Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Weather, day length and physical activity in older adults: Cross-sectional results from the European Prospective Investigation into Cancer and Nutrition (EPIC) Norfolk Cohort

  • Yu-Tzu Wu ,

    Current address: REACH: The Centre for Research in Ageing and Cognitive Health, Department of Psychology, College of Life and Environmental Sciences, University of Exeter Washington Singer Building, Exeter, United Kingdom

    Affiliations Department of Population Health & Primary Care, Norwich Medical School, University of East Anglia, Norwich, United Kingdom, UKCRC Centre for Diet and Activity Research (CEDAR), Institute of Public Health, University of Cambridge, Forvie Site, Robinson Way, Cambridge, United Kingdom

  • Robert Luben,

    Affiliation Department of Public Health and Primary Care, Institute of Public Health, University of Cambridge, Cambridge, United Kingdom

  • Nicholas Wareham,

    Affiliation UKCRC Centre for Diet and Activity Research (CEDAR), Institute of Public Health, University of Cambridge, Forvie Site, Robinson Way, Cambridge, United Kingdom

  • Simon Griffin,

    Affiliation UKCRC Centre for Diet and Activity Research (CEDAR), Institute of Public Health, University of Cambridge, Forvie Site, Robinson Way, Cambridge, United Kingdom

  • Andy P. Jones

    Affiliations Department of Population Health & Primary Care, Norwich Medical School, University of East Anglia, Norwich, United Kingdom, UKCRC Centre for Diet and Activity Research (CEDAR), Institute of Public Health, University of Cambridge, Forvie Site, Robinson Way, Cambridge, United Kingdom



A wide range of environmental factors have been related to active ageing, but few studies have explored the impact of weather and day length on physical activity in older adults. We investigate the cross-sectional association between weather conditions, day length and activity in older adults using a population-based cohort in England, the European Prospective Investigation into Cancer and Nutrition (EPIC) Norfolk study.


Physical activity was measured objectively over 7 days using an accelerometer and this was used to calculate daily total physical activity (counts per minute), daily minutes of sedentary behaviour and light, moderate and vigorous physical activity (LMVPA). Day length and two types of weather conditions, precipitation and temperature, were obtained from a local weather station. The association between these variables and physical activity was examined by multilevel first-order autoregressive modelling.


After adjusting for individual factors, short day length and poor weather conditions, including high precipitation and low temperatures, were associated with up to 10% lower average physical activity (p<0.01) and 8 minutes less time spent in LMVPA but 15 minutes more sedentary time, compared to the best conditions.


Day length and weather conditions appear to be an important factor related to active ageing. Future work should focus on developing potential interventions to reduce their impact on physical activity behaviours in older adults.


There has been particular interest in the potential for the environment to support active ageing, the process of optimising opportunities for health and well-being as individuals grow older [1]. A wide range of environmental factors have been related to physical activity in older adults [2,3]. Although the idea of age-friendly environments has been promoted worldwide, projects have mainly focused on the characteristics of the built and social environment [4]. Nevertheless other environmental conditions that are out of direct control of planners, such as weather conditions and day length, may interact with features of these environments to influence individual activity levels [5] and might thus have an impact on active ageing [6].

Literature in the field of environmental gerontology has proposed the ‘Environmental Press Model’, suggesting that adults with reduced individual competence, such as the aged, are more sensitive to stress from the environment and that this may lead to maladaptive behaviours and poor health [7]. Adverse weather conditions, such as heavy rain, low temperatures and short daylight hours, could be one potential source of environmental stress. Older adults, who are more likely to experience functional and health declines, might be especially sensitive to poor weather conditions, which have been reported to have a detrimental influence on physical activity in younger age groups [5,8,9]. In the elderly there is evidence from qualitative studies that weather and seasonal factors may be associated with concerns such as poor visibility and slippery surfaces [1012]. However, there is a lack of empirical evidence on the nature and magnitude of associations.

