ABSTRACT

OBJECTIVE:

To determine the possible existence of differences in blood pressure change over time according to stature in Mexican adults.

METHODS:

We analyzed the National Household Living Standards Survey databases following household members between 2005 and 2009. We selected participants who were between 20 and 40 years old (n = 7,130). We estimated multilevel models with random intercept to analyze differences in blood pressure changes according to stature. We adjusted the models for age, locality size, geographic region, per capita family income, waist-to-height ratio, physical activity, alcohol consumption, smoking, and use of antihypertensive drugs.

RESULTS:

In both sexes, baseline blood pressure tended to be lower as stature decreased. The differences were maintained in both the crude and adjusted models. In men, the increases in systolic pressure over time tended to be higher as stature increased.

CONCLUSIONS:

Contrary to what studies observed in high-income countries, in Mexico blood pressure was positively associated with stature.

DESCRIPTORS:
Adult; Blood Pressure; Anthropometry; Hypertension; Risk Factors; Longitudinal studies

INTRODUCTION

Arterial hypertension is highly prevalent in Mexico, with almost one third of adults suffering from it in 2012¹1 Gutiérrez J, Rivera-Dommarco J, Shamah-Levy T, Villalpando-Hernández S, Franco A, Cuevas-Nasu L, et al. Encuesta Nacional de Salud y Nutrición 2012: resultados nacionales. Cuernavaca (MX): Instituto Nacional de Salud Pública; 2012.. It is the most common condition in primary care; in 2010, it was the main cause of death and disability adjusted life years. It is a risk factor for several of the main causes of mortality²2 Whelton PK, Carey RM, Aronow WS, Casey DJ Jr, Collins KJ, Himmelfarb CD, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13-e115. https://doi.org/10.1161/HYP.0000000000000065
https://doi.org/10.1161/HYP.000000000000... . This contrasts with the fact that it is a preventable risk factor.

In recent years, cohort studies conducted in mainly European and Brazilian populations have established a negative association between stature and the risk of developing cardiovascular disease³3 De Lucia Rolfe E, França GVA, Vianna CA, Gigante DP, Miranda JJ, Yudkin JS, et al. Associations of stunting in early childhood with cardiometabolic risk factors in adulthood. PloS One. 2018;13(4):e0192196. https://doi.org/10.1371/journal.pone.0192196
https://doi.org/10.1371/journal.pone.019... –⁷7 Grillo LP, Gigante DP, Horta BL, Barros FCF. Childhood stunting and the metabolic syndrome components in young adults from a Brazilian birth cohort study. Eur J Clin Nutr. 2016;70(5):548-53. https://doi.org/10.1038/ejcn.2015.220
https://doi.org/10.1038/ejcn.2015.220... . In different populations, an inverse association has been established between blood pressure and stature, tending to be higher in those who are shorter⁴4 Korhonen PE, Kautiainen H, Eriksson JG. The shorter the person, the higher the blood pressure a birth cohort study. J Hypertens. 2017;35(6):1170-7. https://doi.org/10.1097/HJH.0000000000001300
https://doi.org/10.1097/HJH.000000000000... ,⁵5 Batty GD, Shipley MJ, Gunnell D, Huxley R, Kivimaki M, Woodward M, et al. Height, wealth, and health: an overview with new data from three longitudinal studies. Econ Hum Biol. 2009;7(2):137-52. https://doi.org/10.1016/j.ehb.2009.06.004
https://doi.org/10.1016/j.ehb.2009.06.00... ,⁸8 Das Gupta R, Haider SS, Hashan MR, Rahman MA, Sarker M. Association between height and hypertension in the adult Nepalese population: findings from a nationally representative survey. Health Sci Rep. 2019;2(12):e141. https://doi.org/10.1002/hsr2.141
https://doi.org/10.1002/hsr2.141... –¹¹11 Langenberg C, Hardy R, Breeze E, Kuh D, Wadsworth MEJ. Influence of short stature on the change in pulse pressure, systolic and diastolic blood pressure from age 36 to 53 years: an analysis using multilevel models. Int J Epidemiol. 2005;34(4):905-13. https://doi.org/10.1093/ije/dyi071
https://doi.org/10.1093/ije/dyi071... .

