The Lancet Regional Health - Americas 4 (2021) 10 0 068 Contents lists available at ScienceDirect The Lancet Regional Health - Americas journal homepage: www.elsevier.com/locate/lana Impact of common cardio-metabolic risk factors on fatal and non-fatal cardiovascular disease in Latin America and the Caribbean: an individual-level pooled analysis of 31 cohort studies Cohorts Consortium of Latin America and the Caribbean (CC-LAC) ∗, 1 a r t i c l e i n f o Article history: Received 9 April 2021 Revised 14 July 2021 Accepted 24 August 2021 Available online 17 September 2021 a b s t r a c t Background: Estimates of the burden of cardio-metabolic risk factors in Latin America and the Caribbean (LAC) rely on relative risks (RRs) from non-LAC countries. Whether these RRs apply to LAC remains un- known. Methods: We pooled LAC cohorts. We estimated RRs per unit of exposure to body mass index (BMI), systolic blood pressure (SBP), fasting plasma glucose (FPG), total cholesterol (TC) and non-HDL cholesterol on fatal (31 cohorts, n = 168,287) and non-fatal (13 cohorts, n = 27,554) cardiovascular diseases, adjusting for regression dilution bias. We used these RRs and national data on mean risk factor levels to estimate the number of cardiovascular deaths attributable to non-optimal levels of each risk factor. Results: Our RRs for SBP, FPG and TC were like those observed in cohorts conducted in high-income countries; however, for BMI, our RRs were consistently smaller in people below 75 years of age. Across risk factors, we observed smaller RRs among older ages. Non-optimal SBP was responsible for the largest number of attributable cardiovascular deaths ranging from 38 per 10 0,0 0 0 women and 54 men in Peru, to 261 (Dominica, women) and 282 (Guyana, men). For non-HDL cholesterol, the lowest attributable rate was for women in Peru (21) and men in Guatemala (25), and the largest in men (158) and women (142) from Guyana. Interpretation: RRs for BMI from studies conducted in high-income countries may overestimate disease burden metrics in LAC; conversely, RRs for SBP, FPG and TC from LAC cohorts are similar to those esti- mated from cohorts in high-income countries. Funding: Wellcome Trust (214185/Z/18/Z) © 2021 The Author(s). Published by Elsevier Ltd. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ) H R i h 2 Research In Context Evidence before this study The search query ("Latin America" AND "Caribbean") AND ("relative risks" OR "population attributable fraction OR "PAF") AND ("body mass index" OR "BMI" OR "blood pressure" OR "total cholesterol" OR "fasting glucose") did not retrieve any results in PubMed (June 14 th 2021). It is well known that Latin America and the Caribbean has not had large multi-country cohort studies or cohort pooling ∗ Corresponding Author: Dr. Goodarz Danaei, Harvard T.H. Chan School of Public ealth, Department of Epidemiology, 677 Huntington Avenue, Building 1, 11th Floor, oom 1107, Boston, MA 02115, United States, Phone: + 16174325722 E-mail address: gdanaei@hsph.harvard.edu (Cohorts Consortium of Latin Amer- ca and the Caribbean (CC-LAC)) 1 Authors listed at end. ttps://doi.org/10.1016/j.lana.2021.10 0 068 667-193X/© 2021 The Author(s). Published by Elsevier Ltd. This is an open access article projects. Before this work, the evidence about long-terms ef- fects of cardio-metabolic risk factors in Latin America and the Caribbean was informed by cohorts conducted in North America, Europe and Asia. Added value of this study This work pooled data from several Latin American and the Caribbean cohorts and examined the relative risks of es- tablished cardio-metabolic risk factors for cardiovascular out- comes. We found that the relative risks for systolic blood pressure, fasting glucose and total cholesterol, are similar to those reported by cohort pooling projects carried out in other world regions (e.g., Asia-Pacific Cohort Studies Collab- oration, Prospective Studies Collaboration and Emerging Risk Factors Collaboration); however, for body mass index, the rel- ative risks were slightly smaller in Latin America and the Caribbean. We used the relative risks herein derived to es- timate the mortality attributable to non-optimal levels of under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ) https://doi.org/10.1016/j.lana.2021.100068 http://www.ScienceDirect.com http://www.elsevier.com/locate/lana http://crossmark.crossref.org/dialog/?doi=10.1016/j.lana.2021.100068&domain=pdf http://creativecommons.org/licenses/by/4.0/ mailto:gdanaei@hsph.harvard.edu https://doi.org/10.1016/j.lana.2021.100068 http://creativecommons.org/licenses/by/4.0/ The Lancet Regional Health - Americas 4 (2021) 10 0 068 1 a k o e v r [ i o t c i a a s w t d f t p e o c t m a 2 C W [ W c p ( o s fi p 3 h n 3 3 c s C g r p c e e t w w w ( f f c t 2 c c s w c e 2 m ( h v c a [ 2 2 n r p p I o r F i s 2 f m t the selected cardio-metabolic risk factors. We estimated the largest attributable cardiovascular deaths due to non-optimal systolic blood pressure and non-HDL cholesterol. These risk factors had a larger impact on cardiovascular deaths in the Caribbean, as well as in Southern and Tropical sub-regions. Implications for all the available evidence Our results support using global relative risks for systolic blood pressure, fasting glucose and total cholesterol in Latin America and the Caribbean; for body mass index, however, it seems reasonable to use the relative risks herein proposed. Global relative risks for body mass index may overestimate disease burden metrics in Latin America and the Caribbean. . Introduction Cardiovascular diseases are the leading causes of death glob- lly [1] , and the main causes of these deaths are a set of well- nown cardio-metabolic risk factors such as high blood pressure, verweight/obesity, diabetes and dyslipidemias [2-4] . Supporting vidence on the impact of cardio-metabolic risk factors on cardio- ascular diseases has mostly come from cohort pooling collabo- ations [5] , including the Asia-Pacific Cohort Studies Collaboration 6 , 7 ], the Prospective Studies Collaboration [8-10] , and the Emerg- ng Risk Factors Collaboration [11-13] . The results of these collab- rations have been used to attribute disease burden to risk fac- ors globally, providing inputs for surveillance and monitoring of ardio-metabolic risk factors and diseases. These collaborations have little representation from Latin Amer- ca and the Caribbean (LAC) [14] , a vast region with unique char- cteristics in terms of non-communicable diseases such as diabetes nd raised blood pressure [15] , paired with the fastest rate of tran- ition towards a predominance of urban areas in the developing orld [ 16 , 17 ]. Therefore, the findings of these non-LAC collabora- ions, such as the age-specific relative risks used in global bur- en of disease estimations, may not apply to LAC countries. In act, evidence suggests that the association between some risk fac- ors and cardio-metabolic outcomes may be stronger in LAC com- ared with other world regions [ 4 , 18 ], possibly due to different lev- ls of access to health care [ 19 , 20 ], differences in the distribution f cardio-metabolic risk factors [ 15 , 21-23 ], or incidence of non- ommunicable diseases [24-26] . We identified and pooled prospec- ive cohort studies in LAC to examine the effect of major cardio- etabolic risk factors on cardiovascular outcomes, and to estimate ge-specific relative risks for this world region. . Methods Details about the Cohorts Consortium of Latin America and the aribbean (CC-LAC) have been reported detailed elsewhere [27] . e analysed cohort data pooled and harmonized by the CC-LAC 27] , a network of health researchers and practitioners in LAC. e have harmonised and pooled approximately population-based ohort data on cardio-metabolic risk factors and outcomes, i.e., articipants were not recruited based on cardiovascular diseases CVD) (e.g., cohort of stroke survivors) or risk factor (e.g., cohort f smokers) history only. Cohort studies were identified through a ystematic search and networks of researchers in LAC. We identi- ed 78 approximately population-based cohorts (i.e., did not select articipants on the basis of having previous disease) and excluded 1 cohorts that recruited only young participants (e.g., birth co- orts), did not measure exposures/outcomes of interests, or could ot be accessed by the original investigators [27] . We accessed 2 3 cohorts from 13 countries (37% of LAC countries). From these 3 cohorts, 5 included participants who attended a specific health entre [28-30] or were members of a professional organization uch as The Mexican Teachers’ Cohort [31] and the Health Workers ohort Study [32] . The other cohorts sampled individuals from the eneral population. Individual-level data from each cohort were eceived by the CC-LAC and were subsequently harmonised and ooled for the present analyses [27] . From the pooled 33 cohorts, we excluded two that did not as- ertain either fatal or non-fatal cardiovascular events; we further xcluded 18 cohorts that did not