Effect of prevalence of alcohol consumption and tobacco use in Mexican municipalities on early childhood development Authors: Francisco-Javier Prado-Galbarro, Carlos Sanchez-Piedra, Juan-Manuel Martínez-Núñez Published: 2023 ABSTRACT Abstract:One of the most critical time periods in childhood is from birth to five years of age. Children exposed to alcohol and/or tobacco via family members and neighborhood are at risk for childhood developmental delays. This study evaluated the association of early childhood development with the prevalence of alcohol consumption and tobacco use in Mexican municipalities. This is a cross-sectional study. Early childhood development information from 2,345 children aged from 36 to 59 months was obtained from the 2015 Mexican National Survey of Boys, Girls, and Women (ENIM). Data on alcohol consumption and tobacco use come from the 2016 Mexican National Survey on Drugs, Alcohol, and Tobacco Consumption (ENCODAT). Multilevel logistic models were fitted to evaluate the association of the prevalence of alcohol consumption and tobacco use with the inadequacy of early childhood development. Children living in municipalities with high prevalence of alcohol consumption (OR = 13.410; 95%CI: 2.986; 60.240) and tobacco use (OR = 15.080; 95%CI: 2.040; 111.400) were less likely to be developmentally on track regarding early childhood development after adjustment for individual variables related to the child’s development and other environmental variables at municipal level. Childhood exposure to alcohol and tobacco in the neighborhood may directly contribute to inadequate early childhood development. These findings suggest that there is an urgent need to develop effective interventions aimed at reducing alcohol consumption and tobacco use in municipalities to ensure adequate early childhood development. Introduction Child development is a public health priority worldwide, included as one of the United Nations Sustainable Development Goals 1 . The period from birth to five years of age is considered the most important developmental phase of humans and it is a key stage to improve individual’s life trajectories 2 , 3 , 4 . An adequate early childhood development strongly influences well-being throughout life 3 . It is estimated that more than 200 million children under five years of age from low- and middle-income countries are at risk of not achieving their full developmental potential due to several lacks and barriers related to the environments responsible for fostering child development, such as family, school, and civil society in a broader context 5 . Generally, research regarding child development has focused on the proximal determinants, mainly describing the influence of individual, maternal, familiar, and school predictors on child’s growth and development. There is evidence regarding the role of maternal nutrition, breastfeeding, parenting skills, parent’s education, cultural practices, and intra-family relations on child development 3 , 6 . There is also ample evidence of associations between socioeconomic conditions of the family and child development 7 , 8 , 9 . For instance, maternal educational attainment is strongly associated with better child development: children of mothers with low educational level may have a slower physical and emotional development when compared with children of mothers with complete high-school or higher education 10 , 11 . It has been proposed that mothers and families need supportive environments to be able to provide opportunities that allow their children to fulfill their potentialities. The neighborhood environment has been identified as key for supporting families to overcome adversities that could constrain child development 3 , 12 , 13 . Neighborhood residential instability in childhood can contribute to the development of major depressive disorder and antisocial personality disorder 14 . As in other countries, alcohol consumption and tobacco use are two of the major public health problems in Mexico, especially in reproductive age people. The consumption of alcohol and tobacco use presents repercussions on the consumer’s family and social environment, including the development of children in early childhood 15 . The family is considered the central axis in the early childhood development process since it is the closest and most immediate environment to the child. The