These studies are consistent with the hypothesis proposed by Barker, who postulated that adverse conditions during prenatal life and childhood increase the risk of suffering chronic diseases in adulthood¹²12 Barker D. Fetal origins of coronary heart disease. BMJ. 1995;311(6998):171-4. https://doi.org/10.1136/bmj.311.6998.171
https://doi.org/10.1136/bmj.311.6998.171... . This theory holds that adversities present during early stages of life produce structural and physiological modifications in the organism. Part of these changes result in short stature; those who have deficiencies during development generally reach shorter stature in adulthood³3 De Lucia Rolfe E, França GVA, Vianna CA, Gigante DP, Miranda JJ, Yudkin JS, et al. Associations of stunting in early childhood with cardiometabolic risk factors in adulthood. PloS One. 2018;13(4):e0192196. https://doi.org/10.1371/journal.pone.0192196
https://doi.org/10.1371/journal.pone.019... ,⁵5 Batty GD, Shipley MJ, Gunnell D, Huxley R, Kivimaki M, Woodward M, et al. Height, wealth, and health: an overview with new data from three longitudinal studies. Econ Hum Biol. 2009;7(2):137-52. https://doi.org/10.1016/j.ehb.2009.06.004
https://doi.org/10.1016/j.ehb.2009.06.00... .

It possible to explain the inverse association between stature and blood pressure by different mechanisms: in shorter individuals the diameter of the blood vessels is smaller¹³13 Hwaung P, Heo M, Bourgeois B, Kennedy S, Shepherd J, Heymsfield SB. Greater height is associated with a larger carotid lumen diameter. Medicines (Basel). 2019;6(2):57. https://doi.org/10.3390/medicines6020057
https://doi.org/10.3390/medicines6020057... ,¹⁴14 Norman M. Low birth weight and the developing vascular tree: a systematic review. Acta Pediatr. 2008;97(9):1165-72. https://doi.org/10.1111/j.1651-2227.2008.00904.x
https://doi.org/10.1111/j.1651-2227.2008... ; they have a smaller number of capillaries and less elasticity of the large vessels¹⁴14 Norman M. Low birth weight and the developing vascular tree: a systematic review. Acta Pediatr. 2008;97(9):1165-72. https://doi.org/10.1111/j.1651-2227.2008.00904.x
https://doi.org/10.1111/j.1651-2227.2008... ; as well as greater arterial intima-media thickness and greater incidence of atherosclerosis¹⁴14 Norman M. Low birth weight and the developing vascular tree: a systematic review. Acta Pediatr. 2008;97(9):1165-72. https://doi.org/10.1111/j.1651-2227.2008.00904.x
https://doi.org/10.1111/j.1651-2227.2008... . These mechanisms contribute to a decrease in vascular lumen, increasing peripheral resistance¹³13 Hwaung P, Heo M, Bourgeois B, Kennedy S, Shepherd J, Heymsfield SB. Greater height is associated with a larger carotid lumen diameter. Medicines (Basel). 2019;6(2):57. https://doi.org/10.3390/medicines6020057
https://doi.org/10.3390/medicines6020057... ,¹⁴14 Norman M. Low birth weight and the developing vascular tree: a systematic review. Acta Pediatr. 2008;97(9):1165-72. https://doi.org/10.1111/j.1651-2227.2008.00904.x
https://doi.org/10.1111/j.1651-2227.2008... .

