
      • Nordestgaard B.G.
      • Tybjærg-Hansen A.
      • Frikke-Schmidt R.
      Impact of cardiovascular risk factors and genetics on 10-year absolute risk of dementia: risk charts for targeted prevention.
      ,
      • Rasmussen K.L.
      • Tybjærg-Hansen A.
      • Nordestgaard B.G.
      • Frikke-Schmidt R.
      Plasma levels of Apolipoprotein E, APOE genotype, and all-cause and cause-specific mortality in 105 949 individuals from a white general population cohort.
      ,
      • Jørgensen A.B.
      • Frikke-Schmidt R.
      • Nordestgaard B.G.
      • Tybjærg-Hansen A.
      Loss-of-function mutations in APOC3 and risk of ischemic vascular disease.
      ] Individuals were recruited and examined as in the CGPS. We used information from the first examination the individual participated in. Information on physical activity in leisure time, vital status and disease status was available on 18,745 individuals. Combining the two studies yielded a total of 117,616 individuals, of whom 4,391 developed dementia during a median follow-up of 10 years (range <1–43 years). No individual was lost to follow-up due to the complete Danish registries. Follow-up began at the time of first examination (2003–2015 for CGPS and 1976–78, 1981–83, 1991–1994 or 2001–2003 for CCHS). Follow-up ended at occurrence of dementia (n = 4,391), death (n = 21,465), emigration (n = 587), or on December 13th, 2018 (last update of the registry), whichever came first (Supplementary Fig. 1).

      2.2 Dementia

      In CGPS and CCHS information on births, deaths, emigrations, and immigrations was collected from the national Danish Civil Registration System. Information on diagnoses of dementia was drawn from the national Danish Patient Registry and the national Danish Causes of Death Registry. The national Danish Patient Registry has information on all patient contacts with all clinical hospital departments in Denmark since 1977, including emergency wards and outpatient clinics from 1994. The national Danish Causes of Death Registry contains data on the causes of all deaths in Denmark, as reported by hospitals and general practitioners. Alzheimer's disease was World Health Organization International Classification of Diseases, 8th revision (ICD8) code 290.10 and ICD10 codes F00 and G30. Non-Alzheimer's dementia included vascular dementia (ICD10 code F01) and unspecified dementia (ICD8 code 290.18 and ICD10 code F03) and no record of Alzheimer's disease in the follow-up period. The quality of these registry-based dementia diagnoses has previously been validated [
      • Phung T.K.T.
      • Andersen B.B.
      • Høgh P.
      • Kessing L.V.
      • Mortensen P.B.
      • Waldemar G.
      Validity of dementia diagnoses in the Danish hospital registers.
      ] and was further validated by combining the well-known association between the apolipoprotein E (APOE) ε4 allele and all-cause dementia and Alzheimer's disease in the present cohorts. [
      • Juul Rasmussen I.
      • Rasmussen K.L.
      • Nordestgaard B.G.
      • Tybjærg-Hansen A.
      • Frikke-Schmidt R.
      Impact of cardiovascular risk factors and genetics on 10-year absolute risk of dementia: risk charts for targeted prevention.
      ,
      • Rasmussen K.L.
      • Tybjærg-Hansen A.
      • Nordestgaard B.G.
      • Frikke-Schmidt R.
      Plasma levels of Apolipoprotein E, APOE genotype, and all-cause and cause-specific mortality in 105 949 individuals from a white general population cohort.
      ].

      2.3 Ischemic heart disease

      We used ischemic heart disease as a positive control for our measurement of physical inactivity in leisure time. Ischemic heart disease was ICD8 codes 410–414 and ICD10 codes I20–I25.

      2.4 Genotyping

      TaqMan-based assays (Life Technologies, a part of Thermo Fisher Scientific, Waltham, Massachusetts, USA) were used to genotype for APOE genotypes p.Cys130Arg (rs429358, legacy name Cys112Arg, c.388T>C) and p.Arg176Cys (rs7412, legacy name Arg158Cys, c.526C>T).

