The Beijing Federation of Trade Unions has hired 5, instructors to teach employees how to maximize the eight-minute exercise routine. Higher incomes have meant more people are eating rich, high-caloric diets and following sedentary lifestyles, a scenario that has also translated into a thriving weight loss industry. The weight loss center that hosted Tian for free, for example, has expanded to 1, locations across the country since it opened in Luxury is idleness. All Sections.
About Us. B2B Publishing. Business Visionaries. Even the ancient Greek physician Hippocrates, commonly referred to as the father of medicine , commented on the health risks of obesity during his lifetime, writing: "The men lack sexual desire because of the moistness of their constitution and the softness and coldness of their bellies. In the case of the women, fatness and flabbiness are also to blame Everything from our attitudes toward the large to rumored treatments for obesity have dramatically changed over time:.
Adult obesity rates continue to rise in many states and fell in exactly none in No state has a prevalence of adult obesity that is less than 20 percent , and two -- Mississippi and West Virginia -- have a prevalence of adult obesity of 35 percent or more.
For the most part, childhood obesity rates in the U. Experts currently recommend losing weight slowly and steadily, on average about one to two pounds a week, by a combination of healthy eating and regular physical activity. And every little bit counts: Even among overweight or obese people, losing just 5 to 10 percent of a person's total body weight has been linked with health benefits.
Still, others argue that the overweight and obese are an example of a very limited number of people the general public still finds it acceptable to shame. News U. Year fixed effects control for common time trends shared among all countries, in effect allowing us to estimate our relationship of interest after removing global trends in overweight.
We estimate logistic models for the binary outcome variable of individual overweight and include findings with obesity and BMI as the dependent variables in Tables G—L in S1 Appendix. We also estimate the same model with country-specific age and sex trends in Fig E and wealth quintiles in Fig F in S1 Appendix. After estimating the basic shift in overweight and obesity trends between the poor and wealthy, we projected this shift to We project the overweight and obesity prevalence to of each sex and 5-year age groups, from age 15 to 49, using Eq 1 with a linear time trend.
We use GDP per capita projections from the Institute for Health Metrics and Evaluation to estimate the effect of economic development to [ 19 ]. We incorporated variance in the GDP per capita series and parameter uncertainty from our regression model to quantify an uncertainty interval for our projections.
After estimation of Eq 1 , we took 1, draws from the multivariate normal distribution defined by the model parameter estimates and the variance-covariance matrix of the model. To create our predictions, those were coupled with 1, draws provided by the Institute for Health Metrics and Evaluation for their GDP per capita series. These draws incorporate model, data, and parameter uncertainty by using ensemble modeling techniques, drawing from the sub-model variance-covariance matrices, and adding a random walk of statistical noise to each forecast with variance based on the residual of the observed data for each sub-model.
We weight the predicted age—sex-specific overweight and obesity rates by the United Nations World Population Prospects age—sex-specific population projections to aggregate overweight and obesity rates to the national level by personal wealth decile [ 20 ]. Thus, these projections also capture the effects of aging on overweight and obesity rates. In Table U in S1 Appendix , we report results for out-of-sample validation of these projections.
We collected data from nationally representative surveys with individual-level data for 2. Fig 1 stratifies the overweight 1A and obesity 1B prevalence by level of economic development and ranked quintile of personal wealth. While Fig 2 adjusted for confounders, the shifting pattern was seen from the raw data. Unadjusted overweight 1A and obesity 1B prevalence obtained directly from survey data, stratified by GDP per capita and within-survey personal wealth decile.
The columns represent GDP per capita categories, and the rows represent deciles of within-country wealth. Within each GDP per capita category, deciles with the lowest prevalence are coded in green, and deciles with the highest prevalence are coded in red. All prevalence estimates were obtained using survey weights. GDP, gross domestic product. Each point represents the probability of being overweight 2A or obese 2B relative to the richest decile 90th—th percentile at different GDP per capita cutoffs.
The lines are color coded by wealth decile. The wealth-overweight and wealth-obesity transition zones are denoted by the vertical lines. The first line marks where the richest decile was no longer the most likely to be overweight or obese. The second line marks where the richest decile was less likely than the poorest to be overweight or obese. We evaluate the change in the gradient based on where the other percentiles are statistically significantly greater than zero which means the wealth group has a higher chance of obesity than the 90th—th percentile personal wealth income group.
Fig 2 reports the adjusted probability of being overweight 2A and obese 2B in each individual-level wealth decile relative to the richest decile with increasing GDP per capita. At the GDP per capita of low-income countries, there was an increasing probability of overweight and obesity with personal wealth relative to individuals in the poorest decile. Fig 3 displays the projected relative change in the share of overweight by wealth decile between and , grouped by World Bank income group.
In low-income countries, the richest decile had a The changes in the relative share of overweight were greatest in low- and lower-middle-income countries. In lower-middle-income countries, the share of overweight in the poorest decile was projected to increase In high-income countries, overweight in the richest decile was expected to decrease 6.
Each line displays World Bank Income group changes in overweight burden across a wealth decile. Wealth deciles, where 1 is the poorest and 10 is the richest, are displayed on the x-axis.
The percent change in each wealth decile's overweight burden share from to is on the y-axis. The maps in Fig 4 display the projected change in overweight prevalence, portion of overweight, and overweight population in the bottom quintile of personal wealth in each of our study countries to Fig 4B suggests that from to , the largest increase in the share of the overweight that is relatively poor was in Ethiopia Finally, as shown in Fig 4C , the largest population growth of overweight and relatively poor individuals was projected to occur in Niger from , to 1.
