Burden of Disease Attributable to Multiple Risk Factors
This section presents the disease burden attributable to the joint hazards of the risk factors in table 4.1.
All Selected Risk Factors
Table 4.3 shows the joint contributions of all the risk factors shown in table 4.1 to the total mortality and disease burden in different regions. Globally, an estimated 45 percent of mortality and 36 percent of the disease burden were attributable to the joint effects of the 19 selected risk factors. Sub-Saharan Africa (49 percent of the disease burden) and Europe and Central Asia (46 percent of the disease burden) had the largest regional PAFs, and the Middle East and North Africa (25 percent of the disease burden) and East Asia and the Pacific (27 percent of the disease burden) had the smallest. The regions with large joint PAFs are those where a relatively small number of diseases and their risk factors are responsible for large losses of life, for example, HIV/AIDS and childhood disease risk factors in Sub-Saharan Africa and cardiovascular risks, smoking, and alcohol consumption in Europe and Central Asia. Those with smaller joint PAFs are regions where the causes of health loss are distributed among a larger number of diseases and their risk factors.
[Table .]
Table 4.4 shows the individual and joint contributions of the selected risk factors to the 10 leading diseases in the world and in low- and middle-income and high-income countries. As the table shows, for most diseases the joint effects of these risk factors were substantially less than the crude sum of their individual effects. For example, globally four separate risk factors were each responsible for 88, 50, 20, and 11 percent of the diarrheal disease burden, but with a joint PAF of 92 percent; or seven separate risk factors were each responsible for 45, 46, 18, 28, 21, 17, and 17 percent of ischemic heart disease, but with a joint PAF of 80 percent. This confirms that the joint actions of more than one of these risk factors acting simultaneously or through other factors cause a large proportion of disease.
Globally, large fractions of the burden of HIV/AIDS (96 percent), diarrhea (92 percent), ischemic heart disease (80 percent), lung cancer (74 percent), stroke (65 percent), chronic obstructive pulmonary disease (64 percent), and lower respiratory infections (53 percent) were attributable to the joint effects of the 19 risk factors considered here. The joint PAFs for a number of other important diseases and injuries, such as perinatal and maternal conditions, certain other cancers, and intentional and unintentional injuries, which have more diverse risk factors, were smaller but nonnegligible. Even though the fraction of the total malaria burden attributable to childhood undernutrition was relatively large (59 percent), this was because of the contribution of mortality at younger ages to the malaria burden. No adult malaria was attributed to the risk factors in table 4.1, because the epidemiological literature has focused on quantifying increased risk of malaria as a result of childhood undernutrition only. Finally, with the exception of alcohol and drug dependence, which were fully attributable to their namesake risk factors, small or zero fractions of neuropsychiatric conditions, tuberculosis, congenital anomalies, and a number of other diseases were attributed to the risk factors considered here.
[Table .]
An important finding of this analysis is the key role of nutrition in health worldwide. Approximately 11 percent of the global disease burden was attributable to the joint effects of underweight or micronutrient deficiencies. In addition, almost 16 percent of the burden (28 percent for those aged 30 years and older) can be attributed to risk factors that have substantial dietary determinants (high blood pressure, high cholesterol, overweight and obesity, and low fruit and vegetable intake) and to physical inactivity. These patterns are not uniform within regions, however, and the transition has been healthier in some countries than in others (Lee, Popkin, and Kim 2000; Popkin 2002a; Popkin 2002b; Popkin and others 2001). Furthermore, the major nutritional and related risk factors show interregional heterogeneity, for instance, the relative contributions of blood pressure, cholesterol, and BMI differed across regions.
At the same time, the joint contributions of these risk factors left an important part of the global disease burden unexplained, because only a small fraction of some important diseases was attributable to the risk factors considered here. These include diseases whose determinants (a) are diffuse among environmental and behavioral factors, for example, some cancers, perinatal conditions, and neuropsychiatric diseases; (b) have more complex, multifactor etiology and often heterogeneous determinants in different populations, and are therefore difficult to quantify without data on a small scale, such as tuberculosis and injuries; (c) involve long delays between risk factor exposure and disease outcome; or (d) have limited quantitative research at the population level, for instance, neuropsychiatric diseases, often as a result of the previous three factors as well as difficulties in measuring exposure or outcome (Evans 1976, 1978). The mitigation of many such conditions, including malaria, tuberculosis, and injuries, may be better guided by analyses of the effects of interventions tailored to individual settings than by risk factor analysis.
Risk Factor Clusters
In addition to estimating the joint contributions of all the risk factors in table 4.1 to the all-cause mortality and disease burden, we also examined the role of selected clusters of risks that may be of particular interest to disease prevention policies and programs. The risk factor clusters were those affecting cancers (alcohol use, smoking, low fruit and vegetable intake, indoor smoke from household use of solid fuels, urban air pollution, overweight and obesity, physical inactivity, contaminated injections in health care settings, and unsafe sex), cardiovascular diseases (high blood pressure, high cholesterol, smoking, overweight and obesity, alcohol use, physical inactivity, low fruit and vegetable intake, and urban air pollution), and child mortality (childhood underweight; vitamin A deficiency; zinc deficiency; iron deficiency anemia; unsafe water, sanitation, and hygiene; and indoor smoke from household use of solid fuels). Tables 4.5 through 4.7 show the individual and joint contributions of these risk factors to mortality and to the burden of disease for specific diseases within each cluster.
[Table .]
[Table .]
Globally, the cancers with the largest mortality fraction attributable to the risk factors in table 4.1 were cervix uteri cancer (100 percent); trachea, bronchus, and lung cancers (74 percent); and esophagus cancer (62 percent), and those with the smallest joint PAFs were colon and rectum cancers (13 percent) and leukemia (9 percent) (table 4.5). The largest number of deaths attributable to the joint effects of the risk factors was from trachea, bronchus, and lung cancer (930,000 deaths) and liver cancer (283,000 deaths), which reflects both the relatively large joint PAF and the total number of deaths from these cancers. Except for cervix uteri cancer, which was by definition fully attributable to the risk factor unsafe sex, joint PAFs were larger in high-income countries than in low- and middle-income countries for all cancer sites, mostly because of the higher contribution of smoking and alcohol use. The joint PAFs for all cancers combined, however, were similar for the two groups of countries (34 percent versus 37 percent for the disease burden), because of the distributions of total mortality from various site-specific cancers.
Almost two-thirds of all cardiovascular deaths were attributable to eight of the selected risk factors that affect these outcomes (table 4.6). The joint effects of these risk factors were much lower than the crude sum of individual effects (64 percent versus 126 percent for the disease burden), pointing to the extensive overlap in their hazards for cardiovascular diseases compared with cancers. The overlap is partly because the hazardous effects of some risks are mediated through others and partly because multiple risk factors act in combination. The joint PAF differed little between low- and middle-income and high-income countries, reflecting the high levels of multiple cardiovascular risk factors in many middle-income nations (Ezzati and others 2005). Coupled with substantially more cardiovascular deaths and a larger disease burden in low- and middle-income countries, these risk factors result in a much larger loss of healthy life in these nations.
[Table .]
Worldwide, approximately half of the mortality among children under five years of age (about 5 million deaths) was attributable to six major risk factors, with childhood underweight alone accounting for more than a quarter of all child deaths. Practically all the mortality and disease burden from childhood diseases attributable to major risk factors occurred in low- and middle-income countries (table 4.7). The reasons for this large disparity in the disease burden attributable to risk factors are higher risk factor exposure coupled with lower access to case management, which affects child mortality together with risk factor exposure.