      2.5 Categories of physical activity

      Baseline characteristics for the individuals were collected at the date of first examination from a questionnaire, physical examination, and blood measurements. Physical activity in leisure time and physical activity at work were self-reported. Physical activity in leisure time was in three categories: high physical activity (serving as the reference) including individuals reporting either > 4 h of light physical activity/week and/or ≥ 2 h of vigorous physical activity/week; moderate physical activity including individuals reporting 2 to 4 h of light physical activity/week; and low physical activity including individuals that reported < 2 h of light physical activity/week. Physical activity at work was also in three categories: seated or standing work – occasionally walking; walking – occasionally with lifting; and physically heavy work. Other covariates are described in detail in legend to Table 1.
      Table 1Characteristics of study participants by category of physical activity in leisure time in the Copenhagen General Population Study and the Copenhagen City Heart Study.
      HighModerateLow
      No. of individuals (%)56,019 (48)51,872 (44)9,725 (8)
      Age (years)57 (47–67)59 (49–68)55 (46–65)*
      Female (%)506053*
      Body mass index (kg/m2)25 (23–28)26 (23–29)27 (24–30)*
      Education <8 years (%)111729*
      Diabetes (%)356*
      Hypertension (%)566060*
      Smoking (%)596677*
      High alcohol intake (%)212328*
      Lipid-lowering therapy (%)9119*
      Hormonal replacement therapy (%)†101111*
      Physical inact work, group (%)767777‡
       Sitting/standing (%)767777*
       Walking (%)202018*
       Heavy lifting (%)435*
      APOE-genotype (N = 105,237)
       ϵ22 (%)111
       ϵ32 (%)121212
       ϵ33 (%)555656
       ϵ42 (%)333
       ϵ43 (%)262525
       ϵ44 (%)333
      High physical activity included individuals reporting either > 4 h of light physical activity/week or ≥ 2 h of vigorous physical activity/week, moderate physical activity included individuals reporting 2 to 4 h of light physical activity/week, and low physical activity included individuals that reported < 2 h of light physical activity/week. Values are numbers, median (interquartile range) or percent, and are from the day of enrollment (2003–2014 for the Copenhagen General Population Study and 1976–1978, 1981–1983, 1991–1994, or 2001–2003 for the Copenhagen City Heart Study). Diabetes mellitus was self-reported disease, use of insulin or oral hypoglycemic agents, non-fasting plasma glucose levels of more than 11 mmol/L (198 mg/dL) and/or a diagnosis of diabetes mellitus at baseline from the national Danish Patient Registry. Hypertension was use of anti-hypertensive medication, a systolic blood pressure of 140 mm Hg or greater, and/or a diastolic blood pressure of 90 mm Hg or greater. Smoking was never/ever smoker. High alcohol intake was >14/21 units per week for women/men (1 unit = 12 g alcohol, equivalent to one glass of wine or one beer (33 cL)). Lipid-lowering therapy was primarily statins (yes/no). Women reported menopausal status and use of hormonal replacement therapy. Physical inactivity at work was primarily sitting or standing work. †In women only. *P < 0.001 by Kruskal-Wallis one-way analysis of variance or Pearson's χ2-test. ‡P < 0.05 by Pearson's χ2-test.

      2.6 Statistical analysis

      We used Stata/S.E. v14.2 and 16 (Stata Corp, College Station, TX). Probability values < 0.001 are given as powers of 10. Kruskal-Wallis one-way analysis of variance, Mann-Whitney U test or Pearson's χ2 test were used to evaluate continuous and categorical variables by exposure status. Missing data on covariates were imputed from age, sex, and population by multiple imputing. Missing values were less than 0.75% except for smoking status and physical activity at work where percentages were 2.5% and 27.7%, respectively. Consequently, analyses only including individuals with reported physical activity at work were also performed. Cox proportional hazards regression models (with censoring at death) with age as time scale and left truncation (delayed entry) were used to estimate hazard ratios for the association between physical activity level and risk of non-Alzheimer's dementia and Alzheimer's disease. For Cox regression models, proportionality of hazards over time were assessed by plotting –ln(-ln[survival]) versus ln(analysis time). There was no suspicion of nonproportionality. To test whether reverse causation affected our results, sensitivity analyses were made only including individuals with more than two, five, and ten years of follow-up, respectively. 10-year absolute risks of non-Alzheimer's dementia were calculated using competing risk regression based on Fine and Gray proportional subhazards model [
      • Fine J.P.
      • Gray R.J.
      A proportional hazards model for the subdistribution of a competing risk.
      ], to account for the possibility of death or emigration as competing events. Fine and Gray proportional subhazards model was chosen because a competing event prevents the event of interest, which is highly relevant with diseases of late life, while censoring merely obstructs the observation of the event of interest. Interactions were tested using an interaction term (physical activity x covariate in two groups) in a multifactorially adjusted model. A likelihood ratio test compared the multifactorially adjusted model without and with the interaction term.

      3. Results

      Baseline characteristics of participants are shown by physical activity in leisure time in Table 1 and by physical activity at work in Supplementary Table 1.