ascertain non-fatal cardiovascular vents from analysis of fatal- and non-fatal outcomes. Thus, our es- imates for fatal cardiovascular events were informed by 31 cohorts ith a mean follow-up of 8.8 years (standard deviation = 3.1), hile our estimates for fatal and non-fatal cardiovascular events ere informed by 13 cohorts with a mean follow-up of 8.5 years standard deviation = 5.3). We estimated incidence rate ratios which we will hereafter re- er to as relative risks (RRs) and 95% Confidence Intervals (95% CI) or each selected cardio-metabolic risk factor on fatal and non-fatal ardiovascular diseases. We used these RRs to estimate the propor- ion and number of deaths attributable to each risk factor. .1. Cardiovascular outcomes We analysed two outcomes separately: i) fatal and non-fatal ardiovascular events and ii) fatal cardiovascular events. Non-fatal ardiovascular events were not analysed alone because of the mall number of events in many age groups. Cardiovascular events ere identified using data from vital registration systems, clini- al records or verbal autopsies, and where relevant adjudicated by ach cohort (Supplementary Table 1). .2. Cardio-metabolic risk factors The risk factors of interest were systolic blood pressure (SBP, in mHg), body mass index (BMI, in kg/m 2 ), fasting plasma glucose FPG, in mmol/L), total serum cholesterol (in mmol/L) and non- igh-density lipoprotein cholesterol (non-HDL, in mmol/L). These ariables were collected following standardised protocols in each ohort. We only examined SBP as the relationship between SBP nd CVD outcomes is stronger than that of diastolic blood pressure 8 , 33 ]. .3. Statistical analysis .3.1. Handling of missing data BMI was missing in 12% of the pooled observations, while this umber for SBP, total cholesterol, FPG and non-HDL cholesterol anged from 66% to 80% (Supplementary Table 2). We used multi- le imputation and fitted the regression models in each of 50 im- uted datasets, pooling the estimates following Rubin’s rules [34] . n sensitivity analyses, we used a complete-case dataset, and we bserved minor differences for non-HDL cholesterol but overall the esults were unchanged using multiple imputation (Supplementary igure 1). Detailed methods on multiple imputation are available n Supplementary Material (pp. 7-8). The main results herein pre- ented are based on the multiple imputation data. .3.2. Adjusting for Regression Dilution Bias (RDB) As the selected risk factors have a natural variability during ollow-up, the estimated associations between baseline one-off easures underestimate the effect of “usual” exposure. This is of- en referred to as regression dilution bias [ 35 , 36 ]. To adjust for this The Lancet Regional Health - Americas 4 (2021) 10 0 068 b t m r f t o 2 e ( d a s a r f W l p s d R b P g w p s p 2 t t p b ( w [ c w T i d m W i 2 R e d C t t e w S 2 e h b s r B a s d u o f c p [ 2 l G fi 3 w o I a k c a w h I w S w t 4 m m m n T c ( a d t ( t m v F S d s R d w ( S k ias, we used data from 10 of our cohorts with repeated risk fac- or measurements and used standard analytic methods (MacMahon ethod) [ 35 , 36 ] to calculate correction factors. The estimated cor- ection factors were: 1.10 for BMI, 1.50 for SBP, 1.53 for FPG, 1.75 or total cholesterol, and 1.85 for non-HDL cholesterol. Further de- ails about the RDB methods are available in the expanded meth- ds (Supplementary Material pp. 6-7). .3.3. Survival analysis For each cardio-metabolic risk factor, we fitted a Poisson lin- ar mixed effects regression model for each age group separately 35-44, 45-54, 55-64, 65-74, 75-84, 85 + years) in which the in- ependent variable was the risk factor, adjusted for sex and age t risk (i.e., at follow-up/event) within each age-group. A cohort- pecific random intercept was included as well as the natural log- rithm of the follow-up time as an offset. The coefficient of each isk factor from this model represents the log-incidence rate ratio or one-unit increase in the risk factor in each 10-year age group. e applied the RDB correction factor to these coefficients (in the og scale). For further details, refer to the extended methods (Sup- lementary Material pp. 5-11). To better understand any potential differences in RRs between ub-regions in LAC, we computed RDB-adjusted RRs for fatal car- iovascular diseases for Central America & the Caribbean (Costa ica, Cuba, Dominican Republic, Puerto Rico and Trinidad & To- ago) versus South America (Argentina, Brazil, Chile, Colombia, eru, Uruguay and Venezuela). We included Mexico in the former roup to preserve geographic proximity. To compare our results ith a previous analysis of cohort data pooling studies that re- orted RRs for fatal ischaemic heart disease and stroke subtypes eparately [5] , we weighted their estimated RRs by the relative revalence of these outcomes in LAC. .3.4. Quantifying the population-level impact of risk factors Following a comparative risk assessment approach [37] , we es- imated the population attributable fraction (PAF) for each risk fac- or on cardiovascular deaths in 35 countries of the region com- aring the current mortality burden to the one that would have een observed if the mean levels in the population were optimal Supplementary Material p. 9). The optimal levels in the population ere derived from previous analyses of global burden of disease 5] . Current mean levels of BMI, SBP, total cholesterol and non-HDL holesterol for each country and by five-year age group and sex, ere extracted from the NCD-RisC ( http://ncdrisc.org/ ) [ 15 , 22 , 23 ]. his information was not available for FPG, therefore, we did not nclude non-optimal glucose in these analyses. The number of car- iovascular deaths for the year 2019 was extracted from the esti- ates provided in the Global Burden of Disease (GBD) Study [38] . e used the RRs herein estimated (10-year age groups), and we nterpolated them into 5-year age groups (Supplementary Figure and Supplementary Table 7) [5] . We used the same age-specific Rs for men and women across countries in LAC, as we found no vidence of different RRs by sex. We estimated crude attributable eath rates per 10 0,0 0 0 person-years by multiplying PAFs by total VD deaths and dividing by the adult population of each coun- ry, which was also extracted from the GBD Study. Further de- ails about the comparative risk assessment are available in the xpanded methods (Supplementary Material pp. 8-9). Countries for hich we made estimates are those in common between the GBD tudy and the NCD-RisC (35 countries and territories in LAC). .4. Risk of bias in each study We evaluated three sources of bias. First, selection bias due to nrolment of participants. The risk of selection bias in these co- orts is rather small because inclusion in the study is unlikely to 3 e simultaneously related to exposure and outcome. Second, mea- urement bias. As explained above and in Supplementary Mate- ial p. 04, major variables of interest were measured except for MI which was self-reported in one cohort [ 31 , 39 ]. These variables re commonly measured in cardiovascular cohorts and were mea- ured following standard procedures. Regarding the outcomes, we id not pool cohorts in which this information was not verified sing links to vital registration data or adjudication (Flow Diagram n Supplementary Material p. 04). Third, confounding. We adjusted or age, sex and cohort in all analyses as we were interested in omparing the magnitude of our RRs with those of other global ooling studies which used the same set of potential confounders 5] . .4.1. Role of the funding source The funder of the study had no role in study design, data col- ation, analysis, interpretation, or writing of the report. RMC-L and D had full access to the data in the study. RMC-L and GD had nal responsibility for the decision to submit for publication. . Results In the group of cohorts analysed for fatal outcomes, women ere younger than men (46.1 vs 55.7 years), while in the sec- nd set of cohorts the age was more alike (52.7 vs 52.1 years). n both sets of cohorts (i.e., analysis for fatal as well as for fatal nd non-fatal outcomes), women had higher BMI than men (27.4 g/m 2 vs 26.2 kg/m 2 and 28.6 kg/m 2 vs 26.4 kg/m 2 ); mean total holesterol was also higher in women (5.4 mmol/L vs 5.2 mmol/L nd 5.3 mmol/L vs 5.2 mmol/L). The average non-HDL cholesterol as slightly higher among men than women in both sets of co- orts (4.1 mmol/L vs 4.2 mmol/L and 4.1 mmol/L vs 4.2 mmol/L). n both sets of cohorts, mean SBP was higher among men than omen (131 mmHg vs 134 mmHg and 128 mmHg vs 133 mmHg; upplementary Table 3B). For fatal outcomes, the 31 selected cohort studies contributed ith 168,287 eligible participants aged 20 years old and over. More han four fifths were women (83.7%) and they were on average 7.7 (standard deviation (SD) = 12.2) years old at baseline. The ean BMI was 27.2 kg/m 2 (SD = 4.8), the average SBP was 131 mHg (SD = 22.1), the mean FPG was 5.5 mmol/L (SD = 1.9), the ean total cholesterol was 5.3 mmol/L (SD = 1.3), and the average