The physical characteristics of the Mexican population are not comparable with the studied populations; therefore, it is not possible to assume that the results of these studies are valid for this population. For example, the average stature in Mexico is 156.8 and 169.0 cm for women and men, respectively; while in Brazil it is 160.9 and 173.6 cm; and in the United Kingdom it is 169.5 and 177.5 cm¹⁵15 Longás H. Altura media por países. El País (Madrid). 26 Jul 2016 [cited 2020 Sep 11]. Available from: https://elpais.com/elpais/2016/07/21/media/1469127433_712478.html
https://elpais.com/elpais/2016/07/21/med... . This characteristic is similar with populations from other Latin American countries such as Colombia, Honduras and Bolivia, which have an average stature similar to that of Mexico¹⁶16 Grasgruber P, Hrazdíra E. Nutritional and socio-economic predictors of adult height in 152 world populations. Econ Hum Biol. 2020;37:100848. https://doi.org/10.1016/j.ehb.2020.100848
https://doi.org/10.1016/j.ehb.2020.10084... . Differences in stature are probably the result of genetic characteristics and living conditions. Genetically, the Mexican population differs from the Brazilian and European populations because, although our population is a mixture of Amerindian, European and African genes, the Amerindian component predominates¹⁷17 Ruiz-Linares A, Adhikari K, Acuña-Alonzo V, Quinto-Sanchez M, Jaramillo C, Arias W, et al. Admixture in Latin America: geographic structure, phenotypic diversity and self-perception of ancestry based on 7,342 individuals. PLoS Genet. 2014;10(9):e1004572 https://doi.org/10.1371/journal.pgen.1004572
https://doi.org/10.1371/journal.pgen.100... . In addition to the above, there are genetic differences (higher proportion of indigenous population), differences in living conditions (less human development) and feeding patterns (lower consumption of foods of animal origin) making the Mexican population different from that studied populations¹⁶16 Grasgruber P, Hrazdíra E. Nutritional and socio-economic predictors of adult height in 152 world populations. Econ Hum Biol. 2020;37:100848. https://doi.org/10.1016/j.ehb.2020.100848
https://doi.org/10.1016/j.ehb.2020.10084... .

On the other hand, in Mexico, around 1.5 million children under five years old were chronically undernourished in 2012¹1 Gutiérrez J, Rivera-Dommarco J, Shamah-Levy T, Villalpando-Hernández S, Franco A, Cuevas-Nasu L, et al. Encuesta Nacional de Salud y Nutrición 2012: resultados nacionales. Cuernavaca (MX): Instituto Nacional de Salud Pública; 2012., favoring short stature in adulthood⁵5 Batty GD, Shipley MJ, Gunnell D, Huxley R, Kivimaki M, Woodward M, et al. Height, wealth, and health: an overview with new data from three longitudinal studies. Econ Hum Biol. 2009;7(2):137-52. https://doi.org/10.1016/j.ehb.2009.06.004
https://doi.org/10.1016/j.ehb.2009.06.00... . Therefore, this study sought to determine the existence of differences in changes in blood pressure over time according to stature in Mexican adults, considering the hypothesis that individuals of shorter stature will have a greater increase in blood pressure.

METHODS

To achieve the proposed objective, we carried out a secondary analysis of the databases of the National Household Living Standards Survey (ENNViH - Encuesta Nacional sobre Niveles de Vida de los Hogares) conducted in Mexico. ENNViH is a cohort study following a sample of households with three surveys: 2002 (ENNViH-1), between 2005 and 2006 (ENNViH-2), and between 2009 and 2012 (ENNViH-3). During the follow-ups (2005 and 2009), the participants of the initial round were contacted again. The sample of the ENNViH has national, rural-urban and regional representativeness. This study used data from the second and third surveys; the 2005 information corresponds to our baseline measurement and the 2009 information corresponds to the follow-up. The ENNViH questionnaires were adapted to the Mexican population, and the ethics committee of the National Institute of Public Health approved the study. The informed consent was obtained from all participants¹⁸18 Rubalcava L, Teruel G. Guía del usuario para la Segunda Encuesta Nacional sobre Niveles de Vida de los Hogares. Ciudad de México: UIA/CIDE; 2007..

For the analysis, we included individuals between 20 and 40 years old at the beginning of the study (ENNViH-2, n = 8,106), considering that during this stage of life, stature remains stable. Those who, in baseline measurement (2005), were pregnant or breastfeeding (n = 323); those who did not have systolic or diastolic blood pressure record (n = 312) or whose values were outside the biologically possible range (n = 17); those who did not have a record of stature (n = 19); those whose sex was not specified (n = 201); and those whose follow-up time did not correspond to the time period in which the data were collected (n = 104) were excluded. Follow-up time was obtained from the age difference between the second and first cycle. The shortest follow-up was two years for those surveyed during 2006 after the day and month of their birth, with follow-up during 2009 before the day and month of their birth; while the longer was eight years for those surveyed during 2005 before the day and month of their birth, with follow-up during 2012 after the day and month of their birth. An analytical sample of 7,130 individuals was formed.