      3.1 Physical inactivity in leisure time

      Multifactorially adjusted hazard ratios for low versus high physical activity in leisure time were 1.60 (95% CI 1.40–1.83; p for trend = 4*10−11) for non-Alzheimer's dementia and 0.94 (0.80–1.11; p for trend = 0.19) for Alzheimer's disease (Fig. 1, left panel). When further adjusting for APOE genotype or physical activity at work, corresponding hazard ratios were 1.82 (1.54–2.15; p for trend = 5*10−11) and 1.60 (1.40–1.83; p for trend = 4*10−11) for non-Alzheimer's dementia and 1.02 (0.86–1.22; p for trend = 0.60) and 0.94 (0.80–1.12; p for trend = 0.23) for Alzheimer's disease, respectively (Fig. 1, middle and right panel). When only including individuals with reported physical activity at work, results were similar for non-Alzheimer's dementia (Supplementary Fig. 2, top panel). When stratifying on sex, corresponding multifactorially adjusted hazard ratios for non-Alzheimer's dementia were 1.99 (1.63–2.44; p for trend = 8*10−12) in men and 1.33 (1.11–1.59; p for trend = 0.007) in women (Supplementary Fig. 3, left panel). These results were similar after further adjustment for APOE genotype or physical activity at work (Supplementary Fig. 3, middle and right panel).
      Fig. 1
      Fig. 1Physical inactivity in leisure time and risk of non-Alzheimer's dementia and Alzheimer's disease.
      Individuals with all-cause dementia before blood sampling were excluded, leaving a total of 117,616 individuals for analyses. A total of 105,237 individuals with APOE genotype were included in the middle panel. High physical activity (serving as the reference) included individuals reporting either > 4 h of light physical activity/week or ≥ 2 h of vigorous physical activity/week. Moderate physical activity included individuals reporting 2 to 4 h of light physical activity/week, and low physical activity included individuals that reported < 2 h of light physical activity/week. Hazard ratios were multifactorially adjusted for age (as time scale), sex, body mass index, diabetes mellitus, hypertension, education, smoking, alcohol intake, lipid-lowering therapy, postmenopausal hormonal replacement therapy in women, and study population. Furthermore, the middle panel was adjusted for APOE genotype, and the right panel for physical activity at work. p for trend was from Cox regression. APOE = apolipoprotein E gene; APOE genotype = ε2/ε3/ε4 APOE genotype; CI = confidence interval.
      In the age group 70–79 years, 10-year absolute risks for non-Alzheimer's dementia were 5% in individuals with low physical activity in leisure time and 3% in individuals with high physical activity in leisure time. For the age group 80–100 years, corresponding values were 15% and 8% (Supplementary Table 2).

      3.2 Exclusion of individuals with follow-up less than 2–10 years

      Multifactorially and APOE adjusted hazard ratios for non-Alzheimer's dementia for low versus high physical activity in leisure time after exclusion of individuals with follow-up periods less than two, five, and ten years from baseline were 1.75 (1.47–2.08; p for trend = 2*10−9), 1.54 (1.27–1.86; p for trend = 4*10−5), and 1.40 (1.12–1.76; p for trend = 0.01), respectively (Fig. 2, middle panel). Results were similar without APOE adjustment or with adjustment for physical activity at work (Fig. 2, left and right panel), or when only including individuals with reported physical activity at work (Supplementary Fig. 2, bottom three panels).
      Fig. 2
      Fig. 2Physical inactivity in leisure time and risk of non-Alzheimer's dementia after exclusion of individuals with follow-up periods within two, five, or ten years from baseline.
      Individuals with follow-up periods within two, five, and ten years from baseline were excluded, leaving a total of 116,073, 112,348, and 63,539 individuals for analyses, respectively. High physical activity (serving as the reference) included individuals reporting either > 4 h of light physical activity/week or ≥ 2 h of vigorous physical activity/week. Moderate physical activity included individuals reporting 2 to 4 h of light physical activity/week, and low physical activity included individuals that reported < 2 h of light physical activity/week. Hazard ratios were multifactorially adjusted for age (as time scale), body mass index, diabetes mellitus, hypertension, education, smoking, alcohol intake, lipid-lowering therapy, postmenopausal hormonal replacement therapy in women, and study population. Furthermore, the middle panel was adjusted for APOE genotype, and the right panel for physical activity at work. p for trend was from Cox regression. APOE = apolipoprotein E gene; APOE genotype = ε2/ε3/ε4 APOE genotype; CI = confidence interval.