influence family exerts throughout childhood can last a lifetime 16 . Several studies have analyzed the effect of alcohol consumption and tobacco use during pregnancy on children’s behavioral disorders 17 , 18 . However, these studies have not considered exposure to alcohol and tobacco in neighborhoods as an important risk factor for children’s development. In this study, we explored the association between early childhood development and the prevalence of alcohol and tobacco use in Mexican municipalities, adjusting individual and family-level determinants. The main hypothesis was that children living in municipalities with higher prevalence of alcohol and tobacco use, along with those with higher levels of crime and marginalization, could present more developmental issues. Methods Study design, participants, and setting This is a retrospective study with information extracted from the 2016 Mexican National Survey on Drugs, Alcohol, and Tobacco Consumption (ENCODAT - Encuesta Nacional de Consumo de Drogas) and the 2015 Mexican National Survey of Boys, Girls, and Women (ENIM - Encuesta Nacional de los Niños, Niñas y Mujeres). These surveys used probabilistic multistage stratified cluster sampling design and are representative at the national and regional levels and for urban/rural areas. Furthermore, ENCODAT is also representative at the state level. ENIM 2015 was part of the fifth round of the Multiple Indicator Cluster Survey (MICS 5), and oversampled children under age five and rural areas to estimate indicators for children and women 19 . Four questionnaires were applied to get information on: (a) the household, (b) women aged 15 to 49, (c) children and adolescents aged 5 to 17 years (answered by their mothers), and (d) children under age five (answered by their mothers). The municipality of residence of children who participated in the ENIM 2015 was linked to its prevalence of alcohol consumption and tobacco use, along with other secondary sources, in the same area through unique municipality codes. A total of 2,345 children belonging to 200 municipalities were selected. Outcome variable: early childhood development The primary source of early childhood development data was the Early Childhood Development Index (ECDI), which is a population-based measure developed and validated by the MICS program from the United Nation’s Children Fund (UNICEF) during 2006-2009. The ECDI was conducted in around 80 low- and middle-income countries, being considered a useful tool for identifying inadequate development in children aged 36-59 months. The ECDI represents the percentage of children aged 36-59 months who were developmentally on track (adequate early childhood development) in at least three out of the following four domains: literacy-numeracy, physical development, social-emotional development, and learning. The ECDI is composed of 10 items, each with yes/no responses, covering all domains 20 . Based on ECDI criteria, literacy-numeracy development is considered on track if children can do at least two of the following: identify 10 or more letters of the alphabet; read four or more common simple words; and/or name and recognize all the numbers from 1 to 10. The social-emotional development is considered on track if children can do at least two of the following: get along well with other children; do not get distracted easily, and/or do not hit, bite, or kick other children. The physical development is considered on track if at least one of the following is true about the children: can pick up small objects (like a stick or rock from the ground) with two fingers and/or does not often feel sick to play. Learning development is considered to be on track if children can do at least one of the following: can follow simple instructions on how to do something correctly and/or when given something to do, is able to do it independently. Therefore, each domain was considered as a binary outcome, and children aged 36-59 months were considered to be developmentally on track if they met the age-expected development targets in at least three out of the four domains. Exposure variables: alcohol consumption and tobacco use The prevalence of alcohol consumption and tobacco use by municipality was estimated by implementing small area estimation with ENCODAT data. Auxiliary information for small areas was aggregated to increase survey sampling