During follow-up, 81.1% of the participants included in the initial round were contacted. However, blood pressure measurement was only recorded for 62.7% (n = 4,472). Women who were pregnant or breastfeeding during the second time (n = 202) were not excluded, but only their baseline information was included for analysis, considering them as lost during follow-up. The comparison between those who had complete follow-up information with those who only had information from the first round showed that those who were followed up in general had short stature and higher blood pressure (Supplementary Table 1). Due to the type of analysis, we did not exclude losses in follow-up and those who had missing information, since the results of their first measurement were available.

We measured the blood pressure in the manner established in the guidelines for the detection of arterial hypertension, with the participants seated in a quiet environment after at least five minutes of rest, with the arm at heart level, recording two measurements in each participant²2 Whelton PK, Carey RM, Aronow WS, Casey DJ Jr, Collins KJ, Himmelfarb CD, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13-e115. https://doi.org/10.1161/HYP.0000000000000065
https://doi.org/10.1161/HYP.000000000000... , obtaining the results by means of an OMRON model HEM¹⁹19 Cohen JB, Wong TC, Alpert BS, Townsend RR. Assessing the accuracy of the OMRON HEM-907XL oscillometric blood pressure measurement device in non -dialytic chronic kidney disease patients. J Clin Hypertens (Greenwich). 2017;19(3):296-302. https://doi.org/10.1111/jch.12961
https://doi.org/10.1111/jch.12961... digital baumanometer, recorded in mmHg. We used the average of the two measurements for the analysis. We considered as implausible values those with measurements ≤ 60 or ≥ 300 mmHg for systolic and ≤ 30 or ≥ 150 mmHg for diastolic. We calculated the change in blood pressure by subtracting the values of the first moment from those of the second moment.

Specialized personnel trained for this activity, following standardized procedures²⁰20 Lohman TG, Roche AF, Martorell R, editors. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics Books; 1988., obtained the anthropometric measurements. We measured the stature using a SECA wall stadiometer with a precision of 1.0 mm, with the individual standing barefoot, resting heels, calves, buttocks, back and head on the wall, reporting the results in centimeters and calculating quintiles for each sex.

The covariates considered were the participants’ age, socioeconomic position, as well as habits related to health and an indicator of adiposity, the geographic region and the size of the town where they lived. Age was taken in completed years recorded in the questionnaire and was grouped into five categories: 20.0–24.9 years old, 25.0–29.9 years old, 30.0–34.9 years old, 35.0–39.9 years old and 40.0 years old or more. To measure the socioeconomic position of the participants, we considered family income and parents’ schooling. We obtained the per capita family income from the sum of the income of the family members (business, rent, labor, rural and transfers from another family member or person). We divided the total by the number of household members and reported in Mexican pesos. On the other hand, when data was available, we analyzed the father’s schooling; and when it was not available, we considered the mother’s schooling.

For habits (physical activity, alcohol consumption, smoking and use of antihypertensive drugs), dichotomous variables were generated, distributing the sample as positive or negative according to the responses. We considered physical activity positive for who reported that performs physical activity regularly. In addition, we classified as positive for alcohol consumption those who mentioned they usually drink alcoholic beverages in any of the situations mentioned in the questionnaire. In smoking, likewise, those who reported smoking or having smoked in the past were positive. Those who mentioned being hypertensive and under treatment were classified as positive for the use of antihypertensive drugs. We measured waist circumference with a fiberglass measuring tape 200 cm long and accurate of one decimeter, with the individual standing upright, feet together, abdomen relaxed and arms crossed in front of the chest. The anatomic reference was the iliac crest and the last rib. The measurement was reported in centimeters²⁰20 Lohman TG, Roche AF, Martorell R, editors. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics Books; 1988.. We calculated the waist-height index (WHI) by dividing waist circumference by stature. We considered it as an indicator of adiposity, since it uses to be a good predictor in the evaluation of cardiovascular risk. It was considered as a possible mediator of the relationship between stature and blood pressure, since people of shorter stature tend to have greater adiposity and, in turn, adiposity is one of the strongest predictors of blood pressure²¹21 Deng G, Yin L, Liu W, Liu X, Xiang Q, Qian Z, et al. Associations of anthropometric adiposity indexes with hypertension risk: a systematic review and meta-analysis including PURE-China. Medicine (Baltimore). 2018;97(48):e13262. https://doi.org/10.1097/MD.0000000000013262
https://doi.org/10.1097/MD.0000000000013... .

