Progress in physical activity over the Olympic quadrennium

Summary

On the eve of the 2012 summer Olympic Games, the first Lancet Series on physical activity established that physical inactivity was a global pandemic, and global public health action was urgently needed. The present paper summarises progress on the topics covered in the first Series. In the past 4 years, more countries have been monitoring the prevalence of physical inactivity, although evidence of any improvements in prevalence is still scarce. According to emerging evidence on brain health, physical inactivity accounts for about 3·8% of cases of dementia worldwide. An increase in research on the correlates of physical activity in low-income and middle-income countries (LMICs) is providing a better evidence base for development of context-relevant interventions. A finding specific to LMICs was that physical inactivity was higher in urban (vs rural) residents, which is a cause for concern because of the global trends toward urbanisation. A small but increasing number of intervention studies from LMICs provide initial evidence that community-based interventions can be effective. Although about 80% of countries reported having national physical activity policies or plans, such policies were operational in only about 56% of countries. There are important barriers to policy implementation that must be overcome before progress in increasing physical activity can be expected. Despite signs of progress, efforts to improve physical activity surveillance, research, capacity for intervention, and policy implementation are needed, especially among LMICs.

Introduction

Every 4 years, the summer Olympic Games divert much of the world's attention from the conflicts and tragedy that regularly dominate the news. The sight of talented athletes pushing their bodies to the limits inspires some viewers to greater achievements in sport and life. Health professionals hope that 2 weeks of exposure to images and stories of athletics will lead viewers to make increased efforts to be physically active in their own lives, even if at a much lower level than the athletes. Although no evidence has shown that the Olympics impact physical activity in the host country or elsewhere,¹ the Olympic Games aim a powerful media spotlight on human movement.

As the London Olympic Games were poised to open in July, 2012, the first Lancet Series on physical activity identified physical inactivity as a global pandemic and urgent public health priority. A wide variety of interventions have been shown to be effective, but they have not been widely implemented, so public health agencies were called upon to collaborate with sectors such as transportation, health care, and sport to mount a stronger response to this health challenge.² The 2012 Series was widely covered in media worldwide, and the Series papers have been heavily cited. With the imminent inauguration of the 2016 summer Olympic Games in Rio de Janeiro, we ask how much progress has been made during the Olympic quadrennium in research, practice, and policy regarding physical activity.

This first paper in this second Lancet physical activity Series provides a progress report on the topics covered in the 2012 Series. Different approaches to identifying progress were taken that were deemed appropriate to each topic. The progress reports on physical activity surveillance and national policies to promote physical activity have strong continuity with papers in the first physical activity Series. Rather than provide an update on deaths from physical inactivity-related non-communicable diseases (NCDs), the present section on health effects summarises new evidence on the link between physical activity and dementia. To complement the papers in the first Series, the sections on correlates of physical activity and intervention studies focus specifically on progress in low-income and middle-income countries (LMICs). Authors of each section used different methods because of the diverse nature of the topics.

We used comparable country estimates for physical inactivity from WHO to analyse the evolution of physical activity surveillance over the Olympic quadrennium (panel 1). In 2012, we obtained adult physical inactivity surveillance data from 122 countries representing 88·9% of the world's population.¹¹ For the present analyses, data were available for 146 countries, representing 93·3% of the world's population (figure 1). The increased global population coverage was mainly due to the addition of populous nations such as Nigeria, Egypt, and Tanzania. Data were available from 82% (40 of 49) of high-income countries (HICs), 75% (41 of 55) of upper-middle-income countries (U-MICs), 69% (38 of 55) of lower-middle-income countries (L-MICs), and 77% (27 of 35) of low-income countries (LICs). The proportion of countries contributing surveillance data among adult populations increased in all regions, except southeast Asia: Africa (72–87%), Americas (43–57%), eastern Mediterranean (43–57%), Europe (68–75%), southeast Asia (82%, no change), and western Pacific (70–89%).

Key messages

• In the 4 years since the 2012 Lancet Series that identified physical inactivity as a global pandemic, progress has been made in the breadth of national surveillance, evidence about physical activity as a protective factor for dementia, adoption of national policies and action plans, and research on correlates and interventions in low-income and middle-income countries. However, progress in the implementation of national actions to address one of the biggest health challenges of the 21st century has been insufficient.

• Most countries have done population surveys of physical activity, with an extra 24 countries providing adult data and 15 countries providing adolescent data since 2012. The global prevalence of physical inactivity was about 23% for adults and about 80% for school-going adolescents, although self-report data have limitations. Few countries provided trend data for adults, and trend data for adolescents showed an increase in proportion of people who were physically inactive in most countries.

• In addition to the major impact of physical inactivity on the global burden of non-communicable diseases documented 4 years ago, evidence now shows that almost 300 000 cases of dementia could be avoided annually if all people were adequately active, and this figure is increasing as the global population ages.

• Research examining reasons why people are and are not physically active has increased substantially in middle-income countries, but not in low-income countries. Unlike evidence from high-income countries, urban (vs rural) residence emerged as an inverse correlate of physical activity in low-income and middle-income countries (LMICs), which is a concern given global trends toward urbanisation. These results can be used to design interventions informed by local data.

• Research and evaluation of physical activity interventions has increased in LMICs. Although several examples of effective interventions have been reported, the evidence is still scarce. An important next step is to build capacity for intervention research in LMICs so interventions can be developed or adapted for local conditions, then rigorously assessed.