on-HDL cholesterol was 4.2 mmol/L (SD = 1.3) (Supplementary able 3). The mean follow-up was 8.9 (SD = 3.1) years. In the 31 ohorts analysed for fatal outcomes we observed, 1,710 events (116 95% CI: 111-122) per 10 0,0 0 0 person-years). We observed an age gradient in the magnitude of the RDB- djusted RRs across all cardio-metabolic risk factors for fatal car- iovascular outcomes, with smaller RRs in older ages; this pat- ern was less clear for total cholesterol and non-HDL cholesterol Figure 1 , Supplementary Table 5). The magnitude of the RRs in he youngest group (35-44) was at or above 1.3 ( Figure 1 , Supple- entary Table 5), with the largest estimate for SBP on fatal cardio- ascular events (RR = 1.9, 95% CI: 1.4-2.4); conversely, the RRs for PG in the youngest age group was 1.3 (95% CI: 0.9-1.9) ( Figure 1 , upplementary Table 5). In regional sub-group analyses, we did not observe substantial ifferences in the magnitude of the RRs for fatal outcomes. In both ub-regions, we could not observe a clear age gradient with larger Rs in younger groups; except for SBP where there was an age gra- ient from age 45 years ( Figure 2 , Supplementary Table 6). For fatal and non-fatal outcomes, the 13 cohorts contributed ith 27,554 eligible individuals. Almost two thirds were men 64.1%), and the mean age was 52.3 (10.5) years. The mean BMI, BP, FPG and total cholesterol and non-HDL cholesterol was: 27.2 g/m 2 (SD = 5.1), 131 mmHg (SD = 21.0), 5.3 mmol/L (SD = 1.7), http://ncdrisc.org/ The Lancet Regional Health - Americas 4 (2021) 10 0 068 BMI (5 kg/m^2) SBP (10 mmHg) Glucose (1 mmol/L) Total Cholesterol (1 mmol/L) Non−HDL Cholesterol (1 mmol/L) N o R eg ressio n D ilu tio n B ias R eg ressio n D ilu tio n B ias 0.75 1.00 2.00 0.75 1.00 2.00 0.75 1.00 2.00 0.75 1.00 2.00 0.75 1.00 2.00 85+ 75−84 65−74 55−64 45−54 35−44 85+ 75−84 65−74 55−64 45−54 35−44 RR (95% CI) A ge G ro up s Outcome Non−Fatal + Fatal Fatal Figure 1. Age-specific relative risks for fatal and fatal plus non-fatal cardiovascular disease associated with usual levels of selected cardio-metabolic risk factors. While the upper panel shows estimates without accounting for regression dilution bias, the lower panel shows estimates accounting for regression dilution bias; all estimates were adjusted by sex and age (within each age group). Age groups based on age at risk. Estimates for fatal plus non-fatal events included only the first five age groups (insufficient observations in the eldest age group). RR: relative risk; 95% CI: 95% confidence interval; BMI: body mass index; SBP: systolic blood pressure. The red vertical line at relative risk = 1.5 and the orange vertical line at relative risk = 2.0 on the X-axis. SBP (10 mmHg) BMI (5 kg/m^2) Glucose (1 mmol/L) Total Cholesterol (1 mmol/L) Non−HDL Cholesterol (1 mmol/L) 0.50 0.75 1.00 2.00 4.00 0.50 0.75 1.00 2.00 4.00 0.50 0.75 1.00 2.00 4.00 0.50 0.75 1.00 2.00 4.00 0.50 0.75 1.00 2.00 4.00 85+ 75−84 65−74 55−64 45−54 35−44 RR (95% CI) A ge G ro up s Sub−region Central America & Caribbean South America Figure 2. Age-specific relative risks for fatal cardiovascular disease associated with usual levels of selected cardio-metabolic risk factors by sub-regions. All models were adjusted by sex and age (within each age group). Age groups based on age at risk (i.e. at outcome). RR: relative risk; 95% CI: 95% confidence interval; BMI: body mass index; SBP: systolic blood pressure. Only results adjusted for regression dilution bias are presented. Results as per multiple imputation. Insufficient observations to reliably compute these risk estimates for fatal plus non-fatal cardiovascular events. RR: relative risk; 95% CI: 95% confidence interval. 5 ( y t 1 p a n ( t m f ( ( F m o t F t t 1 a p 5 f c w n p i .3 mmol/L (SD = 1.1) and 4.1 mmol/L (SD = 1.1), respectively Supplementary Table 3). The mean follow-up was 8.5 (SD = 5.3) ears. In the 13 cohorts analysed for fatal and non-fatal events, here were 577 non-fatal events (246 (95% CI: 227-267) per 0 0,0 0 0 person-years) and 677 fatal events (288 (95% CI: 267-311) er 10 0,0 0 0 person-years). We observed an age gradient in the magnitude of the RDB- djusted RRs across all cardio-metabolic risk factors for fatal and on-fatal cardiovascular outcomes, with smaller RRs in older ages Figure 1 , Supplementary Table 5). The magnitude of the RRs in he youngest group (35-44) was at or above 1.2 ( Figure 1 , Supple- entary Table 5). In the youngest age group, the largest RR for atal and non-fatal cardiovascular outcomes was observed for SBP RR = 1.7, 95% CI: 1.2-2.4); conversely, the smallest RR was for FPG RR = 1.2, 95% CI: 0.8-1.9) ( Figure 1 , Supplementary Table 5). The age-specific RRs for fatal and non-fatal CVD for SBP and PG were remarkably similar to those reported from cohorts 4 ostly conducted in high-income countries ( Figure 3 ) [5] . For TC, ur RRs appeared to be smaller for the two youngest age groups, hough these differences were statistically insignificant ( Figure 3 ). or BMI, the RRs were consistently smaller in magnitude for par- icipants younger than 75 years old, with the largest difference for hose in the age group 55-64 years ( Figure 3 ): 1.24 (95% CI: 1.11- .38) vs. 1.50 (95% CI: 1.41-1.61). Non-optimal SBP was responsible for the largest proportion of ttributable cardiovascular deaths across countries, with a pro- ortional effect ranging from 30.7% among Cuban women to 8.0% among men from Grenada. The second largest proportion or attributable CVD mortality was due to non-optimal non-HDL holesterol, which proportional effect varied between 13.9% (Chile, omen) and 31.2% (Guyana, women). The proportional effect of on-optimal BMI and total cholesterol were smaller. For BMI the roportional effect ranged from 6.1% in women from Cuba to 19.6% n men from Saint Kitts and Nevis, whereas for total cholesterol The Lancet Regional Health - Americas 4 (2021) 10 0 068 BMI (5 kg/m^2) Glucose (1 mmol/L) SBP (10 mmHg) Total Cholesterol (1 mmol/L) 0.50 0.75 1.00 2.00 3.00 0.50 0.75 1.00 2.00 3.00 0.50 0.75 1.00 2.00 3.00 0.50 0.75 1.00 2.00 3.00 85+ 75−84 65−74 55−64 45−54 35−44 RR (95% CI) A ge g ro up s source PSC and APCSC CC−LAC Figure 3. Relative risks from the pooled analysis of PSC and APCSC [5] compared with those from LAC cohort pooling. Estimates from LAC cohorts are those adjusted by regression dilution bias and based on multiple imputation for fatal and non-fatal cardiovascular outcomes. Our estimates for fatal plus non-fatal outcomes were computed for the first five age groups only (insufficient observations in the oldest age group). PSC: Prospective Studies Collaboration; APCSC: Asia Pacific Cohort Studies Collaboration. RR: relative risk; 95% CI: 95% confidence interval. 0 20 40 60 0 20 40 60 PAF (%), Men PA F ( % ), W om en Risk Factor BMI Non−HDL SBP TC Sub−region Andean Caribbean Central Southern and Tropical Figure 4. Population attributable fraction (PAF, %) in women compared to men by risk factor and sub-regions. BMI: body mass index; SBP: systolic blood pressure. 5 The Lancet Regional Health - Americas 4 (2021) 10 0 068 20 30 70 20 20 30 60 20 40 90 180 40 30 80 150 50 30 50 100 20 20 40 80 20 40 60 160 30 20 50 110 30 Andean Caribbean Central Southern and Tropical M en W om en 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 0 50 100 150 200 250 Non−HDL TC SBP BMI Non−HDL TC SBP BMI Crude attributable death rate per 100,000 person−years Figure 5. Crude attributable death rates per 10 0,0 0 0 person-years by risk factors, sub-regions and sex. Results at the sub-region level are mean averaged accounting for the population size of the countries within each sub-region. BMI: body mass index; SBP: systolic blood pressure; TC: total cholesterol. The vertical lines along the numbers at the top of the bars represent the 95% credible interval. t w t 3 w 4 p a e F v i o t a v t t A w o f t t n A A p s ( r m w o e t n r c a ( ( e P G H P a d t ( 4 d a o g o g hese numbers were 4.4% (Guatemala, men) and 18.1% (Guyana, omen; Supplementary Figure 3). The proportional effect of non-optimal total cholesterol tended o be larger among women than among men, as was observed in 4 countries ( Figure 4 ); the largest absolute difference between omen and men was observed in Guatemala (11.6% in women vs .4% in men; Supplementary Figure 3). On the other hand, the pro- ortional effect of non-optimal SBP was higher among men than mong women in most countries ( Figure 4 ), with the largest differ- nce in Uruguay (47.5% in men vs 35.4% in women; Supplementary igure 3). Over half a million deaths (502,913 (95% credible inter- al = 340,637-653,242) out of a total of 1,094,795 CVD deaths n LAC) were attributable to non-optimal SBP in 2019. The sec- nd largest effect was estimated for non-optimal non-HDL choles- erol at 224,118 (95% credible interval = 83,755-388,176) deaths nd the lowest for non-optimal BMI at 119,498 (95% credible inter- al = 61,201-200,824) ( Table 1 ). Brazil, Mexico and Argentina, in hat order, had the largest numbers of attributable deaths across he four risk factors for both women