precision. Area-level models described by Fay-Herriot were used 21 , which is a widely used technique to obtain reliable estimates for small areas. Then, estimated averages from the outcomes of interest were combined with secondary data for each municipality, such as crime rate 22 , social backwardness index 23 , and population density 24 . Covariates (i) Child-level covariates: the analysis incorporated the following information on children: sex, age in months, presence of any functional difficulty (seeing, hearing, walking, fine motor coordination, understanding, being understood, learning things, playing, and, where applicable, of controlling behavior), maternal age, maternal education, indigenous origin, and households’ wealth quintiles. (ii) Municipality-level covariates: population density 24 and number of reported homicides 25 , obtained from the Mexican Institute of Geography and Statistics (INEGI). It uses data from death certificates, which tends to be more precise than Mexican police records. Statistical analysis Individual and municipal characteristics by early childhood development status were analyzed using percentages and 95% confidence intervals (95%CI) for categorical variables, along with means and standard deviations (SD) for continuous variables. Moreover, bivariate analyses were performed to identify the association between individual- and municipal-level variables and early childhood development. Then, to test the association between the proportion of alcohol consumption and early childhood development, as well as the proportion of daily tobacco smokers and early childhood development, multilevel multivariable logistic regression models were used with data from individuals nested within municipalities. A series of models were fitted: two models including each exposure separately, and a final model including all exposures together. Then, these models were adjusted for each child- or municipal-level variable with a p-value ≤ 0.20 in the bivariate analysis. The following multivariate regression model was used: yij∼Binomial1,πij ogπij1-πij=β0+∑k=1nβkXki+∑l=1mβlZlj+uj Where π ij is the probability of inadequate early childhood development for child i in municipality j, X i is the set of explanatory variables at the child-level (sex, age in months, functional difficulties, maternal age, maternal education, household wealth quintiles, indigenous origin, and domestic violence), Z j is the set of explanatory variables defined for the municipalities (proportion of alcohol consumption/proportion of daily tobacco smokers, number of homicides and population density), and u j are the residuals of level 2, for which it is assumed that they are independent and follow a normal distribution with mean 0 and a variance of σu2 . The median odds ratio (OR) were computed to quantify the effect of area-level variance, which can be interpreted as the increased risk (in median) of having inadequate early childhood development if the child moves from the municipality at lowest risk to the municipality at highest risk 26 . The median OR is equal to 1 when there is no variation between municipalities. However, if the median OR is greater than 1, then between-municipality variation is significant. All analyses were performed in Stata, version 16.0 (https://www.stata.com). Estimates of descriptive indicators account for the complex survey design and survey weights. In statistical models, weighted multilevel logistic models were estimated by child-level weights at level 1 (pweight). Statistical tests were two-tailed and considered significant below a 0.05 alpha. Ethics approval and informed consent During the data collection of the ENIM and ENCODAT, the participants consented to voluntary and anonymously participate by signing a consent letter. Both studies were conducted in accordance with the Declaration of Helsinki. All analyses in this study were performed with de-identified secondary data. Since it was a secondary data analysis, this study did not require the approval of a Research Ethics Committee. Results A total of 2,345 children aged 36-59 months were analyzed. The study sample has a mean age of 48.71 months (SD: 7.08), 56.41% of the sample was female, 1.3% had functioning difficulties, and 39.6% lived in a poor or very poor household. At baseline, 41.61% of the mothers had complete middle school with a mean age of 29.63 years (SD: 6.59), and 23.27% were indigenous (Table 1). Table 1Individual characteristics of children with inadequate and adequate early childhood development.CharacteristicsEarly childhood development [% (95%CI)] p-valueTotal [% (95%CI)]InadequateAdequate Total [n (%)]484 (17.44)1,861 (82.56) 2,345 (100.00)Functional difficulties No17.14 (14.07; 20.70)82.86 (79.30; 85.93)< 0.00198.70 (98.06; 99.13)Yes40.22 (26.07; 56.20)59.78 (43.80; 73.93) 1.30 (0.87; 1.94)Maternal education Elementary or less21.72 (16.9; 27.46)78.28 (72.54; 83.10)0.01216.44 (13.49; 19.89)Middle school21.23 (17.3; 25.78)78.71 (73.12; 83.40) 41.61 (35.29; 48.22)High school16.49 (13.5; 19.98)84.78 (81.02; 87.90) 22.59 (18.64; 27.11)Higher education8.30 (4.14; 15.97)91.91 (82.58; 96.46) 19.35 (10.28; 25.94)Household wealth quintiles Very poor22.22 (16.84; 28.72)77.78 (71.28; 83.16)0.19715.30 (12.49; 19.38)Poor18.41 (14.73; 22.76)81.59 (77.24; 85.27) 24.30 (19.92; 29.30)Middle19.24 (11.89; 29.59)80.76 (70.41; 88.11) 22.66 (18.30; 27.69)Wealthy17.61 (12.37; 24.46)82.39 (75.54; 87.63) 19.49 (15.63; 24.03)Very wealthy9.48 (4.31; 19.59)90.52 (80.41; 95.69) 17.93 (9.65; 30.88)Sex Boys22.75 (19.86; 25.93)77.25 (77.25; 77.25)0.00543.59 (37.18; 50.21)Girls13.33 (9.20; 18.93)86.67 (81.07; 90.80) 56.41 (49.79; 62.82)Indigenous origin17.50 (13.71; 22.07)82.50 (77.93; 86.29) No17.23 (13.70; 21.44)82.77 (78.56; 86.30)0.92876.73 (71.93; 80.93)Yes17.50 (13.71; 22.07)82.50 (77.93; 86.29) 23.27 (19.07; 28.07)Area Urban16.92 (13.39; 21.15)83.08 (78.85; 86.61)0.32181.82 (75.86; 86.57)Rural19.76 (15.99; 24.18)80.24 (75.82; 84.01) 18.18 (13.43; 24.14)Domestic violence No14.64 (11.27; 18.82)85.36 (81.18; 88.73)0.20632.69 (27.59; 38.24)Yes18.79 (14.37; 24.19)81.21 (75.81; 85.63) 67.31 (61.76; 72.41)Maternal age (years) [mean (SD)]28.27 (7.28)29.93 (6.40)0.01129.64 (6.59)Children age (months) [mean (SD)]45.71 (7.85)49.35 (6.75)< 0.00148.71 (7.08)95%CI: 95% confidence interval; SD: standard deviation. Table 1Individual characteristics of children with inadequate and adequate early childhood development.CharacteristicsEarly childhood development [% (95%CI)] p-valueTotal [% (95%CI)]InadequateAdequate Total [n (%)]484 (17.44)1,861 (82.56) 2,345 (100.00)Functional difficulties No17.14 (14.07; 20.70)82.86 (79.30; 85.93)< 0.00198.70 (98.06; 99.13)Yes40.22 (26.07; 56.20)59.78 (43.80; 73.93) 1.30 (0.87; 1.94)Maternal education Elementary or less21.72 (16.9; 27.46)78.28 (72.54; 83.10)0.01216.44 (13.49; 19.89)Middle school21.23 (17.3; 25.78)78.71 (73.12; 83.40) 41.61 (35.29; 48.22)High school16.49 (13.5; 19.98)84.78 (81.02; 87.90) 22.59 (18.64; 27.11)Higher education8.30 (4.14; 15.97)91.91 (82.58; 96.46) 19.35 (10.28; 25.94)Household wealth quintiles Very poor22.22 (16.84; 28.72)77.78 (71.28; 83.16)0.19715.30 (12.49; 19.38)Poor18.41 (14.73; 22.76)81.59 (77.24; 85.27) 24.30 (19.92; 29.30)Middle19.24 (11.89; 29.59)80.76 (70.41; 88.11) 22.66 (18.30; 27.69)Wealthy17.61 (12.37; 24.46)82.39 (75.54; 87.63) 19.49 (15.63; 24.03)Very wealthy9.48 (4.31; 19.59)90.52 (80.41; 95.69) 17.93 (9.65; 30.88)Sex Boys22.75 (19.86; 25.93)77.25 (77.25; 77.25)0.00543.59 (37.18; 50.21)Girls13.33 (9.20; 18.93)86.67 (81.07; 90.80) 56.41 (49.79; 62.82)Indigenous origin17.50 (13.71; 22.07)82.50 (77.93; 86.29) No17.23 (13.70; 21.44)82.77 (78.56; 86.30)0.92876.73 (71.93; 80.93)Yes17.50 (13.71; 22.07)82.50 (77.93; 86.29) 23.27 (19.07; 28.07)Area Urban16.92 (13.39; 21.15)83.08 (78.85; 86.61)0.32181.82 (75.86; 86.57)Rural19.76 (15.99; 24.18)80.24 (75.82; 84.01) 18.18 (13.43; 24.14)Domestic violence No14.64 (11.27; 18.82)85.36 (81.18; 88.73)0.20632.69 (27.59; 38.24)Yes18.79 (14.37; 24.19)81.21 (75.81; 85.63) 67.31 (61.76; 72.41)Maternal age (years) [mean (SD)]28.27 (7.28)29.93 (6.40)0.01129.64 (6.59)Children age (months) [mean (SD)]45.71 (7.85)49.35 (6.75)< 0.00148.71 (7.08)95%CI: 95% confidence interval; SD: standard deviation. Individual characteristics of children with inadequate and adequate early childhood development. 