• Due largely to the inclusion of physical activity in the WHO Global Action Plan on NCDs and the establishment of a global target to reduce inactivity by 10% by 2025, many countries have now adopted national policies or action plans to increase physical activity. However, implementation appears to be weak. Meaningful action will require increasing the infrastructure and resources for physical activity, including providing capacity-building, country technical assistance, creating effective multisector coalitions, and reaching consensus on a few highest-priority actions for each country.

• Overall, physical activity surveillance, research, and policy adoption worldwide improved. However, policy implementation appears to be poor, and evidence of an increasing trend in global physical activity was absent. Thus, the global pandemic of physical inactivity remains, and the capacity for nations to respond is improving too slowly.

Notably, the algorithm used to estimate physical inactivity among adult populations has changed from that presented in the 2012 Lancet Series¹¹ to align with the new standards used by the WHO Global Health Observatory.⁴ In 2012, inactivity was defined as not achieving 5 days of 30 min of moderate-intensity activity, or 3 days of 20 min of vigorous-intensity activity, per week, or an equivalent combination, according to the recommendations at that time.⁵ Reflecting scientific evidence¹² and following updated physical activity recommendations,³ inactivity was defined for the present analyses as not achieving 150 min of moderate-intensity activity or 75 min of vigorous-intensity activity per week, or an equivalent combination, regardless of the weekly frequency. This recommendation is easier to achieve. Thus, the estimated prevalence of inactivity among adult populations worldwide changed from 31·1% in 2012 to 23·3% in 2016, a reduction that primarily reflects changes in the recommendations rather than a real increase in physical activity. The lack of substantial change is confirmed by findings from the 12 countries with trend data that included domains of leisure, transportation, and occupation. Six countries (Argentina, Belgium, Iran, Kuwait, Mongolia, and Singapore) reported a numerical increase in the prevalence of inactivity, and six countries (Maldives, New Zealand, South Korea, Seychelles, South Africa, and USA) reported a decrease (for references for trends see appendix p 1). Notable disparities remain in the prevalence of physical inactivity between men and women, with 137 of the 146 countries showing higher inactivity among women.4, 13 Older age groups continue to be at higher risk for inactivity, with the oldest age category showing more than double the prevalence of the youngest (aged 80 years or older, 55·3% vs aged 18–29 years, 19·4%).

Improvements in global surveillance coverage of physical activity were also documented for school-going adolescents. In the 2012 publication¹¹ we analysed data of adolescents aged 13–15 years from 105 countries. For the present analyses, estimates were available for adolescents aged 11–17 years from 120 countries⁴ (figure 1), with data mainly from the Global School-based Student Health Survey¹⁴ and the Health Behaviour in School-aged Children Study.¹⁵ The population coverage of adolescent surveillance increased from 68·0% in 2012 to 76·3% in 2016. Availability of self-report data for adolescents was 81·6% in HICs (40 of 49), 70·9% in U-MICs (39 of 55), 60·0% in L-MICs (33 of 55), and 20·0% in LICs (seven of 35). The proportion of countries contributing surveillance data from adolescents increased in all world regions, except Africa and southeast Asia: Africa (30%, no change), Americas (57–77%), eastern Mediterranean (57–76%), Europe (64–68%), southeast Asia (55%, no change), and western Pacific (33–78%). We identified 50 countries that reported comparable trend data for adolescents. For 32 of the 50 countries, the prevalence of inactivity numerically increased, whereas for the other 18, prevalence of inactivity decreased.

Consistent with the 2012 Series, adolescent inactivity prevalence was defined as not achieving at least 60 min of moderate to vigorous physical activity daily.3, 11 Inactivity prevalence continued to be extremely high, with a global average of 78·4% for boys and 84·4% for girls. In the vast majority of countries (115 of 120 countries with data), more than a quarter of school-going adolescents did not reach the recommended level of activity.4, 13 The apparent higher inactivity prevalence for adolescents than adults was partly a result of the higher recommended level for youth. However, prevalence cannot be compared directly across age groups because the questionnaires differed greatly.

Given known limitations of self-reports, the use of objective physical activity measures, such as accelerometers, to estimate national prevalence is growing. A 2015 review¹⁶ of accelerometer studies in adults found 76 studies across 36 countries that had used devices in at least 400 participants, with 13 identified as national population-based cohorts. From this review, eight studies from seven HICs met our definition of reporting national prevalence.17, 18, 19, 20, 21, 22, 23, 24, 25 Prevalence estimates varied from 1% to 52% for meeting physical activity recommendations. However, estimates were not comparable across countries as a result of large variations in data collection methods, data processing, and scoring. Experts agree that standardised accelerometer methods are needed,¹⁶ and prevalence estimates from accelerometers should not be compared with self-report data.²⁵

In children and adolescents (aged 5–19 years) we found accelerometer-based population prevalence estimates of engaging in 60 min or more of physical activity daily in six studies from five HICs.17, 21, 23, 26, 27, 28 Once again, prevalence estimates were not comparable and reflected methodological inconsistencies. The International Children's Accelerometry Database (ICAD) had accelerometry data from 20 studies worldwide, and allowed comparisons because of standardised methods,²⁹ but most samples were not nationally representative. ICAD data also showed large between-country physical activity prevalence variations ranging between 15% and 28%.²⁹

More countries are collecting physical activity surveillance data, although reporting on adolescents in LICs has not improved much. About a quarter of adults and 80% of adolescents were not meeting guidelines according to self-report data. Though trend data were scarce, no evidence has shown that physical inactivity declined globally. More countries are using objective measures for surveillance, demonstrating feasibility. To promote wide use of objective measures for surveillance, methods should be standardised, and data collection in LICs should be supported.