and men; Colombia displaced rgentina from the third place regarding non-HDL cholesterol in omen (Supplementary Figure 3). Most of the cardiovascular disease deaths attributable to non- ptimal BMI were premature ( < 70 years of age; Table 1 ), ranging rom 72% among women in Southern and Tropical Latin America, o 83% among men in the Caribbean ( Table 1 ). On the other ex- reme, the 44% of all cardiovascular disease deaths attributable to on-optimal SBP was premature; this ranged from 31% (women in ndean Latin America) to 54% (men in Southern and Tropical Latin merica; Table 1 ). Across the four sub-regions, the crude attributable death rate er 10 0,0 0 0 person-years due to non-optimal risk factors was con- istently larger among men for BMI, SBP and non-HDL cholesterol Figure 5 ). Consistently across all sub-regions, non-optimal SBP was 6 esponsible for the largest number of attributable deaths for both en and women. Similarly, non-optimal non-HDL cholesterol al- ays ranked second ( Figure 5 ). For men, country-specific attributable death rates to non- ptimal SBP ( Figure 6 A, Supplementary Figure 3), was the small- st in Peru (54 per 10 0,0 0 0 person-years) and Guatemala (67), and he largest in Dominica (249) and Guyana (282). For non-optimal on-HDL cholesterol, we observed the smallest attributable death ates in Guatemala (25) and Peru (27), and the largest in Domini- an Republic (107) and Guyana (158). Finally, for BMI, the lowest ttributable death rates were observed in Peru (13) and Guatemala 14), and the largest in Saint Lucia (64) and Saint Kitts and Nevis 75). For women ( Figure 6 B, Supplementary Figure 3), the small- st attributable death rate to non-optimal SBP was estimated in eru (38 per 10 0,0 0 0 person-years) and Guatemala (64), while uyana (228) and Dominica (261) had the largest rates. For non- DL cholesterol, the lowest attributable rates were observed in eru (21) and Chile (30), while the largest rates were in Haiti (109) nd Guyana (142). For non-optimal BMI, the lowest attributable eath rates were observed in Peru (10) and Panama (14), whilst he largest rates were in Saint Kitts and Nevis (51) and Guyana 76). . Discussion Benefiting from a unique database of pooled individual-level ata from 31 cohort studies in 13 LAC countries [27] , we estimated ge-specific RRs for major cardiovascular disease risk factors. We bserved smaller RRs for BMI compared with those used in the lobal estimates of disease burden based on cohort collaborations riginating mostly in high-income countries. We observed an age radient whereby young people had higher RRs than older indi- T h e La n cet R eg io n a l H ea lth - A m erica s 4 (2 0 2 1 ) 10 0 0 6 8 Table 1 Number of cardiovascular deaths in 2019 attributable to each risk factor by sub-region and sex in Latin America and the Caribbean Region Sex BMI SBP TC Non-HDL Estimate Lower CI Upper CI Estimate Lower CI Upper CI Estimate Lower CI Upper CI Estimate Lower CI Upper CI All ages (ages 20 and above) Andean Latin America Men 3299 1605 5687 13590 8107 18597 3118 562 7135 6841 2548 11916 Andean Latin America Women 3124 1602 5450 11216 5757 16342 3532 622 8193 6189 2025 11376 Caribbean Men 6176 2832 10630 27503 16887 36885 6178 1123 13847 12863 5229 21835 Caribbean Women 5134 2312 9475 24135 12787 34567 7710 1593 17115 13125 4588 23590 Central Latin America Men 21147 11158 34447 81481 53892 107084 16548 3844 35438 39095 15439 65842 Central Latin America Women 16734 8649 28854 67962 40603 93829 20079 4416 43640 35646 11593 64701 Southern and Tropical Latin America Men 36418 19372 58019 155845 121477 187436 30617 9504 58140 61208 25953 99449 Southern and Tropical Latin America Women 27466 13672 48261 121181 81127 158501 32048 7881 66512 49151 16381 89467 Premature (below age 70) Andean Latin America Men 2659 1494 3920 6072 3905 7854 1851 473 3519 3886 2003 5631 Andean Latin America Women 2390 1474 3335 3466 1848 4883 1580 495 2855 2813 1490 4051 Caribbean Men 5136 2651 7880 14327 9244 18297 3828 930 7373 7824 4119 11345 Caribbean Women 3932 2104 5920 9617 5259 13180 3697 1248 6596 6342 3397 9102 Central Latin America Men 17209 10396 24328 40272 28791 49875 10278 3269 18462 22925 12156 32755 Central Latin America Women 12342 7816 16888 21806 14077 28469 8553 3340 14340 15098 8294 21297 Southern and Tropical Latin America Men 29324 17847 41325 83297 69701 94929 19949 7806 33399 38343 20426 54830 Southern and Tropical Latin America Women 19765 12201 27444 41732 31403 50817 14050 5875 22882 22272 11925 31983 Premature-to-all-ages ratio (%) Andean Latin America Men 80.60 93.13 68.92 44.68 48.16 42.23 59.37 84.25 49.33 56.81 78.60 47.26 Andean Latin America Women 76.52 92.02 61.19 30.91 32.10 29.88 44.74 79.54 34.85 45.45 73.56 35.61 Caribbean Men 83.16 93.61 74.13 52.09 54.74 49.61 61.97 82.80 53.24 60.82 78.78 51.96 Caribbean Women 76.59 90.99 62.47 39.85 41.12 38.13 47.95 78.35 38.54 48.32 74.04 38.59 Central Latin America Men 81.38 93.17 70.62 49.42 53.42 46.58 62.11 85.04 52.10 58.64 78.74 49.75 Central Latin America Women 73.75 90.36 58.53 32.09 34.67 30.34 42.60 75.62 32.86 42.35 71.54 32.92 Southern and Tropical Latin America Men 80.52 92.13 71.23 53.45 57.38 50.65 65.16 82.14 57.45 62.64 78.71 55.13 Southern and Tropical Latin America Women 71.96 89.25 56.86 34.44 38.71 32.06 43.84 74.54 34.40 45.31 72.80 35.75 BMI: body mass index; SBP: systolic blood pressure; TC: total cholesterol; Non-HDL: non-HDL cholesterol; CI: 95% credible interval. All ages included observations aged ≥20 years, whereas premature refers to ages between 20 and 69 years. The premature-to-all-ages ratio quantifies the ratio of the estimated attributable deaths below age 70 to the corresponding estimate for all ages expressed as a percentage. 7 The Lancet Regional Health - Americas 4 (2021) 10 0 068 BMI Non−HDL SBP TC Bolivia Ecuador Peru Belize Bermuda Antigua and Barbuda Bahamas Barbados Cuba Dominica Dominican Republic Saint Vincent and the Grenadines Trinidad and Tobago Guyana Grenada Puerto Rico HaitiSaint Lucia Jamaica Suriname Saint Kitts and Nevis Colombia Costa Rica El Salvador Guatemala Honduras Mexico Nicaragua Panama Venezuela Argentina Brazil Uruguay Paraguay Chile 10 80 130 180 280 Crude attributable death rate per 100,000 person−years, men Andean Latin America Caribbean Central Latin America Southern Latin America Figure 6. Crude attributable death rates per 10 0,0 0 0 person-years by risk factor and country in Latin America and the Caribbean in men (A) and women (B). Countries are clustered within sub-regions (Andean Latin America, the Caribbean, Central Latin America as well as Southern Latin America). Colour scale allows comparison within each wheel (risk factor) as well as within each column (country). v C A r c v r c [ o h h g r a f R d t r i o s d i p i h s t e o p i c f a e r u i d 1 b [ v t l H l r iduals. Our results suggested that the RRs did not differ between entral America & the Caribbean sub-region compared with South merica. The largest attributable CVD deaths across the selected isk factors were due to non-optimal SBP, followed by non-HDL holesterol. These risk factors had a much larger impact on cardio- ascular deaths in the Caribbean and Southern and Tropical sub- egions. The age gradient of the estimated RRs in our analysis is onsistent with prior pooled analysis of large cohort studies 2 , 5 , 9 , 10 , 13 , 40 ]. The magnitude of age-specific RRs was similar in ur analyses compared with prior pooling projects of cohorts in igh-income countries and those of the Asia-Pacific region [5] ; owever, for BMI, our estimated RRs were smaller for many age roups below the age of 75, particularly for people aged 55-64. The eported RRs for BMI from the Prospective Studies Collaboration, nd the Asia Pacific Cohort Studies Collaboration did not account or RDB [ 10 , 41 ]. Adjusting for RDB, would have led to even higher Rs compared with ours. Such similarity may reflect the same un- erlying biology of these risk factors and lack of major modifica- ions by lifestyle or environmental risk factors that do differ across egions. In fact, where patterns or lengths of exposure matter as t is the case for smoking and alcohol use, RRs of cardiovascular utcomes differ substantially by region [25] . In contrast, the ob- erved differences in RRs for BMI may be explained by the shorter uration of the weight gain in the LAC region compared with high- ncome countries. That is, high-income populations have been ex- osed to non-optimal BMI levels longer than most populations 8 n LAC [22] , and are therefore experiencing the larger cumulative armful effects of BMI on cardiovascular health. Alternatively, the ame level of BMI may correspond to a healthier body fat distribu- ion in LAC compared with high-income populations. The current vidence on such a difference in fat distribution at the same level f BMI is mixed [ 42 , 43 ] and further research is needed using larger opulation-based surveys with measurements of body composition n LAC. Our RRs for non-HDL cholesterol are consistent with a re- ent analysis of