      3.3 Combined physical activity in leisure time and at work

      Multifactorially adjusted hazard ratios for non-Alzheimer's dementia as a function of physical activity in leisure time (high, moderate, low) stratified by increasing physical activity at work (sitting/standing at work, walking at work, or heavy lifting at work) are shown in Fig. 3, upper panel. The risk of non-Alzheimer's dementia increased with lower levels of physical activity in leisure time and possibly with higher levels of physical activity at work, independent of each other (Fig. 3; p for interaction = 0.87). Compared with those with high physical activity in leisure time and low physical activity at work (sitting/standing at work, left green bar), individuals with low physical activity in leisure time (red bars) and increasing physical activity at work had stepwise higher hazard ratios of, respectively, 1.34 (1.08–1.66), 1.49 (1.07–2.07), and 1.89 (1.11–3.21) for non-Alzheimer's dementia (Fig. 3, red bars left to right: p for trend = 5*10−4).
      Fig. 3
      Fig. 3Combined effect of physical activity in leisure time and at work on risk of non-Alzheimer's dementia.
      Only individuals with reported physical activity in leisure time and at work, and individuals without all-cause dementia before blood sampling were included in the analyses (N = 84,708). High physical activity in leisure time included individuals reporting either > 4 h of light physical activity/week or ≥ 2 h of vigorous physical activity/week. Moderate physical activity included individuals reporting 2 to 4 h of light physical activity/week, and low physical activity included individuals that reported < 2 h of light physical activity/week. Physical activity at work was in three categories: seated or standing work, primarily walking at work, and work with heavy lifting. Hazard ratios were multifactorially adjusted for age (as time scale), sex, body mass index, diabetes mellitus, hypertension, education, smoking, alcohol intake, lipid-lowering therapy, postmenopausal hormonal replacement therapy in women, and study population. P for trend was from Cox regression with high physical activity in leisure time and sitting/standing at work as the reference through the categories of first high physical activity in leisure time, then through moderate physical activity in leisure time and lastly through the categories of low physical activity in leisure time. p for interaction was from likelihood ratio test.
      For physical activity at work alone, multifactorially adjusted hazard ratios for heavy lifting at work versus sitting/standing at work were 1.21 (0.93–1.58; p for trend = 0.26) for non-Alzheimer's dementia (Supplementary Fig. 4, upper, left panel) and 1.41 (1.00–1.99; p for trend = 2*10−4) for Alzheimer's disease (Supplementary Fig. 4, lower, left panel). When further adjusting for APOE genotype or physical activity in leisure time results were similar (Supplementary Fig. 4, middle and right panels).

      3.4 Positive control

      The multifactorially adjusted hazard ratio for ischemic heart disease for low versus high physical activity in leisure time was 1.40 (1.32–1.49; p for trend = 1*10−25) (Fig. 4, top left panel). After exclusion of individuals with follow-up periods less than two, five, and ten years from baseline, corresponding multifactorially adjusted hazard ratios were 1.37 (1.29–1.46; p for trend = 4*10−19), 1.32 (1.23–1.42; p for trend = 2*10−12), and 1.35 (1.23–1.48; p for trend = 2*10−9), respectively (Fig. 4, left panel). These results were similar after further adjustment for APOE genotype or physical activity at work (Fig. 4, middle and right panel). When stratifying on sex, these results were similar for both men and women (Supplementary Fig. 5).
      Fig. 4
      Fig. 4Physical inactivity in leisure time and risk of ischemic heart disease.
      Individuals with ischemic heart disease before blood sampling were excluded, leaving a total of 111,555 individuals for analyses in the top panel. Individuals with follow-up periods within two, five, and ten years from baseline were excluded, leaving a total of 109,036, 104,205, and 53,218 individuals for analyses in the three bottom panels, respectively. High physical activity (serving as the reference) included individuals reporting either > 4 h of light physical activity/week or ≥ 2 h of vigorous physical activity/week. Moderate physical activity included individuals reporting 2 to 4 h of light physical activity/week, and low physical activity included individuals that reported < 2 h of light physical activity/week. Hazard ratios were multifactorially adjusted for age (as time scale), body mass index, diabetes mellitus, hypertension, education, smoking, alcohol intake, lipid-lowering therapy, postmenopausal hormonal replacement therapy in women, and study population. Furthermore, the middle panel was adjusted for APOE genotype, and the right panel for physical activity at work. p for trend was from Cox regression. APOE = apolipoprotein E gene; APOE genotype = ε2/ε3/ε4 APOE genotype; CI = confidence interval.

      3.5 Interaction

      No interactions between low versus high physical activity in leisure time and covariates in the association with risk of non-Alzheimer's dementia fulfilled a Bonferroni corrected level of significance (p-level divided by number of covariates: <0.05/9 = 0.0056). This was supported by largely similar hazard ratios after stratifying on covariate categories (Supplementary Fig. 6); however, we cannot completely exclude that the hazard ratio is larger in men than in women.