We divided the regions according to the national development plan 2001–2006²²22 Gobierno de los Estados Unidos Mexicanos, Presidencia de la República. Plan Nacional de Desarrollo 2001-2006. Ciudad de México; 2001. p.157., resulting in four regions: center, midwest, south-southeast and north (including northeast and northwest). We formed four categories according to the size of the locality: > 100,000; 15,001–100,000; 2,500–15,000; and < 2,500 inhabitants.

We used the Stata version 14 program for database cleaning and statistical analysis. We considered the expansion factor at the individual level to obtain weighted relative frequencies of the categorical variables for each moment in each sex. We used ANOVA testing to look for differences according to stature in baseline blood pressure and changes over time.

We used multilevel models²³23 Twisk JWR. Applied longitudinal data, analysis for epidemiology. a practical guide. 2. ed. Cambridge (UK): Cambridge University Press; 2013., considering participant identification as a random intercept, to look for differences in blood pressure (baseline and over time) according to stature. With multilevel models it is recognized that there is intra-subject correlation. The interaction of the cycle with stature allows us to recognize between cross-sectional and longitudinal differences. The intraclass correlation coefficient for systolic and diastolic pressure in men was 0.20 and 0.16; while in women it was 0.25 and 0.18, respectively. The cycle, which identifies to which round each measurement belongs, was included as a time variable. We added the quadratic term of stature and its interaction with cycle in each of the models to evaluate the quadratic effect of stature. We used crude models adjusted for sociodemographic data, WHI and health-related habits (physical activity, alcohol consumption, smoking and use of antihypertensive drugs). The parental schooling variable had a high proportion of missing data (women, 15.8%; men, 18.2%), so models were estimated with and without it.To improve the understanding of the results, marginal means were estimated and plotted from the fitted models^aaSupplementary figure may be requested to the authors.. We performed the analysis stratifying by sex.

RESULTS

Women participants were 56.1%. The mean age was 29.5 years at first, increasing to 35.0 years during follow-up. The average follow-up time was 4.3 years.

In both sexes, half of the participants were between 30 and 39 years old during 2005 (Table 1). For the second time, the majority were over 35 years old. Most of the participants lived in cities and were located mainly in the central region. In both cycles, a higher proportion of the population did not practice physical activity or smoke. The proportion of men who were physically active, smoked or consumed alcohol was higher than the proportion of women. At follow-up, in men, the proportion of smokers increased, while alcohol consumption decreased. In women both, the proportion of smokers and alcohol consumption increased.

Per capitafamily income was higher in men than in women in both cycles, but while in men an increase in average income is present, in women it decreased slightly. The average systolic and diastolic pressure in both years was higher in men than in women. Similarly, the increase in systolic pressure among men was greater than in women, while the increase in diastolic pressure was slightly greater in women. In men, the average stature was greater than in women, by 12.5 cm. In 2009 the average stature in women decreased by 0.8 cm, while in men it decreased by 0.6 cm.

In both sexes, baseline blood pressure was lowest in the shorter stature quintile, while it was highest in the taller stature quintile (Table 2). However, the differences were significant only in men, where blood pressure tended to be lower as the stature quintile was shorter.

In each of the multilevel models (Tables 3 and 4), in both sexes, systolic and diastolic pressures were higher at follow-up compared to baseline measurement, except for systolic pressure in women, where this difference was significant only in the crude model (model 0). In men (Table 3 and supplementary figure), both pressures, at the start of follow-up, tended to be lower as stature became shorter. We observed this trend both in the crude models and in those adjusted for different covariates (models 0 to 3). When adjusted for WHI this relationship became stronger (model 2), both in systolic and diastolic. In evaluating the quadratic effect, we identified the increases in systolic pressure over time tending to be higher as stature increased, in each of the models (models 0 to 3). The effect on diastolic pressure was similar but only when adjusting for WHI and all variables (models 2 and 3).