Only a small amount of research has used objective measures of physical activity to investigate associations with weather, climate and physical activity in older adults [1315]. Amongst 1324 German older adults, Klenk et al [14] showed linear relationships between the duration of walking and a wide range of weather variables including daylight, maximum temperature, total global radiation, average precipitation, average wind speed and average humidity. In a rural Scottish study of 548 adults, Witham et al [13] explored potential effect modifiers on the association between weather conditions and accelerometer derived activity levels and found higher daily temperature and longer day length were associated with higher activity levels. More recently, Prins & van Lenthe [15] used a GPS logger to determine associations between hourly weather conditions and walking and cycling behaviour among 43 older adults in the Netherlands, reporting a positive relationship between hourly temperature, walking and cycling minutes per hour.

Whilst these recent studies have given new insights into the potential role of weather as a determinant of physical activity in older adults, they have a number of limitations. One is that none took into account temporal autocorrelation when examining the association between weather and physical activity. This is a key methodological limitation because daily trends in both weather and physical activity are like to follow a temporally autocorrelated time series pattern whereby observations for one day are likely to be associated with those on the subsequent day. Failure to account for this in model specification leads to biases in model results [16]. Further, none of the studies examined different intensities of activity and in particular none looked at sedentary behaviour. Time spent sedentary has been particularly related to physical function, disability and metabolic syndrome in older age [17,18].

Using appropriate statistical methodologies for the analysis for time-series data, this study explores associations between weather conditions, day length and physical activity amongst large well-characterised population of older adults. The analysis is based on a population-based cohort in England: the European Prospective Investigation into Cancer and Nutrition (EPIC) Norfolk study, which collected objective measures of physical activity in over 4000 older adults between 2006 and 2011.

Materials and methods

Study population

The European Prospective Investigation into Cancer and Nutrition (EPIC) Norfolk study is one of population-based cohorts from the 10-county collaboration of the European Prospective Investigation into Cancer and Nutrition (EPIC), which was originally designed to examine the associations between diet and cancer. The scope of data collection has since been expanded to investigate major determinants of chronic disease, disability and death in middle and later life [19].

Details of the EPIC sampling and recruitment have been described elsewhere [20]. Briefly, EPIC Norfolk participants were recruited at baseline aged 45–74 between 1993 and 1997 from general practices across the county of Norfolk. In total 77630 individuals were invited and 30445 consented to take part. At the third health check, between September 2006 and December 2011, 8623 attended a health examination. Of these 4207 wore an accelerometer to measure their physical activity. The EPIC Norfolk study was approved by the Norfolk Local Research Ethics Committee (05/Q0101/191) and East Norfolk and Waveney National Health Service Research Governance Committee (2005EC07L) and written consent was obtained from participants [19]. This secondary data analysis does not require new IRB approval.

Measurement of physical activity

Physical activity was measured using a commercial accelerometer (Actigraph GT1M, Florida USA), which was set to a 5 second epoch. The EPIC Norfolk participants attending the third health check were invited to wear the accelerometer to measure their daily physical activity. Those who agreed to take part were instructed to wear the equipment for seven continuous days. Valid days were defined as those with evidence that the accelerometer was worn for at least 10 hours after screening out period of non-wear time, which was defined as continuous zero strings of ≥90 minutes duration. Participants with less than four valid days were excluded from the analysis. After excluding non-valid days and those with insufficient data, a total of 27446 person days of accelerometery were available for this research. The mean wear time was 869 (SD: 95) minutes per day in spring, 875 (SD: 89) in summer, 869 (SD: 99) in autumn and 865 (SD: 96) in winter.

Three types of physical activity measures were generated for each participant day using the accelerometer data. Mean daily counts per minute, a summarised indicator of daily activity level, were calculated using the total daily counts as recorded by the Actigraph divided by total wear minutes. Sedentary behaviour was defined by valid periods below 100 counts per minutes. As older adults are typically not vigorously active, the analysis presented here focused on light, moderate and vigorous physical activity (LMVPA) as opposed to the commonly employed moderate to vigorous activity (MVPA). LMVPA, which was defined as that over 1000 accelerometery counts per minute, includes any activities ranging from slow walking to vigorous exercise.

Environmental conditions: Day length and weather

Day length, precipitation and temperature have previously been suggested to be related to physical activity in older adults [13,14]. Hourly measurements of temperature and precipitation data were obtained from the Marham Norfolk weather station, which was the closest to the study area. Data from the weather station was obtained for each day during the study period, and used to calculate daily cumulative precipitation (mm) from 6am to 10pm and identify the maximum and minimum daytime temperature (Celsius) for the study period. In addition, day length (hours) was computed based on an algorithm that used latitude [21].