      4. Discussion

      The principal finding of this study is that physical inactivity in leisure time is associated with increased risk of non-Alzheimer's dementia, independent of physical activity at work and APOE genotype. The association remained after exclusion of individuals with follow-up periods within two, five, or ten years from baseline, suggesting that the present finding is not due to reverse causation. Combining physical activity in leisure time and at work, physical activity in leisure time had the strongest relationship with risk of non-Alzheimer's dementia. These findings are novel.
      To the best of our knowledge, this is the first large prospective study to assess the association of physical inactivity in leisure time alone, adjusted for, or in combination with physical activity at work with risk of non-Alzheimer's dementia, a disease primarily thought to be caused by vascular pathology. Previous studies have focused on all-cause dementia and/or Alzheimer's disease and report physical inactivity to be associated with increased risk of these disease entities. [
      • Sofi F.
      • Valecchi D.
      • Bacci D.
      • et al.
      Physical activity and risk of cognitive decline: a meta-analysis of prospective studies.
      ,
      • Hamer M.
      • Chida Y.
      Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence.
      ,
      • Groot C.
      • Hooghiemstra A.M.
      • Raijmakers P.G.H.M.
      • et al.
      The effect of physical activity on cognitive function in patients with dementia: a meta-analysis of randomized control trials.
      ,
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ,
      • Sabia S.
      • Dugravot A.
      • Dartigues J.F.
      • et al.
      Physical activity, cognitive decline, and risk of dementia: 28 year follow-up of Whitehall II cohort study.
      ,
      • Beydoun M.A.
      • Beydoun H.A.
      • Gamaldo A.A.
      • Teel A.
      • Zonderman A.B.
      • Wang Y.
      Epidemiologic studies of modifiable factors associated with cognition and dementia: systematic review and meta-analysis.
      ] These observations were, however, likely due to reverse causation – the fact that people might stop exercising due to prodromal phases of dementia – since by restricting analyses to a minimum follow-up of 10 years from the reported level of physical activity and assessment of dementia diagnoses, the associations disappeared. [
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ] Interestingly, the observations tended to remain in individuals with comorbid cardiometabolic disease [
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ], indirectly supporting the present results, where the signal is clear for non-Alzheimer's dementia even after exclusion of initial follow-up periods. This may suggest involvement of physical activity in causal pathways - whether it is directly causal or whether the associations are mediated by causal cardiovascular risk factors remains to be firmly established. The causal nature of the association can be tested using Mendelian randomization strategies, where genetic variants are used as unconfounded proxies for the modifiable exposure of interest, when powerful, non-pleiotropic genetic instruments are available. [
      • Holmes M.V.
      • Ala-Korpela M.
      • Smith G.D.
      Mendelian randomization in cardiometabolic disease: challenges in evaluating causality.
      ] One Mendelian randomization study investigating objectively measured physical activity and risk of Alzheimer's disease concluded non-causality. [
      • Baumeister S.E.
      • Karch A.
      • Bahls M.
      • Teumer A.
      • Leitzmann M.F.
      • Baurecht H.
      Physical activity and risk of Alzheimer's disease: a two-sample Mendelian randomization study.
      ] One genome-wide association study identified genetic variants associated with self-reported physical activity [
      • Klimentidis Y.C.
      • Raichlen D.A.
      • Bea J.
      • et al.
      Genome-wide association study of habitual physical activity in over 377,000 UK Biobank participants identifies multiple variants including CADM2 and APOE.
      ], these genetic variants were, however, very weak instruments and were thus not optimal for Mendelian randomization analyses. [
      • Baumeister S.E.
      • Karch A.
      • Bahls M.
      • Teumer A.
      • Leitzmann M.F.
      • Baurecht H.
      Physical activity and risk of Alzheimer's disease: a two-sample Mendelian randomization study.
      ] When strong non-pleiotropic genetic variants are identified and when powerful GWAS studies are available for non-Alzheimer's dementia, we will revisit the Mendelian randomization strategies.
      There are several possible biological mechanisms behind an effect of physical activity. Physical activity is thought to act indirectly through modification of other joint dementia and cardiovascular risk factors [
      • Sofi F.
      • Valecchi D.
      • Bacci D.
      • et al.
      Physical activity and risk of cognitive decline: a meta-analysis of prospective studies.
      ,
      • Hamer M.
      • Chida Y.
      Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence.
      ,
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ] e.g. by lowering blood pressure, plasma lipoproteins, and inflammatory markers and possibly prevent obesity and diabetes. [
      • Kivipelto M.
      • Ngandu T.
      • Fratiglioni L.
      • et al.
      Obesity and vascular risk factors at midlife and the risk of dementia and Alzheimer disease.
      ,
      • Rosendorff C.
      • Beeri M.S.
      • Silverman J.M.
      Cardiovascular risk factors for Alzheimer's disease.
      ] Of direct mechanisms, physical activity is thought to have neurotrophic effects yielding angiogenesis, synaptogenesis and neuronal growth and survival, possibly leading to increased brain plasticity and cognitive reserve, and maintained blood-brain barrier integrity. [
      • Sofi F.
      • Valecchi D.
      • Bacci D.
      • et al.
      Physical activity and risk of cognitive decline: a meta-analysis of prospective studies.
      ,
      • Hamer M.
      • Chida Y.
      Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence.
      ,
      • Wolters F.J.
      • Zonneveld H.I.
      • Hofman A.
      • et al.
      Cerebral perfusion and the risk of dementia: a population-based study.
      ,
      • Bailey D.M.
      • Marley C.J.