We projected that the number of individuals who are poor and overweight in India will grow from Each map displays country-level projections in overweight prevalence inequality. Map A shows the percent change in overweight prevalence among the relatively poor, defined as individuals in the bottom quintile of the personal wealth distribution. Map B displays the percent change in the share of overweight individuals who are relatively poor.
This differs from Map A by quantifying where the reversal of the wealth-overweight gradient will occur the fastest. Map C displays the percent change in the overweight and relatively poor population. As countries develop economically, we observed that, along with rising overweight rates, a larger fraction of the global overweight and obese populations become relatively poor. This is important for policy makers, as public health systems may be asked to shoulder the health and social consequences of overweight and obesity among those unable to cover the costs themselves.
The relatively poor in many countries have limited access to healthcare and worse health outcomes, and this transition is one of both growing average burden and increasing health inequities.
The intuitive implications are that cardiometabolic diseases and related conditions associated with overweight and obesity could be shifting to the relatively poor, and this has been occasionally documented [ 21 , 22 ]. Moreover, we found that the prevalence of overweight and obesity in the wealthiest decile had changed relatively little with increasing GDP per capita during the last 20 years.
This arguably unexpected finding would be consistent with a pattern in which individuals in wealthier strata are less affected by the mechanisms influencing the rise in overweight and obesity among the poor as countries develop economically. One hypothesis for future work is that the wealthy may not change their food consumption as much as the poor in response to food prices, or their physical activity levels may not change much with labor choices as economic development occurs.
It is also possible that the poor face changing supply-side factors for example, food deserts or food swamps as economic development occurs [ 23 ], while the rich do not. Our projections, while speculative, suggest that if current trends and relationships continue, the rise in the number of relatively poor individuals who are also overweight will be most pronounced in low- and lower-middle-income countries over the next decade.
Countries with currently very low GDP per capita, particularly those in sub-Saharan Africa, are not projected to realize a full wealth-overweight transition by , yet there will still be substantial growth in overweight burden. In addition, lower-income countries are projected to experience population growth not expected in higher-income countries [ 20 ].
As life expectancy continues to rise, especially in lower-income countries, one implication of this study is that the growing burden of obesity among the poor will translate to increased reliance on public health systems for cardiovascular disease, type 2 diabetes, and related chronic conditions [ 9 ]. These projections highlight that policy makers, facing a growing burden of overweight-related diseases among the relatively poor, will need to design and implement interventions for a different target population than in the past.
Specifically, as economic development and globalization progress in low-income countries, overweight-related care may be more widely demanded by the relatively poor. If governments want to tackle the burden of overweight and associated conditions, they must thus target their interventions at the poorest, which may require different approaches and financing than if they were targeting the wealthy.
By planning for this reversal of the wealth gradient and subsequently different interventions, policy makers will use public health resources in a cost-effective manner, in addition to reducing health disparities. Previous evidence suggests that sources of calorie intake differ across the relatively rich and poor, with the poorest having lower-quality diets than the richest [ 24 ].
While there are many possible explanations for this, if within a particular country, most of the health and economic burden of overweight and obesity is concentrated in the poor, then national policy makers would focus on the sources of calorie intake that the poor are most likely to consume.
The same would be true in a different country where the burden is concentrated in the relatively rich population. There are multiple policy instruments that might be effective: tackling food deserts in low-income neighborhoods [ 23 ]; subsidizing fruit and vegetables [ 25 ]; taxing sugar with a word of caution that this could be seen as a regressive tax with potentially positive health effects but negative effects on income inequality [ 26 ]; working with industry to reformulate their products where the industry and products on which a country would focus are distinct if overweight is concentrated in the rich or the poor [ 27 ]; even potentially ensuring safe places for people in low-income neighborhoods to exercise and be outside [ 28 ].
From a public health perspective, anticipating the disease burden is of the utmost importance. This is true universally, and especially in resource scarce environments. This research and the accompanying wealth-overweight transition highlight when overweight and obesity burdens are expected to grow, and therefore equips domestic public health programs with key information needed regarding when to act and what to prioritize.
This same information is useful to health donors concentrated on efficiently disbursing resources for the largest health gain possible, in the moment and when planning for the future. This study has several limitations. First, although the World Health Surveys rely mostly on in-person surveys, which may reduce bias, its anthropometric measures are self-reported.
Sensitivity analyses in Fig C and Tables H, J, and L in S1 Appendix suggest that this does not represent a significant source of bias, but ideally the outcome variable would be recorded anthropometrically in all instances. By including mostly self-reported data for high-income countries, we might observe lower BMI values in these countries, but comparisons of self-reported and national estimates suggest the bias averages to zero, and an indicator variable for self-reported data suggests it is only lower by 1.
Moreover, our results remain similar using only the Demographic and Health Surveys. Second, all surveys in this analysis are cross-sectional, so we cannot follow individuals over time to observe their changing personal wealth and anthropometric measurements. Third, there is a complex relationship between overweight, obesity, and personal wealth.
Obesity may limit one's ability to attain wealth [ 29 ], or it may be itself an indication of wealth [ 30 , 31 ], depending on context. Fourth, expected overweight and obesity projections are based on a relatively simple model and highlight how expected changes in wealth may translate to overweight and obesity rates, given past trends and relationships continuing in the future.
Some important determinants of overweight and obesity are not observed in our data. Reduced physical activity—as a result of changes in occupation, urbanization, transportation, or obesity status—may be responsible for weight gain among many populations.
Similarly, we do not observe food prices and food availability at an individual level.
0コメント