95%CI: 95% confidence interval; SD: standard deviation. Table 1 shows the characteristics of the children with adequate and inadequate early childhood development. A total of 17.78% of children presented inadequate early childhood development, with boys showing a higher percentage of inadequate early childhood development than girls (22.75% and 13.33%, respectively). Overall, children and mothers in the inadequate early childhood development group tended to be younger than in the adequate group. A higher proportion of children with functional difficulties and disabilities had inadequate early childhood development than those without these problems (40.22% and 17.14%, respectively). The proportion of children with inadequate early childhood development was significantly higher for mothers with primary education or less (21.72%) and lower for those with complete tertiary education (8.3%) than for those with complete secondary education (21.23% and 16.49%, respectively). Children who lived in the poorest households (22.22%) and in rural areas (19.76%) had a higher prevalence of inadequate early childhood development than those from the wealthiest households (9.48%) and urban areas (16.92%). Finally, a higher prevalence of inadequate early childhood development was observed in children who were exposed to domestic violence than in those who did not (18.79% vs. 14.64%). Table 2 shows the distribution of the environmental characteristics of the municipalities. Children with inadequate early childhood development lived in municipalities with a greater prevalence of alcohol consumption and a higher proportion of daily tobacco smokers than children with adequate early childhood development (alcohol: 51.96% vs. 50.53%; and daily smokers: 7.84% vs. 7.4%, respectively). Children with inadequate early childhood development lived in municipalities with lower population density than those with adequate early childhood development (an average of 1,908.12 vs. 2,789.69 inhabitants per km2). Other variables, although not statistically significant, were associated with early childhood development, including the marginalization index and the number of homicides. Table 2Characteristics for the municipalities of children with inadequate and adequate early childhood development. Early childhood development [mean (SD)] p-valueTotal [mean (SD)]InadequateAdequate Alcohol consumption (%)51.96 (10.35)50.53 (8.89)0.15850.77 (9.16)Daily tobacco smokers (%)7.84 (7.04)7.40 (6.07)0.4547.48 (6.24)Marginalization index-1.19 (0.69)-1.22 (0.68)0.582-1.22 (0.68)Number of homicides122.4 (190.44)105.17 (148.28)0.165108.17 (155.79)Population density by municipality (inhabitants per km2)1,908.12 (3,847.94)2,789.69 (4,444.87)0.0072,635.98 (4,389.02)SD: standard deviation. Table 2Characteristics for the municipalities of children with inadequate and adequate early childhood development. Early childhood development [mean (SD)] p-valueTotal [mean (SD)]InadequateAdequate Alcohol consumption (%)51.96 (10.35)50.53 (8.89)0.15850.77 (9.16)Daily tobacco smokers (%)7.84 (7.04)7.40 (6.07)0.4547.48 (6.24)Marginalization index-1.19 (0.69)-1.22 (0.68)0.582-1.22 (0.68)Number of homicides122.4 (190.44)105.17 (148.28)0.165108.17 (155.79)Population density by municipality (inhabitants per km2)1,908.12 (3,847.94)2,789.69 (4,444.87)0.0072,635.98 (4,389.02)SD: standard deviation. Characteristics for the municipalities of children with inadequate and adequate early childhood development. SD: standard deviation. Table 3 shows the results of the multilevel model to estimate the association between early childhood development and alcohol consumption and tobacco use, adjusted by child- and municipality-level control variables. In the Model 1, the increase in one unit of proportion of daily tobacco smokers was associated with a 15.08% increase in the odds of inadequate early childhood development (OR = 15.080; 95%CI: 2.040; 111.400). A one figure SD increase in homicides was associated with a 17.9% increase in the odds of inadequate early childhood development (OR = 1.179; 95%CI: 1.051; 1.312), whereas a one-unit increase in population density was associated with 5.3% lower odds of having an inadequate early childhood development (OR = 0.947; 95%CI: 0.906; 0.989). In the Model 2, the increase in one unit of proportion of alcohol consumption was associated with an increase in inadequate early childhood development (OR = 13.410; 95%CI: 2.986; 60.240). A one SD increase in homicides was associated with a 19.6% increase in the odds of inadequate early childhood development (OR = 1.196; 95%CI: 1.053; 1.358). Population density was positively associated with adequate early childhood development since a one-unit increase in population density was associated with 5.4% lower odds of having an inadequate early childhood