      • Brugniaux J.V.
      • et al.
      Elevated aerobic fitness sustained throughout the adult lifespan is associated with improved cerebral hemodynamics.
      ,
      • Zlokovic B.V.
      Neurovascular pathways to neurodegeneration in Alzheimer's disease and other disorders.
      ] Furthermore, physical activity is suggested to attenuate the inevitable, age-related decline in cerebral blood flow and a consistent relationship between observed maximal oxygen uptake (VO2max) and cerebral blood flow confirms that the benefits of aerobic exercise are not limited to the cardiovascular circulation but also applies to the cerebrovascular circulation, supporting exercise prescription in the elderly. [
      • Wolters F.J.
      • Zonneveld H.I.
      • Hofman A.
      • et al.
      Cerebral perfusion and the risk of dementia: a population-based study.
      ,
      • Bailey D.M.
      • Marley C.J.
      • Brugniaux J.V.
      • et al.
      Elevated aerobic fitness sustained throughout the adult lifespan is associated with improved cerebral hemodynamics.
      ].
      Strengths of our study include the prospective design and the large, well-characterized, ethnically homogeneous cohorts with follow-up of up to 43 years and no losses to follow-up, enabling us to address the possibility of reverse causation in the association between physical inactivity and risk of non-Alzheimer's dementia. This is important, since prior longitudinal analyses with repeat data have shown that physical activity declines in the preclinical stage of dementia up to nine years before diagnosis. [
      • Sabia S.
      • Dugravot A.
      • Dartigues J.F.
      • et al.
      Physical activity, cognitive decline, and risk of dementia: 28 year follow-up of Whitehall II cohort study.
      ] Further, the availability of robust information on physical activity at work enabled us to confirm that increasing levels of work-related physical activity is associated with increased risk of dementia. [
      • Nabe‐Nielsen K.
      • Holtermann A.
      • Gyntelberg F.
      • et al.
      The effect of occupational physical activity on dementia: results from the Copenhagen Male Study.
      ] We extended these findings to both non-Alzheimer's dementia and Alzheimer's disease and observed similar findings in both women and men. That increased physical activity at work is associated with increased risk - in contrast to increased physical activity in leisure time - is possibly explained by a too low intensity of the physical activity at work to maintain or improve cardiovascular health. [
      • Nabe‐Nielsen K.
      • Holtermann A.
      • Gyntelberg F.
      • et al.
      The effect of occupational physical activity on dementia: results from the Copenhagen Male Study.
      ,
      • Holtermann A.
      • Krause N.
      • Van Der Beek A.J.
      • Straker L.
      The physical activity paradox: six reasons why occupational physical activity (OPA) does not confer the cardiovascular health benefits that leisure time physical activity does.
      ] This phenomenon is also observed for cardiovascular disease. [
      • Holtermann A.
      • Schnohr P.
      • Nordestgaard B.G.
      • Marott J.L.
      The physical activity paradox in cardiovascular disease and all-cause mortality: the contemporary Copenhagen General Population Study with 104,046 adults.
      ,
      • Holtermann A.
      • Krause N.
      • Van Der Beek A.J.
      • Straker L.
      The physical activity paradox: six reasons why occupational physical activity (OPA) does not confer the cardiovascular health benefits that leisure time physical activity does.
      ,
      • Holtermann A.
      • Marott J.L.
      • Gyntelberg F.
      • et al.
      Self-reported occupational physical activity and cardiorespiratory fitness: importance for cardiovascular disease and all-cause mortality.
      ] However, when combining the two types of physical activity, physical activity in leisure time had the strongest relationship with risk of non-Alzheimer's dementia making it the most efficient and likely the most realistic physical activity instrument to target for intervention.
      The present study has potential limitations that need to be addressed. The use of registry-based diagnoses suffers from potential underdiagnosis, since only patients referred to a hospital are included. Using hospital record data, however, reduces attrition bias [
      • Kivimäki M.
      • Singh-Manoux A.
      • Pentti J.
      • et al.
      Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
      ] and the national Danish registries are regarded among the best of its kind [
      • Lynge E.
      • Sandegaard J.L.
      • Rebolj M.
      The Danish national patient register.
      ,
      • Helweg-Larsen K.
      The Danish register of causes of death.
      ], with high quality of the Alzheimer's disease diagnosis. [
      • Phung T.K.T.
      • Andersen B.B.
      • Høgh P.
      • Kessing L.V.
      • Mortensen P.B.
      • Waldemar G.
      Validity of dementia diagnoses in the Danish hospital registers.
      ] Also in support of a reasonable quality of diagnoses in the present cohorts, we ensured by scrutinizing department codes for all events that 91–93% of all-cause dementia diagnoses were registered at dementia clinics, other neurological outpatient clinics, departments of neurology and neurosurgery, in- and outpatient clinics at departments of geriatrics, or departments of internal medicine. [
      • Rasmussen K.L.
      • Tybjærg-Hansen A.
      • Nordestgaard B.G.
      • Frikke-Schmidt R.
      Plasma apolipoprotein E levels and risk of dementia: a Mendelian randomization study of 106,562 individuals.
      ] The generalizability of the study may be limited as we only studied white individuals from the general population of Denmark.
      In conclusion, in a general population setting physical inactivity in leisure time was associated with increased risk of non-Alzheimer's dementia, but not Alzheimer's disease, independent of modifiable risk factors and physical activity at work. The association remained after exclusion of individuals with follow-up periods within two, five, and ten years from baseline, suggesting that these findings are not due to reverse causation. The present study thus provides evidence for public health advice on physical activity in leisure time for the large part of dementia that mainly is due to vascular factors.