Trends across the variables were examined by classifying them into categories. Since a large number of days had no precipitation, days without rain (i.e. 0mm) were grouped into one category and those with some rain were divided into non-zero tertiles. The other three measures, maximum and minimum daytime temperature and day length, were categorised into quartiles.


Demographic information on gender and education was collected at the baseline. Education was divided into four levels: no education, O-level (10–11 years), A-level (12–13 years) and university degree or equivalent. Since adults with poor health tend to have lower level of physical activity, measures of self-rated health were obtained from the third health check questionnaires. Self-rated health was measured by the question “How would you rate your general health?”. Adults reporting excellent, very good and good health were categorised into one group and those reporting fair and poor were in the other group. This single question has been widely used in health research [22] and has also been recognised as a predictor of mortality [23].

Statistical analysis

The association between daily physical activity and daily weather conditions in the cohort was examined using regression models fitted a two level multilevel structure of days nested within individuals. Between days, the association between weather and physical activity was anticipated to exhibit temporal autocorrelation and hence multilevel first-order autoregressive modelling was employed [24].

Three types of models were fitted to the three measures of daily counts per minutes, sedentary time and LMVPA time. First, unadjusted associations between physical activity and weather conditions were examined, and then these were adjusted for individual level factors including age, gender, education and self-rated health. Finally, a full model including both individual level factors and weather conditions was fitted to investigate the independent association between weather conditions and physical activity. As variation in accelerometer wear time is likely to cause differences in recorded physical activity, daily minutes of time participants wore the accelerometers was added as a covariate for the models of sedentary behaviour and LMVPA. A significance level of p<0.05 was used in this study.


Table 1 shows descriptive characteristics of those 4051 participants with at least four valid days of physical activity data. The mean age was 69.0yrs with a range from 49 to 92yrs. The cohort was relatively well educated; almost 65% of participants had an A-level education or better. Under 15% of participants reported fair or poor health.

Table 1. Distributions of demographic factors and health status in the study sample (N,%).

The mean of daily counts per minute recorded was 256.1 (SD: 150.6) with a range from 3.8 to 1744.8. Mean recorded minutes of daily sedentary behaviour was high at nearly 679 mins (SD: 101.8), which equates to about 11 hours. The mean time spent in LMVPA per day was 73 mins (SD: 43.4) with a maximum of 374.4 mins recorded by one participant. Older age, being female, lower education and poorer self-rated health were generally associated with lower level of physical activity and increased daily sedentary time (Table 2).

Table 2. The associations between physical activity and demographic factors.

Precipitation, temperature and day length over the period of investigation are charted in Fig 1. Daily precipitation ranged from 0 to 26.2 mm with 54% of days being totally dry. The mean maximum and minimum daytime temperature in the study areas was 14.3 and 8.6 degrees Celsius with these two measures being highly correlated (r = 0.93, p<0.001). Day length ranged from 7.6 to 16.9 hours.

Fig 1. Changes in environmental conditions over the period of the study (red/blue: Maximum/Minimum temperature (°C); green: precipitation (mm); orange: day length (hour)).

Weather conditions, day length and physical activity

Fig 2 depicts mean values of the physical activity measures by different conditions examined. Daily counts per minute and LMVPA were higher with higher minimum and maximum temperature and day length and were lower with higher levels of daily precipitation. Daily LMVPA showed similar patterns to daily counts per minutes. Minutes of sedentary behaviour were higher with higher precipitation and lower with higher temperature and longer day length.

Fig 2. Mean of physical activity measures by different environmental conditions.