In women (Table 4 and supplementary figure), baseline pressures were also lower as stature was shorter, especially after adjusting for WHI and habits related to health (models 2 and 3). No differences in changes over time were identified, nor when using quadratic terms (Supplementary table 2).

The results of the models when adjusting for parental schooling (Supplementary tables 3 and 4) did not differ from those obtained without including it (Tables 3 and 4). Therefore, the manuscript reports only the latter.

DISCUSSION

Considering the epidemiological evidence from other populations³3 De Lucia Rolfe E, França GVA, Vianna CA, Gigante DP, Miranda JJ, Yudkin JS, et al. Associations of stunting in early childhood with cardiometabolic risk factors in adulthood. PloS One. 2018;13(4):e0192196. https://doi.org/10.1371/journal.pone.0192196
https://doi.org/10.1371/journal.pone.019... –⁵5 Batty GD, Shipley MJ, Gunnell D, Huxley R, Kivimaki M, Woodward M, et al. Height, wealth, and health: an overview with new data from three longitudinal studies. Econ Hum Biol. 2009;7(2):137-52. https://doi.org/10.1016/j.ehb.2009.06.004
https://doi.org/10.1016/j.ehb.2009.06.00... ,⁸8 Das Gupta R, Haider SS, Hashan MR, Rahman MA, Sarker M. Association between height and hypertension in the adult Nepalese population: findings from a nationally representative survey. Health Sci Rep. 2019;2(12):e141. https://doi.org/10.1002/hsr2.141
https://doi.org/10.1002/hsr2.141... –¹¹11 Langenberg C, Hardy R, Breeze E, Kuh D, Wadsworth MEJ. Influence of short stature on the change in pulse pressure, systolic and diastolic blood pressure from age 36 to 53 years: an analysis using multilevel models. Int J Epidemiol. 2005;34(4):905-13. https://doi.org/10.1093/ije/dyi071
https://doi.org/10.1093/ije/dyi071... and the anatomical and physiological changes described¹³13 Hwaung P, Heo M, Bourgeois B, Kennedy S, Shepherd J, Heymsfield SB. Greater height is associated with a larger carotid lumen diameter. Medicines (Basel). 2019;6(2):57. https://doi.org/10.3390/medicines6020057
https://doi.org/10.3390/medicines6020057... ,¹⁴14 Norman M. Low birth weight and the developing vascular tree: a systematic review. Acta Pediatr. 2008;97(9):1165-72. https://doi.org/10.1111/j.1651-2227.2008.00904.x
https://doi.org/10.1111/j.1651-2227.2008... , an inverse relationship between stature and baseline blood pressure would be expected. However, in Mexican adults, baseline blood pressure tended to be higher as stature was taller in both sexes, with wider differences among men than among women. In the changes in blood pressure, we found differences only in men when evaluating the quadratic effects; contrary to expectations, as stature increased, the increase in blood pressure was higher.

Differences in the genetic composition of the population could help explain why the behavior of blood pressure was not as expected. The ethnic mix of our country resulted from the combination of three ancestries: Amerindian, European and African. European ancestry has been associated with taller stature¹⁷17 Ruiz-Linares A, Adhikari K, Acuña-Alonzo V, Quinto-Sanchez M, Jaramillo C, Arias W, et al. Admixture in Latin America: geographic structure, phenotypic diversity and self-perception of ancestry based on 7,342 individuals. PLoS Genet. 2014;10(9):e1004572 https://doi.org/10.1371/journal.pgen.1004572
https://doi.org/10.1371/journal.pgen.100... as well as with a higher risk of presenting cardiovascular disease²⁴24 Allison MA, Peralta CA, Wassel CL, Aboyans V, Arnett DK, Cushman M, et al. Genetic ancestry and lower extremity peripheral artery disease in the Multi-Ethnic Study of Atherosclerosis. Vasc Med. 2010;15(5):351-9. https://doi.org/10.1177/1358863X10375586
https://doi.org/10.1177/1358863X10375586... compared to Amerindian ancestry. On the other hand, Amerindian ancestry is associated with a greater probability of having short stature²⁵25 Mendez N, Barrera-Pérez TLM, Palma-Solis M, Zavala-Castro J, Dickinson F, Azcorra H, et al. Ethnicity and income impact on BMI and stature of school children living in urban southern Mexico. J Biosoc Sci. 2016;48(2):143-57. https://doi.org/10.1017/S0021932015000127
https://doi.org/10.1017/S002193201500012... . In other words, it is possible that the Euro-descendant population in Mexico is the one that is taller and has higher blood pressure than the rest of the population, and this is what explains the positive relationship between these two variables observed in our analysis. Unfortunately, the ENNViH did not measured markers of ancestry, so it is not possible to evaluate this possibility. On the other hand, the relationship between blood pressure and stature may not be linear, so that in populations of smaller stature compared to European countries, it may be contrary to that reported in populations of taller stature.