development (OR = 0.946; 95%CI: 0.904; 0.991). Finally, in Model 3, the increase in one unit of proportion of alcohol consumption was associated with an increase in inadequate early childhood development (OR = 9.415; 95%CI: 1.856; 47.770). A one SD increase in homicides was associated with a 19.4% increase in the odds of inadequate early childhood development (OR = 1.194; 95%CI: 1.057; 1.350). Similar to the previous model, the population density was also positively associated with adequate early childhood development; a one-unit increase in population density was associated with 5.4% lower odds of having an inadequate early childhood development (OR = 0.939; 95%CI: 0.897; 0.984). Table 3Adjusted associations between overall inadequate early childhood development and alcohol and tobacco consumption.PredictorModel 1 * Model 2 * Model 3 * OR95%CIOR95%CIOR95%CIProportion of daily tobacco smokers15.080 **2.040; 111.400 5.5270.476; 64.220Proportion of alcohol consumption 13.410 **2.986; 60.2409.415 **1.856; 47.770Number of homicides ***1.179 **1.051; 1.3231.196 **1.053; 1.3581.194 **1.057; 1.350Population density by municipality # 0.947 **0.906; 0.9890.946 **0.904; 0.9910.939 **0.897; 0.984Functional difficulties No1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Yes1.013 **1.005; 1.0201.012 **1.004; 1.0201.012 **1.004; 1.020Maternal education Elementary or less1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Middle school1.0020.648; 1.5501.0020.650; 1.5460.9950.645; 1.534High school0.628 **0.397; 0.9920.617 **0.390; 0.9750.620 **0.392; 0.981Higher education0.443 **0.223; 0.8800.430 **0.217; 0.8540.434 **0.219; 0.861Household wealth quintiles Very poor1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Poor0.6470.412; 1.0150.614 **0.390; 0.9670.612 **0.390; 0.962Middle0.6540.385; 1.1120.6130.364; 1.0320.6110.364; 1.026Wealthy0.8100.432; 1.5170.7780.413; 1.4650.7660.408; 1.440Very wealthy0.6960.359; 1.3490.6530.333; 1.2790.6410.327; 1.255Maternal age0.9760.948; 1.0040.9760.948; 1.0050.9760.948; 1.005Sex Male1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Female0.547 **0.384; 0.7790.546 **0.384; 0.7770.546 **0.384; 0.777Children age0.927 **0.904; 0.9510.928 **0.905; 0.9520.928 **0.904; 0.952Domestic violence No1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Yes1.493 **1.030; 2.1651.473 **1.017; 2.1321.481 **1.023; 2.145Median OR Null model2.0821.598; 2.5672.0821.598; 2.5672.0821.598; 2.567Adjusted model1.8291.488; 2.1701.8771.521; 2.2331.8201.482; 2.15995%CI: 95% confidence interval; OR: odds ratio.* All models were adjusted by the set of individual variables for each child: sex, age in months, functional difficulties, maternal age, maternal education, household wealth quintiles, and domestic violence;** Significant value (p < 0.05);*** Re-scaled using z-scores, so that a one-unit change represents a one standard deviation change in municipal services of municipalities; # Per 1,000 inhabitants. Table 3Adjusted associations between overall inadequate early childhood development and alcohol and tobacco consumption.PredictorModel 1 * Model 2 * Model 3 * OR95%CIOR95%CIOR95%CIProportion of daily tobacco smokers15.080 **2.040; 111.400 5.5270.476; 64.220Proportion of alcohol consumption 13.410 **2.986; 60.2409.415 **1.856; 47.770Number of homicides ***1.179 **1.051; 1.3231.196 **1.053; 1.3581.194 **1.057; 1.350Population density by municipality # 0.947 **0.906; 0.9890.946 **0.904; 0.9910.939 **0.897; 0.984Functional difficulties No1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Yes1.013 **1.005; 1.0201.012 **1.004; 1.0201.012 **1.004; 1.020Maternal education Elementary or less1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Middle school1.0020.648; 1.5501.0020.650; 1.5460.9950.645; 1.534High school0.628 **0.397; 0.9920.617 **0.390; 0.9750.620 **0.392; 0.981Higher education0.443 **0.223; 0.8800.430 **0.217; 0.8540.434 **0.219; 0.861Household wealth quintiles Very poor1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Poor0.6470.412; 1.0150.614 **0.390; 0.9670.612 **0.390; 0.962Middle0.6540.385; 1.1120.6130.364; 1.0320.6110.364; 1.026Wealthy0.8100.432; 1.5170.7780.413; 1.4650.7660.408; 1.440Very wealthy0.6960.359; 1.3490.6530.333; 1.2790.6410.327; 1.255Maternal age0.9760.948; 1.0040.9760.948; 1.0050.9760.948; 1.005Sex Male1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Female0.547 **0.384; 0.7790.546 **0.384; 0.7770.546 **0.384; 0.777Children age0.927 **0.904; 0.9510.928 **0.905; 0.9520.928 **0.904; 0.952Domestic violence No1.000 (Reference) 1.000 (Reference) 1.000 (Reference) Yes1.493 **1.030; 2.1651.473 **1.017; 2.1321.481 **1.023; 2.145Median OR Null model2.0821.598; 2.5672.0821.598; 2.5672.0821.598; 2.567Adjusted model1.8291.488; 2.1701.8771.521; 2.2331.8201.482; 2.15995%CI: 95% confidence interval; OR: odds ratio.