      Financial support

      This work was supported by the Danish Heart Foundation , the Lundbeck Foundation (grant no. R278-2018-804 ) and Jeppe and Ovita Juhl's Fund. The funding sources had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. In addition, the researchers were fully independent from the funding sources and all authors had full access to all of the data in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis.

      Author contributions

      I.J.R.: Study concept and design, acquisition of data, statistical analysis, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, final approval for submission.
      K.L.R.: Study concept and design, acquisition of data, statistical analysis, analysis and interpretation of data, critical revision of the manuscript for important intellectual content, final approval for submission.
      J.Q.T.: Acquisition of data, statistical analysis, analysis and interpretation of data, critical revision of the manuscript for important intellectual content, final approval for submission.
      B.G.N.: Acquisition of data, critical revision of the manuscript for important intellectual content, obtained funding, administrative, technical, and material support, final approval for submission.
      P.S.: Acquisition of data, critical revision of the manuscript for important intellectual content, obtained funding, administrative, technical, and material support, final approval for submission.
      A.T.-H.: Acquisition of data, critical revision of the manuscript for important intellectual content, obtained funding, administrative, technical, and material support, final approval for submission.
      R.F.-S.: Study concept and design, acquisition of data, statistical analysis, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content, obtained funding, administrative, technical, and material support, study supervision, final approval for submission, accountable for all aspects of the work.

      Declaration of interest

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Acknowledgements

      We thank the staff and participants of the CCHS and CGPS for their important contributions.

      Appendix A. Supplementary data

      The following is the Supplementary data to this article:

      References

        • Livingston G.
        • Huntley J.
        • Sommerlad A.
        • et al.
        Dementia prevention, intervention, and care: 2020 report of the Lancet Commission.
        Lancet. 2020; 396: 413-446
        • Sofi F.
        • Valecchi D.
        • Bacci D.
        • et al.
        Physical activity and risk of cognitive decline: a meta-analysis of prospective studies.
        J. Intern. Med. 2011; 269: 107-117
        • Hamer M.
        • Chida Y.
        Physical activity and risk of neurodegenerative disease: a systematic review of prospective evidence.
        Psychol. Med. 2008; 39: 3-11
        • Groot C.
        • Hooghiemstra A.M.
        • Raijmakers P.G.H.M.
        • et al.
        The effect of physical activity on cognitive function in patients with dementia: a meta-analysis of randomized control trials.
        Ageing Res. Rev. 2016; (25(2016)): 13-23
        • Kivimäki M.
        • Singh-Manoux A.
        • Pentti J.
        • et al.
        Physical inactivity, cardiometabolic disease, and risk of dementia: an individual-participant meta-analysis.
        BMJ. 2019; 365: l1495
        • Xu W.
        • Wang H.F.
        • Wan Y.
        • Tan C.C.
        • Yu J.T.
        • Tan L.
        Leisure time physical activity and dementia risk: a dose-response meta-analysis of prospective studies.
        BMJ Open. 2017; 7: 1-10
        • Young J.
        • Angevaren M.
        • Rusted J.
        • Tabet N.
        Aerobic exercise to improve cognitive function in older people without known cognitive impairment.
        Cochrane Database Syst. Rev. 2015; 4
        • Zotcheva E.
        • Bergh S.
        • Selbæk G.
        • et al.
        Midlife physical activity, psychological distress, and dementia risk: the HUNT study.
        J. Alzheim. Dis. 2018; 66: 825-833
        • Petersen C.B.
        • Grønbæk M.
        • Helge J.W.
        • Thygesen L.C.
        • Schnohr P.
        • Tolstrup J.S.
        Changes in physical activity in leisure time and the risk of myocardial infarction, ischemic heart disease, and all-cause mortality.
        Eur. J. Epidemiol. 2012; 27: 91-99
        • Holtermann A.
        • Schnohr P.
        • Nordestgaard B.G.
        • Marott J.L.
        The physical activity paradox in cardiovascular disease and all-cause mortality: the contemporary Copenhagen General Population Study with 104,046 adults.
        Eur. Heart J. 2021; 42: 1499-1511
        • Juul Rasmussen I.
        • Rasmussen K.L.
        • Nordestgaard B.G.
        • Tybjærg-Hansen A.
        • Frikke-Schmidt R.
        Impact of cardiovascular risk factors and genetics on 10-year absolute risk of dementia: risk charts for targeted prevention.
        Eur. Heart J. 2020; 41: 4024-4033
        • Rasmussen K.L.
        • Tybjærg-Hansen A.
        • Nordestgaard B.G.
        • Frikke-Schmidt R.
        Plasma levels of Apolipoprotein E, APOE genotype, and all-cause and cause-specific mortality in 105 949 individuals from a white general population cohort.
        Eur. Heart J. 2019; 40: 2813-2824
        • Jørgensen A.B.
        • Frikke-Schmidt R.
        • Nordestgaard B.G.
        • Tybjærg-Hansen A.
        Loss-of-function mutations in APOC3 and risk of ischemic vascular disease.
        N. Engl. J. Med. 2014; 371: 32-41
        • Phung T.K.T.
        • Andersen B.B.
        • Høgh P.
        • Kessing L.V.
        • Mortensen P.B.
        • Waldemar G.
        Validity of dementia diagnoses in the Danish hospital registers.
        Dement. Geriatr. Cognit. Disord. 2007; 24: 220-228
        • Fine J.P.
        • Gray R.J.
        A proportional hazards model for the subdistribution of a competing risk.
        J. Am. Stat. Assoc. 1999; 94: 496-509
        • Sabia S.
        • Dugravot A.
        • Dartigues J.F.
        • et al.
        Physical activity, cognitive decline, and risk of dementia: 28 year follow-up of Whitehall II cohort study.
        BMJ. 2017; 357: 1-12
        • Beydoun M.A.
        • Beydoun H.A.
        • Gamaldo A.A.
        • Teel A.
        • Zonderman A.B.
        • Wang Y.
        Epidemiologic studies of modifiable factors associated with cognition and dementia: systematic review and meta-analysis.
        BMC Publ. Health. 2014; 14: 643
        • Holmes M.V.
        • Ala-Korpela M.
        • Smith G.D.
        Mendelian randomization in cardiometabolic disease: challenges in evaluating causality.
        Nat. Rev. Cardiol. 2017; 14: 577-599
        • Baumeister S.E.
        • Karch A.
        • Bahls M.
        • Teumer A.
        • Leitzmann M.F.
        • Baurecht H.
        Physical activity and risk of Alzheimer's disease: a two-sample Mendelian randomization study.
        Neurology. 2020; 95: e1897-e1905
        • Klimentidis Y.C.
        • Raichlen D.A.
        • Bea J.
        • et al.
        Genome-wide association study of habitual physical activity in over 377,000 UK Biobank participants identifies multiple variants including CADM2 and APOE.
        Int. J. Obes. 2018; 42: 1161-1176
        • Kivipelto M.
        • Ngandu T.
        • Fratiglioni L.
        • et al.
        Obesity and vascular risk factors at midlife and the risk of dementia and Alzheimer disease.
        Arch. Neurol. 2005; 62: 1556-1560
        • Rosendorff C.
        • Beeri M.S.
        • Silverman J.M.
        Cardiovascular risk factors for Alzheimer's disease.
        Am. J. Geriatr. Cardiol. 2007; 16: 143-149
        • Wolters F.J.
        • Zonneveld H.I.
        • Hofman A.
        • et al.
        Cerebral perfusion and the risk of dementia: a population-based study.
        Circulation. 2017; 136: 719-728
        • Bailey D.M.
        • Marley C.J.
        • Brugniaux J.V.
        • et al.
        Elevated aerobic fitness sustained throughout the adult lifespan is associated with improved cerebral hemodynamics.
        Stroke. 2013; 44: 3235-3238
        • Zlokovic B.V.
        Neurovascular pathways to neurodegeneration in Alzheimer's disease and other disorders.
        Nat. Rev. Neurosci. 2011; 12: 723-738
        • Nabe‐Nielsen K.
        • Holtermann A.
        • Gyntelberg F.
        • et al.
        The effect of occupational physical activity on dementia: results from the Copenhagen Male Study.
        Scand. J. Med. Sci. Sports. 2020; (00): 1-10
        • Holtermann A.
        • Krause N.
        • Van Der Beek A.J.
        • Straker L.
        The physical activity paradox: six reasons why occupational physical activity (OPA) does not confer the cardiovascular health benefits that leisure time physical activity does.
        Br. J. Sports Med. 2018; 52: 149-150
        • Holtermann A.
        • Marott J.L.
        • Gyntelberg F.
        • et al.
        Self-reported occupational physical activity and cardiorespiratory fitness: importance for cardiovascular disease and all-cause mortality.
        Scand. J. Work. Environ. Health. 2016; 42: 291-298
        • Lynge E.
        • Sandegaard J.L.
        • Rebolj M.
        The Danish national patient register.
        Scand. J. Publ. Health. 2011; 39: 30-33
        • Helweg-Larsen K.
        The Danish register of causes of death.
        Scand. J. Publ. Health. 2011; 39: 26-29
        • Rasmussen K.L.
        • Tybjærg-Hansen A.
        • Nordestgaard B.G.
        • Frikke-Schmidt R.
        Plasma apolipoprotein E levels and risk of dementia: a Mendelian randomization study of 106,562 individuals.
        Alzheimer's Dementia. 2018; 14: 71-80