Table 3 reports both unadjusted and adjusted associations between physical activity and weather conditions. Significant associations remained with all measures of environmental conditions after adjusting for individual level factors. Daily counts per minute were 26.0 (95% CI: -29.9, -22.0) lower in days with the greatest precipitation (>2.8 mm) compared to dry days. Similar trends in daily counts per minute were observed across the quartile groups for maximum temperature (-29.1; 95% CI: -35.3, -22.9) and day length (-25.9; 95% CI: -34.4, -17.4). On days in the highest precipitation quartile, sedentary time was nearly 15 minutes higher than dry days (14.4; 95% CI: 12.7, 16.2) and time spent in LMVPA time was 8 minutes lower (-8.2; 95% CI: -9.3, -7.1). On the coldest days (maximum temperature <10.0 Celsius), time spent sedentary was nearly 20 minutes higher (19.2; 95% CI: 16.4, 22.0) and LMVPA time by 10 minutes lower (-10.8, 95% CI: -12.6, -9.1) than days with a maximum temperature of over 19 Celsius, although the difference across the quartile groups for minimum temperature was smaller. There was higher sedentary time (20.7; 95% CI: 16.6, 24.8) recorded on the shortest days (<9.3 hours) along with less time spent in LMVPA (-10.0; 95% CI: -12.5, -7.6).

Table 3. The associations between physical activity, day length and weather.

The models in Table 4 include all individual level factors and three measures of environmental conditions: precipitation, maximum temperature and day length together. Since maximum and minimum temperatures were strongly correlated, this fully adjusted model only included maximum temperature which generally had a lager effect size than minimum temperature. The three measures were still significantly associated with physical activity jointly although the effect sizes for maximum temperature and day length were attenuated (by 40~50% in the highest quartile) compared to that observed before joint adjustment.

Table 4. Fully adjusted models showing associations between physical activity and joint environmental conditions.


Main findings

This study investigated the association between day length and weather conditions (precipitation and temperature), physical activity (daily counts per minute and LMVPA time) and sedentary behaviour in older English audits. Short day length and poorer weather conditions, particularly heavy rain and lower temperatures, were associated with up to a 10% reduction in physical activity (25 counts per minute per day or 8 minutes of LMVPA) and a 2% more time sedentary (corresponding to 15 minutes) compared to the average of the whole study population. The associations between day length, weather conditions and physical activity were largely independent of individual level factors and were attenuated but remained after joint adjustment.

Strengths and limitations

Strengths of the study include the fact that it was based on a large population-based cohort of older English adults with objective measures of physical activity for seven days. Objectively measured physical activity can improve limitations of self-reported data and reduce potential recall bias. Unlike previous studies [1315], this study used multilevel time-series modelling to take into account two-level data structure as well as temporal autocorrelation inherent in this type of data.

In terms of limitations, the study population lived in Norfolk, an area situated in East of England. Although daily weather changed throughout the year, the overall climate of this area is mild with less extreme weather conditions compared to some regions in England or other countries. The impact of day length and weather on daily activity could thus be larger in localities with more extreme conditions. Although the literature has suggested that older people are more active in the morning than afternoon or evening [25], this study did not explore the relationship between physical activity and hourly weather conditions as only a single weather station provided data. Although the Marham is only located 50 km from Norwich, the largest city in Norfolk, hourly weather data may not have been representative of the whole study area at any point in time.

This analysis only used a single question to measure health status and did not include complete information from medical records or health examinations. However, self-rated health has been related to mortality and can provide valid insight into individual health in general [23]. A relatively small proportion (15%) of participants reported fair or poor health and this suggests there may have been some selection bias whereby healthier individuals were more likely to remain in the cohort. In common with most other studies, we chose a 7-day wear period for the accelerometer. However, it is possible that patterns of habitual physical activity may not be completely captured over a monitoring period of this length.

Physical activity, sedentary behaviour and weather conditions

The findings of this study show that weather conditions were independently associated with physical activity in older adults. Our results correspond to those of previous studies in other settings variation in climate and other local factors [1315]. We found heavier rain, lower temperatures and shorter day length were associated with lower physical activity. Although we were unable to differentiate time spent indoors and outdoors in this work, we suspect these conditions might strongly affect outdoor activity, which has been shown to substantially contribute to daily activity level in older age [25]. Findings from qualitative research suggest that concerns over safety, fear of falling and injury are potential barriers to outdoor activity in older adults [26,27]. Poor weather conditions may hence increase these worries and lead to reduction in outdoor activity in older adults.

In this older population, much of the time each day was spent sedentary (on average around 11 hours). Sedentary behaviour has been suggested to be related to poor health and act as a risk factor for mortality, cardiovascular diseases and metabolic syndrome [28,29]. A recent meta-analysis evaluating interventions to reduce sedentary time showed a mean reduction of 22 minutes per day among 51 studies [30]. In this study, we observed nearly 15 minutes higher sedentary time on a day with poorest weather compared to best; rather similar to the effect of the reviewed interventions focusing on individual lifestyle and behaviour factors. This suggests that alleviating the negative influence of poor weather may be a possible direction for public health interventions in older adults.