* All models were adjusted by the set of individual variables for each child: sex, age in months, functional difficulties, maternal age, maternal education, household wealth quintiles, and domestic violence;** Significant value (p < 0.05);*** Re-scaled using z-scores, so that a one-unit change represents a one standard deviation change in municipal services of municipalities; # Per 1,000 inhabitants. Adjusted associations between overall inadequate early childhood development and alcohol and tobacco consumption. 95%CI: 95% confidence interval; OR: odds ratio. * All models were adjusted by the set of individual variables for each child: sex, age in months, functional difficulties, maternal age, maternal education, household wealth quintiles, and domestic violence; ** Significant value (p < 0.05); *** Re-scaled using z-scores, so that a one-unit change represents a one standard deviation change in municipal services of municipalities; # Per 1,000 inhabitants. In addition, in all three models, several child-level control variables were significant. Children of mothers with complete secondary or tertiary education showed lower odds of presenting inadequate early childhood development than those who mothers had primary education or less; children with functional difficulties and/or children exposed to domestic violence in their household had greater odds of presenting inadequate early childhood development; girls had a lower risk of inadequate early childhood development than boys; and a one-unit (one-month) increase in children’s age was associated with an 8% decrease in the odds of inadequate early childhood development. In Models 2 and 3, children living in poor households had lower odds of inadequate early childhood development than those in very poor households. Table 3 also show the values of the median OR both from the null model and from the adjusted models for child- and municipality-level variables. In the null model, the median OR was equal to 2.082 (95%CI: 1.598; 2.567), i.e., in the median case, the residual heterogeneity between municipalities increased by 2.082 times. In the adjusted models, the median OR decreased (ORModel 1 = 1.829; 95%CI: 1.488; 2.170; ORModel 2 = 1.877; 95%CI: 1.521; 2.233; and ORModel 3 = 1.820; 95%CI: 1.482; 2.159). Overall, the models fit indices indicated that the model fits the data well (p < 0.001). Discussion This study investigated the association of the prevalence of alcohol consumption or tobacco use in Mexican municipalities with early childhood development for children under five years. Our findings revealed greater risks of inadequate early childhood development in children living in areas with higher prevalence of use of these substances. We found that greater alcohol consumption or tobacco use in Mexican municipalities was associated with higher odds of inadequate early childhood development. These results are consistent with a growing body of evidence suggesting that these factors are essential to understand developmental delays in children. Children of alcoholic parents have shown behavioral problems 27 , 28 , which has been associated with increased cognitive and behavioral impairment in two recent meta-analyses. Likewise, a recent study from Brazil reported that the concomitant use of alcohol and tobacco increased the risk of motor and cognitive delays in children 29 . We found that living in a municipality with a greater number of homicides was associated with greater odds of inadequate early childhood development. Some studies suggest that neighborhood violence is an important risk factor for child development 30 , 31 . Finkelhor et al. 32 demonstrated that exposure to several types of violence could influence children’s physical, social-emotional and cognitive development. Another analysis from the United States suggests that tobacco shops and off-sale alcohol outlets were proximal to violent crime activity, controlling for poverty, renters, resident