the PURE study, which did not find substantial dif- erences in RRs for non-HDL cholesterol between high-, middle- nd low-income countries [4] . The observed differences in RRs for BMI may explain the differ- nces in our estimates of attributable deaths to cardio-metabolic isk factors in LAC versus those reported by the GBD Study, which ses RRs mostly informed by epidemiological studies in high- ncome countries. For example, we estimated a crude attributable eath rate for non-optimal BMI in women in Peru of 10 per 0 0,0 0 0, compared with 18 cardiovascular disease deaths reported y the GBD Study [44] for Guyana we estimated 76 compared to 86 44] . Notably, the GBD Study risk estimates include other cardio- ascular outcomes besides those herein analysed- partly explaining he differences. Our results show that non-optimal SBP was responsible for the argest number of cardiovascular disease deaths, followed by non- DL cholesterol, total cholesterol and BMI. This ranking is simi- ar to the one proposed by the GBD Study in 2019, in which SBP anked first, followed by LDL-Cholesterol, fasting plasma glucose The Lancet Regional Health - Americas 4 (2021) 10 0 068 BMI Non−HDL SBP TC Bolivia Ecuador Peru Belize Bermuda Antigua and Barbuda Bahamas Barbados Cuba Dominica Dominican Republic Saint Vincent and the Grenadines Trinidad and Tobago Guyana Grenada Puerto Rico HaitiSaint Lucia Jamaica Suriname Saint Kitts and Nevis Colombia Costa Rica El Salvador Guatemala Honduras Mexico Nicaragua Panama Venezuela Argentina Brazil Uruguay Paraguay Chile 10 80 160 240 260 Crude attributable death rate per 100,000 person−years, women Andean Latin America Caribbean Central Latin America Southern Latin America Figure 6. Continued a e f b t v d l p s a w a s m r t w C r s t f L d a j o t w c s g a A o A e t l p c i r c f h f a w a a p w b a b C nd BMI [45] . This suggests that the ranking based on global risk stimates still apply to LAC, yet the burden attributable to each risk actor may be different. That difference, as herein proposed, may e overestimating the cardiovascular disease mortality attributable o non-optimal BMI in LAC. Arguably, LAC-based risk estimates -particularly for BMI- pro- ide more valid metrics for countries in LAC to quantify the bur- en of key cardio-metabolic risk factors. This evidence could al- ow prioritizing the risk factor(s) with the largest burden, develop olicies and interventions to address these priorities, and set up urveillance systems to monitor the progress towards international nd local goals. Our results could be taken as parameters upon hich goals can be set to reduce cardiovascular burden in LAC nd in each country in the region given that metrics to mea- ure the progress and surveillance of cardiovascular diseases were ostly informed by countries outside LAC. Considering the sharp ise in obesity and diabetes in the region [15] , despite our evidence hat shows lower RRs compared with high-income countries, over- eight/obesity remains one of the highest-ranking risk factors for VD; obesity control and prevention policies should continue to emain top priorities. Our work has several strengths. The risk estimates are age- pecific and were computed following consistent methods using he largest pooled database of cohorts in LAC. We analysed data rom 13 countries including at least one from each sub-region in AC, a work never conducted before. Analysing individual level ata, in contrast to published estimates [ 18 , 46 ], allowed us to ex- mine interactions between different variables. The RRs were ad- usted for regression dilution bias using LAC data providing the RR f “usual” exposure to risk factors. We used multiple imputation 9 o handle missing data for risk factors at baseline. Nevertheless, e acknowledge several limitations. Due to data availability, we ould not study other risk factors such as LDL-cholesterol. Likewise, ome outcomes were not available, preventing us from disentan- ling, for example, ischaemic from haemorrhagic stroke. We were lso unable to examine RRs in all sub-regions (e.g., Andean Latin merica and southern Latin America) due to the small numbers f events. We therefore only explored risk estimates from South merica with those from Central America and the Caribbean, and ven in this case, confidence intervals were wide, particularly in he youngest and oldest age groups. Many cohorts did not col- ect data on non-fatal events (possibly due to the younger age of articipants or complexities and costs of identifying and adjudi- ating non-fatal events), precluding a separate analysis. The lim- ted number of non-fatal events could have also affected the main esults (RRs for both fatal and non-fatal CVD), as these estimates ould have been mostly driven by fatal events; however, results or fatal outcomes only showed the same age pattern and the RRs ad a similar magnitude as those including both fatal and non- atal events. Mortality risk may also be confounded by health care ccess and control of non-communicable diseases, variables that ere not included in the regression models. A few variables had large proportion of missing values across cohorts mostly because subset of cohorts did not include these measurements in their rotocol, as opposed to non-response or missing measurements ithin each cohort. We used modelled estimates of CVD deaths y country, age and sex from the GBD 2019 Study to calculate the ttributable number of deaths which makes our results compara- le and consistent across countries [38] . However, the estimated VD mortality may be biased in countries especially if local data is The Lancet Regional Health - Americas 4 (2021) 10 0 068 n t t f ( p s i s a d L v h d d ( r c b t b B w w s i a f s d o s o D p r s a p 5 a b o R c v 6 s b 7 ( S t ( ( B C P T d P ( b c l t d d e e v ( d ( l t t C P d M G o C l p S B P G R d d C V E M R B H v s d t t C t e E M M e c a p v ot incorporated in the GBD analyses and/or if modelling assump- ions are not valid for a particular region/sub-region. Also in rela- ion to the GBD Study, we acknowledge that GBD deliver estimates or several years whereas we only used their most recent estimates 2019); we focused on the most recent year because we aimed to rovide estimates to inform policies and goal setting, rather than howing time patterns. Cohorts herein analysed for fatal outcomes ncluded more women than men; interpretation of these estimates hould be made in light of this profile. We only presented results t the country level. Future work should also study cardiovascular isease burden at the subnational level, ideally in all countries in AC considering its substantial geographical and socioeconomic di- ersity. We encourage researchers in LAC to use the risk estimates erein reported to conduct subnational analysis of cardiovascular isease burden. We pooled multiple cohorts which included a ran- om sample of the general population or studied specific groups e.g., The Mexican Teachers’ Cohort). We studied cardio-metabolic isk factors (e.g., blood pressure and total cholesterol) which were ollected following objective, standard and comparable methods etween cohorts. The risk of selection bias is quite low because he probability of being selected in these studies is unlikely to e simultaneously related to the exposure and outcome. Regarding MI, except for one cohort we used measured weight and height hich reduces measurement error; this is method is consistent ith other cohort pooling projects. In conclusion, using data from the first pooling project of cohort tudies in LAC we found that RRs of cardiovascular disease per unit ncrease in blood pressure, glucose and cholesterol are remark- bly similar to previous pooling projects that used data mostly rom high-income countries. In contrast, we observed smaller age- pecific RRs for BMI. The estimated RRs offer region-specific evi- ence that can be used to update estimates of attributable burden f disease to better inform regional policies and goals. One of the trategic lines of action in Pan American Health Organization’s Plan f Action for the Prevention and Control of Non-communicable iseases in the Americas 2013-2019, was to strengthen country ca- acity for surveillance on non-communicable diseases and their isk factors [47] . Our results can help improve the validity of uch surveillance effort s by emphasizing the use of local data nd evidence in prioritizing and implementing CVD prevention rograms. . Contributions GD, RMCL and ME conceived the study. RMCL curated the data nd conducted all analysis with input from GD and ME. All mem- ers of the CC-LAC Steering committee contributed to the design f the analysis and interpretation of the results and conclusions. MCL wrote the first draft of the manuscript and all co-authors ontributed to the revisions. GD and RMCL have access to and have erified the underlying data. . Data sharing Data is currently only available to CC-LAC collaborators. Expres- ions of interest to access the CC-LAC data are welcomed and will e handled by the CC-LAC steering committee. . Cohorts Consortium