Future research directions and public health implications

To support active ageing and develop possible interventions, future research may explore the mechanism by which weather acts as a determinant of physical activity. For example, since older adults may be more hesitant to leave home in poor conditions due to safety concerns [26,27], a potential intervention could be to improve outdoor environments to be more resilient to poor weather such as adding anti-slip surfaces for pavements or lighting in certain areas. An alternative approach could be to increase individual competence to maintain activity level in days with poor weather. Improving clothing and equipment for wet weather might address some concerns in older adults, whilst enhancing motivation for physical activity could be another direction. An example could be encouraging dog ownership where appropriate, as this has been suggested to help protect against declines in physical activity during periods of poor weather [31].

Supporting information


This work was funded by the Medical Research Council (Grant Number: G0401527; URL: The funder had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. We would like to thank Dr. Flo Harrison for her advice on weather data and statistical modelling.

Author Contributions

  1. Conceptualization: YTW APJ.
  2. Data curation: RL.
  3. Formal analysis: YTW.
  4. Funding acquisition: NW SG APJ.
  5. Investigation: YTW.
  6. Methodology: YTW APJ.
  7. Supervision: APJ.
  8. Writing – original draft: YTW APJ.
  9. Writing – review & editing: YTW RL NW SG APJ.


  1. 1. World Health Organization. Active ageing: a policy framework. Geneva: World Health Organization. 2002.
  2. 2. Van Cauwenberg J, De Bourdeaudhuij I, De Meester F, Van Dyck D, Salmon J, Clarys P, et al. Relationship between the physical environment and physical activity in older adults: A systematic review. Health & Place. 2011;17(2):458–69.
  3. 3. Cunningham GO, Michael YL. Concepts guiding the study of the impact of the built environment on physical activity for older adults: a review of the literature. Am J Health Promot. 2004;18(6):435–43. pmid:15293929
  4. 4. European Commission. Innovation for age-friendly buildings, cities and environments. Brussels: European Commission 2013.
  5. 5. Chan CB, Ryan DA. Assessing the Effects of Weather Conditions on Physical Activity Participation Using Objective Measures. Int J Environ Res Public Health. 2009;6:2639–54. pmid:20054460
  6. 6. Clarke P, Yang T, Keusch F. The impact of weather on mobility and independence in older Americans. Proceedings of the Gerontological Society of America 67th Annual Scientific Meeting. 2014.
  7. 7. Lawton MP & Nahemow L. Ecology and the aging process. In Eisdorfer C., & Lawton M.P. (Eds.), The Psychology of Aging and Adult Development (pp. 619–674). Washington, DC: American Psychological Association. 1973.
  8. 8. Tucker P, Gilliland J. The effect of season and weather on physical activity: A systematic review. Public Health. 2007;121(12):909–22. pmid:17920646
  9. 9. Harrison F, Jones A, Bentham G, van Sluijs E, Cassidy A, Griffin S. The impact of rainfall and school break time policies on physical activity in 9–10 year old British children: a repeated measures study. International Journal of Behavioral Nutrition and Physical Activity. 2011;8(1):47.
  10. 10. Eronen J, von Bonsdorff MB, Törmäkangas T, Rantakokko M, Portegijs E, Viljanen A, et al. Barriers to outdoor physical activity and unmet physical activity need in older adults. Preventive Medicine. 2014;67(0):106–11.
  11. 11. Marquez DX, Aguinaga S, Campa J, Pinsker EC, Bustamante EE, Hernandez R. A Qualitative Exploration of Factors Associated With Walking and Physical Activity in Community-Dwelling Older Latino Adults. Journal of Applied Gerontology. 2014.
  12. 12. Rosenberg DE, Huang DL, Simonovich SD, Belza B. Outdoor built environment barriers and facilitators to activity among midlife and older adults with mobility disabilities. Gerontologist. 2013;53(2):268–79. pmid:23010096