mobility, and ethnic/racial heterogeneity 33 . Our results also showed that living in a municipality with a high population density was associated with decreased odds of inadequate early childhood development. Several studies found that children are healthier in urban areas that in rural areas 34 . More densely populated places are more likely to have easier access to relevant health and education services for child survival and development. Furthermore, people in urban areas tend to be richer and better educated 35 , 36 , 37 . However, these results should be interpreted carefully, considering that population density is a proxy for access to services and economic opportunities. Within the individual level, age and female sex were associated with inadequate early childhood development. Children evaluated in this study showed that, as they grew, they had better results in the development of their abilities and skills, which corresponds to the continuous process of maturation where the development of various cognitive, communication, motor, and social-emotional skills is progressively built 5 . Several studies examined gender differences, with boys usually presenting lower scores in certain areas of development 38 , 39 . Mother’s education and household socioeconomic status have shown important influence on child development. Several studies found that mothers with a higher level of education have more intellectual and emotional tools to care for their children 40 , 41 . We also found that Mexican children with functional difficulties are more likely to have inadequate early childhood development. Children with disabilities often face barriers that affect their ability to participate fully and effectively in society, such as poverty, stigma, discrimination, violence, and limited access to programs 41 . Finally, children exposed to violent discipline had higher risk of inadequate early childhood development. Black et al. 5 found that prolonged exposure to violent discipline may generate toxic levels of stress that affect development in the first years of life. We highlight that our study presents several limitations. The first limitation is inherent to the cross-sectional study design, which does not allow for establishing causal relationships. Secondly, the environmental factors were measured at the municipality level since our ability to geocode the participants in the surveys was restricted to that level. Municipalities are fairly large and heterogeneous areas. However, the response rates of the two surveys were high and so were the sample sizes, which reduces the possibility of misclassification bias. Thirdly, we lacked information on the alcohol and tobacco consumption of the children’s parents, and whether the children were exposed at home. Therefore, we had to estimate the prevalence of consumption for each municipality using ENCODAT data with small area estimation methodology since the survey is not representative at the municipality level, which is widely used by statistical agencies in various countries. Finally, social determinants cover a wide spectrum of attributes, including housing circumstances; interpersonal interactions among children and parents; sociodemographic factors; day-care and school learning; among others. Our analysis partially covers all these compounds due to lack of availability of some variables in our database. Despite these limitations, our findings are consistent with prior research and this study served to identify potential associations between inadequate early childhood development and alcohol consumption and tobacco use, as well as other contextual factors. Future work based on longitudinal studies should try to confirm these associations with inadequate early childhood development. Conclusion We found that children who lived in municipalities with higher exposure to alcohol consumption and tobacco use were more likely to have inadequate early childhood development in Mexico. Considering that some characteristics at municipality-level are notably modifiable by public health policies, all efforts should be taken to develop policies aimed at reducing alcohol consumption and tobacco use in municipalities to ensure adequate early childhood development, as well as reducing the number of homicides and emphasizing the importance of targeting these issues with effective interventions within neighborhoods.