In addition to genetics, diet quality could be another element that helps to explain why taller stature participants had higher blood pressure. In Mexico, the availability and consumption of ultra-processed foods differ according to geographic region and socioeconomic level (similar to what occurs with stature), with higher consumption in the north and by people with a higher socioeconomic level²⁶26 Romo-Avilés M, Ortiz-Hernández L. Contribution of NOVA food groups to energy and nutrient supply in Mexican households. Salud Publica Mex. 2019;61(2):155-65. https://doi.org/10.21149/8923
https://doi.org/10.21149/8923... . Ultra-processed foods have a higher sodium content compared to those of natural origin²⁶26 Romo-Avilés M, Ortiz-Hernández L. Contribution of NOVA food groups to energy and nutrient supply in Mexican households. Salud Publica Mex. 2019;61(2):155-65. https://doi.org/10.21149/8923
https://doi.org/10.21149/8923... ,²⁷27 Mendonça RD, Lopes ACS, Pimenta AM, Gea A, Martínez-González MA, Bes-Rastrollo M. Ultra-processed food consumption and the incidence of hypertension in a Mediterranean cohort: The Seguimiento Universidad de Navarra Project. Am J Hypertens. 2017;30(4):358-66. https://doi.org/10.1093/ajh/hpw137
https://doi.org/10.1093/ajh/hpw137... , which explains why their consumption is associated with elevated blood pressure²⁷27 Mendonça RD, Lopes ACS, Pimenta AM, Gea A, Martínez-González MA, Bes-Rastrollo M. Ultra-processed food consumption and the incidence of hypertension in a Mediterranean cohort: The Seguimiento Universidad de Navarra Project. Am J Hypertens. 2017;30(4):358-66. https://doi.org/10.1093/ajh/hpw137
https://doi.org/10.1093/ajh/hpw137... . Therefore, it is possible that people with taller stature tend to have a higher socioeconomic level²⁸28 Castro-Porras LV, Rojas-Russell ME, Aedo-Santos A, Wynne-Bannister EG, López-Cervantes M. Stature in adults as an indicator of socioeconomic inequalities in Mexico. Rev Panam Salud Publica. 2018;42:e29. https://doi.org/10.26633/RPSP.2018.29
https://doi.org/10.26633/RPSP.2018.29... and thus may have a higher consumption of ultra-processed foods, favoring higher blood pressure as stature increases. We did not explore this possibility, since ENNViH has no information related to diet.

We confirmed the participation of WHC as a mediator in the relationship between stature and blood pressure. WHC was inversely related to stature; those with shorter stature had higher WHC, as reported in a study conducted in southern Mexico²⁵25 Mendez N, Barrera-Pérez TLM, Palma-Solis M, Zavala-Castro J, Dickinson F, Azcorra H, et al. Ethnicity and income impact on BMI and stature of school children living in urban southern Mexico. J Biosoc Sci. 2016;48(2):143-57. https://doi.org/10.1017/S0021932015000127
https://doi.org/10.1017/S002193201500012... . In the models where it was not included, the differences according to stature were smaller but, adding to the models, the relationship of stature with baseline systolic and diastolic pressure was stronger²¹21 Deng G, Yin L, Liu W, Liu X, Xiang Q, Qian Z, et al. Associations of anthropometric adiposity indexes with hypertension risk: a systematic review and meta-analysis including PURE-China. Medicine (Baltimore). 2018;97(48):e13262. https://doi.org/10.1097/MD.0000000000013262
https://doi.org/10.1097/MD.0000000000013... .