  13. 13. Witham MD, Donnan PT, Vadiveloo T, Sniehotta FF, Crombie IK, Feng Z, et al. Association of day length and weather conditions with physical activity levels in older community dwelling people. PLoS One. 2014;9(1):e85331. pmid:24497925
  14. 14. Klenk J, Büchele G, Rapp K, Franke S, Peter R, Group tAS. Walking on sunshine: effect of weather conditions on physical activity in older people. Journal of Epidemiology and Community Health. 2012;66(5):474–6. pmid:21325149
  15. 15. Prins RG, van Lenthe FJ. The hour-to-hour influence of weather conditions on walking and cycling among Dutch older adults. Age and Ageing. 2015;44(5):886–90. pmid:26259948
  16. 16. Chatfield C. The Analysis of Time Series: An Introduction. CRC Press, 2013
  17. 17. Dunlop DD, Song J, Arnston EK, Semanik PA, Lee J, Chang RW, Hootman JM. Sedentary time in US older adults associated with disability in activities of daily living independent of physical activity. J Phys Act Health. 2015;12(1):93–101. pmid:24510000
  18. 18. Mankowskia RT, Aubertin-Leheudrea M, Beaversd DP, Botoseneanue A, Buforda TW, Churchf T, et al. Sedentary time is associated with the metabolic syndrome in older adults with mobility limitations—The LIFE Study. Experimental Gerontology. 2015;70:32–36 pmid:26130060
  19. 19. Hayat SA, Luben R, Keevil VL, Moore S, Dalzell N, Bhaniani A, et al. Cohort Profile: A prospective cohort study of objective physical and cognitive capability and visual health in an ageing population of men and women in Norfolk (EPIC-Norfolk 3). International Journal of Epidemiology. 2014;43(4):1063–72. pmid:23771720
  20. 20. Day N, Oakes S, Luben R et al. EPIC-Norfolk: study design and characteristics of the cohort. European Prospective Investigation of Cancer. Br J Cancer 1999;80(Suppl 1):95–103.
  21. 21. Time and Date AS. Available:
  22. 22. Bowling A. Just one question: if one question works, why ask several? Journal of Epidemiology and Community Health 2005;59:342–345 pmid:15831678
  23. 23. DeSalvo KB, Bloser N, Reynolds K, He J, Muntner P. Mortality prediction with a single general self-rated health question: a meta-analysis. Journal of General Internal Medicine 2005;21(3): 267–275 pmid:16336622
  24. 24. Goldstein H, Healy M, Rasbash J. Multilevel time series models with applications to repeated measures data. Statistics in Medicine. 1994;I3:1643–1655
  25. 25. Davis MG, Fox KR, Hillsdon M, Sharp DJ, Coulson JC, Thompson JL. Objectively measured physical activity in a diverse sample of older urban UK adults. Med Sci Sports Exerc. 2011;43(4):647–54. pmid:20689449
  26. 26. Lees FD, Clark PG, Nigg CR, Newman P. Barriers to exercise behavior among older adults: a focus-group study. Journal of Aging and Physical Activity. 2005;13:23–33 pmid:15677833
  27. 27. Sugiyama T & Ward Thompson C. Outdoor Environments, Activity and the Well-Being of Older People: Conceptualising Environmental Support. Environ Plan A. 2007;39(8): 1943–1960
  28. 28. Katzmarzyk PT. Physical Activity, Sedentary Behavior, and Health: Paradigm Paralysis or Paradigm Shift? Diabetes. 2010; 59(11): 2717–2725. pmid:20980470
  29. 29. Biswas A, Oh PI, Faulkner GE, Bajaj RR, Silver MA, Mitchell MS, Alter DA. Sedentary time and its association with risk for disease incidence, mortality, and hospitalization in adults: a systematic review and meta-analysis. Ann Intern Med. 2015 Jan 20;162(2):123–32. pmid:25599350
  30. 30. Martin A, Fitzsimons C, Jepson R, Saunders DH, van der Ploeg HP, Teixeira PJ, Gray CM, Mutrie N on behalf of the EuroFIT consortium. Interventions with potential to reduce sedentary time in adults: systematic review and meta-analysis. Br J Sports Med
  31. 31. Lail P, McCormack GR, Rock M. Does dog-ownership influence seasonal patterns of neighbourhood-based walking among adults? A longitudinal study. BMC Public Health 2011;11:148 pmid:21375754