of Latin America and the Caribbean CC-LAC) Steering committee Rodrigo M Carrillo-Larco (Imperial College London, UK); Dalia tern (National Institute of Public Health, Mexico); Ian R Hamble- on (The University of the West Indies, Barbados); Anselm Hennis w 10 Pan American Health Organization, USA); Mariachiara Di Cesare Middlesex University, UK); Paulo Lotufo (University of São Paulo, razil); Catterina Ferreccio (Pontificia Universidad Católica de Chile, hile); Vilma Irazola (Institute for Clinical Effectiveness and Health olicy, Argentina); Pablo Perel (London School of Hygiene and ropical Medicine, UK); Edward W Gregg (Imperial College Lon- on, UK); J Jaime Miranda (Universidad Peruana Cayetano Heredia, eru); Majid Ezzati (Imperial College London, UK); Goodarz Danaei Harvard T.H. Chan School of Public Health, USA) Country and Regional Data ( ∗ equal contribution; listed alpha- etically by surname) Carlos A Aguilar-Salinas (Instituto Nacional de Ciencias Médi- as y Nutrición, México) ∗; Ramón Alvarez-Váz (Universidad de a República, Uruguay) ∗; Marselle B Amadio (Centro Universi- ario Senac Santo Amaro, Brazil) ∗; Cecilia Baccino (Universidad e la República, Uruguay) ∗; Claudia Bambs (Pontificia Universi- ad Católica de Chile, Chile) ∗; João Luiz Bastos (Universidade Fed- ral de Santa Catarina, Brazil) ∗; Gloria Beckles (Centers for Dis- ase Control and Prevention, USA) ∗; Antonio Bernabe-Ortiz (Uni- ersidad Peruana Cayetano Heredia, Perú) ∗; Carla DO Bernardo The University of Adelaide, Australia) ∗; Katia V Bloch (Universi- ade Federal do Rio de Janeiro (UFRJ), Brazil) ∗; Juan E Blümel Universidad de Chile, Chile) ∗; Jose G Boggia (Universidad de a República, Uruguay) ∗; Pollyanna K Borges (Universidade Es- adual de Ponta Grossa, Brazil) ∗; Miguel Bravo (MELISA Insti- ute, Chile) ∗; Gilbert Brenes-Camacho (Universidad de Costa Rica, osta Rica) ∗; Horacio A Carbajal (Universidad Nacional de la lata, Argentina) ∗; Maria S Castillo Rascon (Universidad Nacional e Misiones, Argentina) ∗; Blanca H Ceballos (Hospital Dr Ramon adariaga, Argentina) ∗; Veronica Colpani (Federal University of Rio rande do Sul, Brazil) ∗; Jackie A Cooper (Queen Mary University f London, UK) ∗; Sandra Cortes (Pontificia Universidad Católica de hile, Chile) ∗; Adrian Cortes-Valencia (National Institute of Pub- ic Health, Mexico) ∗; Roberto S Cunha (Federal University of Es- írito Santo, Brazil) ∗; Eleonora d’Orsi (Universidade Federal de anta Catarina, Brazil) ∗; William H Dow (University of California, erkeley, USA) ∗; Walter G Espeche (Universidad Nacional de la lata, Argentina) ∗; Flavio D Fuchs (Universidade Federal do Rio rande do Sul, Brazil) ∗; Sandra C Fuchs (Universidade Federal do io Grande do Sul, Brazil) ∗; Suely GA Gimeno (Universidad Federal e São Paulo, Brazil) ∗; Donaji Gomez-Velasco (Instituto Nacional e Ciencias Médicas y Nutrición, México) ∗; David A Gonzalez- hica (The University of Adelaide, Australia) ∗; Clicerio Gonzalez- illalpando (Instituto Nacional de Salud Pública, México) ∗; María- lena Gonzalez-Villalpando (Centro de Estudios en Diabetes A.C., éxico) ∗; Gonzalo Grazioli (Hospital Churruca Visca, Argentina) ∗; icardo O Guerra (Federal University of Rio Grande do Norte, razil) ∗; Laura Gutierrez (Institute for Clinical Effectiveness and ealth Policy, Argentina) ∗; Fernando L Herkenhoff (Federal Uni- ersity of Espírito Santo, Brazil) ∗; Andrea RVR Horimoto (Univer- ity of São Paulo, Brazil) ∗; Andrea Huidobro (Universidad Católica el Maule, Chile) ∗; Elard Koch (MELISA Institute, Chile) ∗; Mar- in Lajous (Harvard T.H. Chan School of Public Health, USA; Na- ional Institute of Public Health, Mexico) ∗; Maria Fernanda Lima- osta (Oswaldo Cruz Foundation, Brazil) ∗; Ruy Lopez-Ridaura (Na- ional Institute of Public Health, Mexico) ∗; Alvaro CC Maciel (Fed- ral University of Rio Grande do Norte, Brazil) ∗; Betty S Manrique- spinoza (National Institute of Public Health, Mexico) ∗; Larissa P arques (Universidade Federal de Santa Catarina, Brazil) ∗; Jose G ill (Federal University of Espírito Santo, Brazil) ∗; Leila B Mor- ira (Universidade Federal do Rio Grande do Sul, Brazil) ∗; Os- ar M Muñoz (Pontificia Universidad Javeriana, Colombia) ∗; Lari- ne M Ono (Universidade Federal do Paraná, Brazil) ∗; Karen Op- ermann (Passo Fundo University, Brazil) ∗; Karina M Paiva (Uni- ersidade Federal de Santa Catarina, Brazil) ∗; Sergio V Peixoto (Os- aldo Cruz Foundation, Brazil) ∗; Alexandre C Pereira (University of The Lancet Regional Health - Americas 4 (2021) 10 0 068 S S i o ( P h M R b A C A H o M d d H v C V n D r l s e t S f R [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ [ ão Paulo, Brazil) ∗; Karen G Peres (NDRIS/NDCS Duke-NUS Medical chool, Singapore) ∗; Marco A Peres (NDRIS/NDCS Duke-NUS Med- cal School, Singapore) ∗; Paula Ramírez-Palacios (IMSS Epidemiol- gy and Health Services Research Unit, Mexico) ∗; Cassiano R Rech Universidade Federal de Santa Catarina, Brazil) ∗; Berenice Rivera- aredez (National Autonomous University of Mexico, Mexico) ∗; No- ora I Rodriguez (Clinica de Marly, Colombia) ∗; Rosalba Rojas- artinez (Instituto Nacional de Salud Pública, México) ∗; Luis osero-Bixby (Universidad de Costa Rica, Costa Rica) ∗; Adolfo Ru- instein (Institute for Clinical Effectiveness and Health Policy, rgentina) ∗; Alvaro Ruiz-Morales (Pontificia Universidad Javeriana, olombia) ∗; Martin R Salazar (Universidad Nacional de la Plata, rgentina) ∗; Aaron Salinas-Rodriguez (National Institute of Public ealth, Mexico) ∗; Jorge Salmerón (National Autonomous University f Mexico, Mexico) ∗; Ramon A Sanchez (Universidad Nacional de isiones, Argentina) ∗; Nelson AS Silva (Universidade Federal do Rio e Janeiro (UFRJ), Brazil) ∗; Thiago LN Silva (Universidade federal o Rio de Janeiro (UFRJ), Brazil) ∗; Liam Smeeth (London School of ygiene & Tropical Medicine, UK) ∗; Poli M Spritzer (Federal Uni- ersity of Rio Grande do Sul, Brazil) ∗; Fiorella Tartaglione (Hospital hurruca Visca, Argentina) ∗; Jorge Tartaglione (Hospital Churruca isca, Argentina) ∗; Rafael Velázquez-Cruz (National Institute of Ge- omic Medicine (INMEGEN), Mexico) ∗ eclaration of Competing Interest The authors declare no conflict of interests. The funders had no ole in study design, data collation and analysis, decision to pub- ish, or preparation of the manuscript. The authors alone are re- ponsible for the views expressed in this paper, which do not nec- ssarily represent the views, decisions, or policies of the institu- ions with which the authors are affiliated. upplementary materials Supplementary material associated with this article can be ound, in the online version, at doi: 10.1016/j.lana.2021.10 0 068 . eferences [1] GBD 2019 Demographics CollaboratorsGlobal age-sex-specific fertility, mor- tality, healthy life expectancy (HALE), and population estimates in 204 countries and territories, 1950-2019: a comprehensive demographic analy- sis for the Global Burden of Disease Study 2019. Lancet (London, England) 2020;396(10258):1160–203 . [2] GBD 2019 Risk Factors Collaborators. Global burden of 87 risk factors in 204 countries and territories1990-2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet (London, England) 2020;396(10258):1223–49 . [3] Global Burden of Metabolic Risk Factors for Chronic Diseases CollaborationCar- diovascular disease, chronic kidney disease, and diabetes mortality burden of cardiometabolic risk factors from 1980 to 2010: a comparative risk assessment. The lancet Diabetes & endocrinology 2014;2(8):634–47 . [4] Yusuf S , Joseph P , Rangarajan S , et al. Modifiable risk factors, cardiovascular disease, and mortality in 155 722 individuals from 21 high-income, middle-in- come, and low-income countries (PURE): a prospective cohort study. Lancet (London, England) 2020;395(10226):795–808 . [5] Singh GM , Danaei G , Farzadfar F , et al. The age-specific quantitative effects of metabolic risk factors on cardiovascular diseases and diabetes: a pooled anal- ysis. PLoS One 2013;8(7):e65174 . [6] Woodward M , Barzi F , Martiniuk A , et al. Cohort profile: the Asia Pacific Cohort Studies Collaboration. Int J Epidemiol 2006;35(6):1412–16 . [7] Woodward M , Huxley R , Ueshima H , Fang X , Kim HC , Lam TH . The Asia pacific cohort studies collaboration: a decade of achievements. Global heart 2012;7(4):343–51 . [8] Lewington S , Clarke R , Qizilbash N , Peto R , Collins R . Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet (London, England) 2002;360(9349):1903–13 . [9] Lewington S , Whitlock G , Clarke R , et al. Blood cholesterol and vascu- lar mortality by age, sex, and blood pressure: a meta-analysis of individ- ual data from 61 prospective studies with 55,0 0 0 vascular deaths. Lancet 2007;370(9602):1829–39 . [10] Whitlock G , Lewington S , Sherliker P , et al. Body-mass index and cause-specific mortality in 900 000 adults: collaborative analyses of 57 prospective studies. Lancet (London, England) 2009;373(9669):1083–96 . 