Among the limitations of the study, it is important to point out that ENNViH was not designed to test our hypothesis, so it was necessary to adapt the available information. In addition, with the exception of anthropometric measurements, we obtained the information using questionnaires. The average follow-up time of this study was four years, while several of the studies conducted have a follow-up time of more than 15 years⁵5 Batty GD, Shipley MJ, Gunnell D, Huxley R, Kivimaki M, Woodward M, et al. Height, wealth, and health: an overview with new data from three longitudinal studies. Econ Hum Biol. 2009;7(2):137-52. https://doi.org/10.1016/j.ehb.2009.06.004
https://doi.org/10.1016/j.ehb.2009.06.00... ,⁶6 Horta BL, Barros FC, Victora CG, Cole TJ. Early and late growth and blood pressure in adolescence. J Epidemiol Community Health. 2003;57(3):226-30. https://doi.org/10.1136/jech.57.3.226
https://doi.org/10.1136/jech.57.3.226... ,¹¹11 Langenberg C, Hardy R, Breeze E, Kuh D, Wadsworth MEJ. Influence of short stature on the change in pulse pressure, systolic and diastolic blood pressure from age 36 to 53 years: an analysis using multilevel models. Int J Epidemiol. 2005;34(4):905-13. https://doi.org/10.1093/ije/dyi071
https://doi.org/10.1093/ije/dyi071... . Finally, blood pressure changes over time may be significant at another stage of life. That is, differences in blood pressure changes were probably not identified because the follow-up period was at a time when blood pressure rises uniformly, regardless of stature. It is possible to identify the differences at earlier stages⁶6 Horta BL, Barros FC, Victora CG, Cole TJ. Early and late growth and blood pressure in adolescence. J Epidemiol Community Health. 2003;57(3):226-30. https://doi.org/10.1136/jech.57.3.226
https://doi.org/10.1136/jech.57.3.226... or at later ages than those included in the investigation⁴¹.

Despite the aforementioned limitations, this study provided an overview of the association between stature and blood pressure in the Mexican population. We measured the blood pressure according to the recommendations of the guidelines for the detection of hypertension²2 Whelton PK, Carey RM, Aronow WS, Casey DJ Jr, Collins KJ, Himmelfarb CD, et al. 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018;71(6):e13-e115. https://doi.org/10.1161/HYP.0000000000000065
https://doi.org/10.1161/HYP.000000000000... . Although it may be useful to have more than two measurements²⁹29 Muntner P, Shimbo D, Carey RM, Charleston JB, Gaillard T, Misra S, et al. Measurement of blood pressure in humans: a scientific statement from the American Heart Association. Hypertension. 2019;73(5):e35-e66. https://doi.org/10.1161/HYP.0000000000000087
https://doi.org/10.1161/HYP.000000000000... , the increase in the number of measurements has been of minimal benefit in the detection of cardiovascular risk³⁰30 Paige E, Barrett J, Pennells L, Sweeting M, Willeit P, Di Angelantonio E, et al. Use of repeated blood pressure and cholesterol measurements to improve cardiovascular disease risk prediction: an individual-participant-data meta-analysis. Am J Epidemiol. 2017;186(8):899-907. https://doi.org/10.1093/aje/kwx149
https://doi.org/10.1093/aje/kwx149... . In addition, we analyzed data from a nationally representative study with a longitudinal design. The use of multilevel models made it possible to distinguish between cross-sectional differences in changes over time. Therefore, the relationships found are causally stronger than those that could have been found in a cross-sectional study.

In summary, in the Mexican population, the effect of stature on blood pressure was contrary to the expected, since it was positively related. In the context of the epidemiological transition, the predominance of chronic undernutrition in Latin American populations could be a factor that increases the burden associated with chronic diseases. Our analysis shows that this may not be entirely true. In contrast to the above, in Mexican adults, blood pressure is related in the expected direction to traditional risk factors such as adiposity. Therefore, in order to address the problem of arterial hypertension and its consequences in the country, it is necessary to continue investigating other environmental, genetic and dietary factors, as well as to address the modifiable factors already known. It would be advisable to carry out studies with a similar design, but at different stages of life, to evaluate the blood pressure and identify the times when it shows a greater change.

REFERENCES

Publication Dates

History