11 [11] Di Angelantonio E , Sarwar N , Perry P , et al. Major lipids, apolipoproteins, and risk of vascular disease. JAMA 2009;302(18):1993–2000 . [12] Sarwar N , Gao P , Seshasai SR , et al. Diabetes mellitus, fasting blood glucose concentration, and risk of vascular disease: a collaborative meta-analysis of 102 prospective studies. Lancet (London, England) 2010;375(9733):2215–22 . [13] Wormser D , Kaptoge S , Di Angelantonio E , et al. Separate and com- bined associations of body-mass index and abdominal adiposity with car- diovascular disease: collaborative analysis of 58 prospective studies. Lancet 2011;377(9771):1085–95 . [14] Carrillo-Larco RM , Demaio AR , Miranda JJ . Addressing NCDs: is it really a global coalition? Lancet (London, England) 2013;381(9883):2081 discussion -2 . [15] NCD Risk Factor Collaboration (NCD-RisC)—Americas Working GroupTrends in cardiometabolic risk factors in the Americas between 1980 and 2014: a pooled analysis of population-based surveys. The Lancet Global health 2020;8(1):e123–ee33 . [16] United Nations | Department of Economic and Social Affairs. 2018 Revision of World Urbanization Prospects. URL: https://www.un.org/development/desa/ publications/2018- revision- of- world- urbanization- prospects.html (Accessed July 11, 2020). [17] World Economic Forum. Latin Americas cities are ready to take off. But their infrastructure is failing them. URL: https://www.weforum.org/agenda/2018/ 06/latin- america- cities- urbanization- infrastructure- failing- robert- muggah/ (Accessed July 11, 220). [18] Carrillo-Larco RM , Barengo NC , Albitres-Flores L , Bernabe-Ortiz A . The risk of mortality among people with type 2 diabetes in Latin America: A sys- tematic review and meta-analysis of population-based cohort studies. Dia- betes/metabolism research and reviews 2019;35(4):e3139 . [19] Atun R , de Andrade LO , Almeida G , et al. Health-system reform and universal health coverage in Latin America. Lancet (London, England) 2015;385(9974):1230–47 . 20] Kruk ME , Gage AD , Joseph NT , Danaei G , Garcia-Saiso S , Salomon JA . Mortal- ity due to low-quality health systems in the universal health coverage era: a systematic analysis of amenable deaths in 137 countries. Lancet (London, Eng- land) 2018 . [21] NCD Risk Factor Collaboration (NCD-RisC)Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4.4 million par- ticipants. Lancet (London, England) 2016;387(10027):1513–30 . 22] NCD Risk Factor Collaboration (NCD-RisC)Worldwide trends in body-mass in- dex, underweight, overweight, and obesity from 1975 to 2016: a pooled anal- ysis of 2416 population-based measurement studies in 128.9 million children, adolescents, and adults. Lancet 2017;390(10113):2627–42 . 23] NCD Risk Factor Collaboration (NCD-RisC)Worldwide trends in blood pressure from 1975 to 2015: a pooled analysis of 1479 population-based measurement studies with 19.1 million participants. Lancet 2017;389(10064):37–55 . 24] O’Donnell MJ , Chin SL , Rangarajan S , et al. Global and regional effects of potentially modifiable risk factors associated with acute stroke in 32 countries (INTERSTROKE): a case-control study. Lancet (London, England) 2016;388(10046):761–75 . 25] Yusuf S , Hawken S , Ounpuu S , et al. Effect of potentially modifiable risk factors associated with myocardial infarction in 52 countries (the INTERHEART study): case-control study. Lancet (London, England) 2004;364(9438):937–52 . 26] Yusuf S , Rangarajan S , Teo K , et al. Cardiovascular risk and events in 17 low-, middle-, and high-income countries. N Engl J Med 2014;371(9):818–27 . 27] Cohorts Consortium of Latin America and the Caribbean (CC-LAC)Cohort Pro- file: The Cohorts Consortium of Latin America and the Caribbean (CC-LAC). International journal of epidemiology 2020;49(5):1437g . 28] Muñoz V OM , ÁJ Ruiz Morales , Mariño Correa A , Bustos CMM . Concordan- cia entre los modelos de SCORE y Framingham y las ecuaciones AHA/ACC como evaluadores de riesgo cardiovascular. Revista Colombiana de Cardiología 2017;24(2):110–16 . 29] Blümel JE , Carrillo-Larco RM , Vallejo MS , Chedraui P . Multimorbidity in a co- hort of middle-aged women: Risk factors and disease clustering. Maturitas 2020;137:45–9 . 30] Tartaglione JE , Grazioli GC , Sarmiento MP , LMJRAdC Goldstraj . Eventos cardio- vasculares en una población cerrada: Seguimiento a 10 años. Revista Argentina de Cardiologia 2008;76:347–51 . [31] Lajous M , Ortiz-Panozo E , Monge A , et al. Cohort Profile: The Mexican Teach- ers’ Cohort (MTC). International journal of epidemiology 2017;46(2):e10 . 32] Denova-Gutiérrez E , Flores YN , Gallegos-Carrillo K , et al. Health workers cohort study: methods and study design. J Salud Pública de México 2016;58:708–16 . 33] Lawes CM , Bennett DA , Parag V , et al. Blood pressure indices and cardiovascu- lar disease in the Asia Pacific region: a pooled analysis. Hypertension (Dallas, Tex: 1979) 2003;42(1):69–75 . 34] van Buuren S . Groothuis-Oudshoorn K. mice: Multivariate Imputation by Chained Equations in R. Journal of Statistical Software 2011;1(3):2011 . 35] Frost C , Thompson SG . Correcting for Regression Dilution Bias: Comparison of Methods for a Single Predictor Variable. Journal of the Royal Statistical Society Series A (Statistics in Society) 20 0 0;163(2):173–89 . 36] Clarke R , Shipley M , Lewington S , et al. Underestimation of risk associations due to regression dilution in long-term follow-up of prospective studies. Am J Epidemiol 1999;150(4):341–53 . 37] Ezzati M , Lopez AD , Rodgers A , Vander Hoorn S , Murray CJ . Selected major risk factors and global and regional burden of disease. Lancet (London, England) 2002;360(9343):1347–60 . 38] Global Burden of Disease Collaborative Network. Global Burden of Disease Study 2019 (GBD 2019) Results. Seattle, United States: Institute for Health Met- https://doi.org/10.1016/j.lana.2021.100068 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0001 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0002 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0003 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0004 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0004 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0004 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0004 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0004 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0005 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0005 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0005 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0005 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0005 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0006 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0006 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0006 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0006 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0006 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0007 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0008 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0008 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0008 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0008 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0008 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0008 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0009 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0009 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0009 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0009 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0009 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0010 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0010 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0010 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0010 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0010 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0011 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0011 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0011 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0011 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0011 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0012 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0012 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0012 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0012 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0012 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0013 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0013 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0013 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0013 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0013 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0014 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0014 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0014 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0014 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0015 https://www.un.org/development/desa/publications/2018-revision-of-world-urbanization-prospects.html https://www.weforum.org/agenda/2018/06/latin-america-cities-urbanization-infrastructure-failing-robert-muggah/ http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0018 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0018 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0018 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0018 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0018 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0019 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0019 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0019 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0019 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0019 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0020 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0021 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0022 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0023 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0024 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0024 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0024 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0024 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0024 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0025 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0025 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0025 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0025 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0025 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0026 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0026 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0026 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0026 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0026 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0027 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0028 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0028 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0028 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0028 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0028 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0029 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0029 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0029 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0029 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0029 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0030 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0030 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0030 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0030 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0030 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0031 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0031 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0031 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0031 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0031 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0032 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0032 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0032 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0032 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0032 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0033 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0033 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0033 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0033 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0033 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0034 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0034 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0035 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0035 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0035 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0036 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0036 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0036 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0036 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0036 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0037 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0037 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0037 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0037 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0037 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0037 The Lancet Regional Health - Americas 4 (2021) 10 0 068 [ [ [ [ [ [ [ rics and Evaluation (IHME), 2020. Available from http://ghdx.healthdata.org/ gbd- results- tool . 39] Ortiz-Panozo E , Yunes-Diaz E , Lajous M , Romieu I , Monge A , Lopez-Ridaura R . Validity of self-reported anthropometry in adult Mexican women. Salud pub- lica de Mexico 2017;59(3):266–75 . 40] Lawes CM , Rodgers A , Bennett DA , et al. Blood pressure and cardiovascular disease in the Asia Pacific region. Journal of hypertension 2003;21(4):707–16 . [41] Ni Mhurchu C , Rodgers A , Pan WH , Gu DF , Woodward M . Body mass index and cardiovascular disease in the Asia-Pacific Region: an overview of 33 co- horts involving 310 0 0 0 participants. International journal of epidemiology 2004;33(4):751–8 . 42] Carroll JF , Chiapa AL , Rodriquez M , et al. Visceral fat, waist circumference, and BMI: impact of race/ethnicity. Obesity (Silver Spring, Md) 20 08;16(3):60 0–7 . 43] Fernández JR , Heo M , Heymsfield SB , et al. Is percentage body fat differentially related to body mass index in Hispanic Americans, African Americans, and Eu- ropean Americans? The American journal of clinical nutrition 2003;77(1):71–5 . 12 44] Institute for Health Metrics | GHDx | Viz Hub. GBD Compare. URL: http: //ihmeuw.org/5f70 . 45] Institute for Health Metrics | GHDx | Viz Hub. GBD Compare. URL: http: //ihmeuw.org/5f71 . 46] Carrillo-Larco RM , Albitres-Flores L , Barengo NC . Bernabe-Ortiz A. The associ- ation between serum lipids and risk of premature mortality in Latin Amer- ica: a systematic review of population-based prospective cohort studies. PeerJ 2019;7:e7856 . [47] Bogers RP , Bemelmans WJ , Hoogenveen RT , et al. Association of overweight with increased risk of coronary heart disease partly independent of blood pressure and cholesterol levels: a meta-analysis of 21 cohort studies includ- ing more than 300 000 persons. ArchInternMed 2007;167(16):1720–8 . http://ghdx.healthdata.org/gbd-results-tool http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0039 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0040 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0040 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0040 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0040 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0040 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0041 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0041 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0041 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0041 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0041 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0041 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0042 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0042 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0042 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0042 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0042 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0043 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0043 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0043 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0043 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0043 http://ihmeuw.org/5f70 http://ihmeuw.org/5f71 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0046 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0046 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0046 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0046 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0047 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0047 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0047 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0047 http://refhub.elsevier.com/S2667-193X(21)00064-8/sbref0047 Impact of common cardio-metabolic risk factors on fatal and non-fatal cardiovascular disease in Latin America and the Caribbean: an individual-level pooled analysis of 31 cohort studies Evidence before this study Added value of this study Implications for all the available evidence 1 Introduction 2 Methods 2.1 Cardiovascular outcomes 2.2 Cardio-metabolic risk factors 2.3 Statistical analysis 2.3.1 Handling of missing data 2.3.2 Adjusting for Regression Dilution Bias (RDB) 2.3.3 Survival analysis 2.3.4 Quantifying the population-level impact of risk factors 2.4 Risk of bias in each study 2.4.1 Role of the funding source 3 Results 4 Discussion 5 Contributions 6 Data sharing 7 Cohorts Consortium of Latin America and the Caribbean (CC-LAC) Declaration of Competing Interest Supplementary materials References