Abstract
The Mexican Health and Aging Study (MHAS) is one of the largest ongoing longitudinal studies of aging in Latin America, with six waves over 20 years. MHAS includes sociodemographic, economic, and health data from a nationally representative sample of adults 50 years and older in urban and rural Mexico. MHAS is designed to study the impact of diseases on adults’ health, function, and mortality. As Mexico is experiencing rapid population aging, providing adequate information to study this phenomenon is vital for designing and implementing public policies. The availability of biomarker and genetic data and longitudinal survey data elevates opportunities for research on aging in a low–middle-income country. This manuscript describes the profile of biomarkers and genetic data available in the MHAS study, including sample sizes and sociodemographic characteristics of participants who provided biospecimens for biomarker analyses, emphasizing recent genetic data. The sample size of individuals with anthropometric biomarkers was 2 707 (Wave 1—2001), 2 361 (Wave 2—2003), 2 086 (Wave 3—2012), and 2 051 (2016). Capillary blood samples were collected from 2 063 participants in 2012 (Wave 3) and 1 141 in 2016. Venous blood samples for blood-based biomarkers were collected from 2 003 participants in 2012 (Wave 3) and 752 in 2016. Venous blood samples were also collected for genetic data from 2 010 participants in 2012 (Wave 3) and 750 in 2016. A total of 7 821 participants provided saliva in 2018, and 2 671 provided hair in 2018. From these samples, a total of 7 204 have genome-wide genetic data, 8 600 have apolipoprotein-E genotype data, and 7 156 have genetic ancestry data.
Keywords: Anthropometric measures, Biomarkers, Genetic data, Mexico, MHAS
Introduction to the Mexican Health and Aging Study
Mexico is experiencing rapid population aging. The percentage of adults aged 60 and over was 11% in 2020; by 2050, it is expected to be close to 25% (1). Understanding the health and aging process of older adults in Mexico requires comprehensive data, especially as social and public health challenges arise from this accelerated aging.
The Mexican Health and Aging Study (MHAS) is a nationally representative panel study that prospectively evaluates the impact of aging on the health, function, and mortality of adults over 50 in urban and rural areas of Mexico. The study was designed to examine the aging process and its disease and disability burden among older Mexicans using a broad socioeconomic perspective (2,3). MHAS was the first study in a low–middle-income country designed to be comparable to the U.S. Health and Retirement Study (HRS) (4) and is part of the HRS network of studies, which now includes 17 studies in more than 40 countries that can be used for cross-national comparisons (2,5). The MHAS participants are drawn using multistage probabilistic sampling procedures to select a nationally representative sample in urban and rural areas. Interviews are conducted in Spanish and in participants’ homes by experienced interviewers from the Mexican Statistical Bureau (INEGI) (6). Interviewers from INEGI are specifically trained to follow the protocol and conduct the MHAS interviews. The survey instruments collect health, sociodemographic, economic, and social data, producing a rich phenotype.
Definition of Biomarkers in Mexican Health and Aging Study
Biomarkers are broadly defined as characteristics that can be measured as indicators of normal biological or pathogenic processes or processes that develop in response to an exposure or intervention, including therapeutic interventions (7). Thinking about biological and physiological changes related to aging requires considering complex mechanisms, many of which are not completely understood (8–11). In population-based aging studies, biomarkers have been used to examine biological processes that change with age, diseases where the onset seems to be related to age, and the process of aging itself (10). Social scientists use biomarkers in population-based studies to examine how social, psychological, and behavioral factors over the life course influence biological processes and, consequently, health outcomes (10,12–15). Some of the biomarkers described below are proxy measures for biological processes or physiological functions and are correlated with specific conditions (16). Overall, these biomarkers were selected for the MHAS because they can enhance our understanding of intrinsic biological changes associated with aging and how they relate to lifestyle, environmental, and socioeconomic factors (3,16). This information is instrumental for the study of healthy aging.
One of the main objectives of MHAS is to provide new information that the research community can use to study aging in Mexico comprehensively. Therefore, in addition to collecting comprehensive phenotype data for over 20 years, the study has also collected biomarker and genetic data on participants to study cardiovascular and metabolic health, genetic risk factors for medical conditions like Alzheimer’s disease and related dementias, and exposure to environmental contaminants. This manuscript presents the MHAS biomarker and genetic data profile to describe the data available in the study, including anthropometric biomarkers (anthropometric and performance measures), data from blood-based and hair-based samples, and genetic data extracted from blood or saliva. This profile highlights the study waves when different biomarkers were collected, the number of participants with biomarker and genetic data, and summarizes procedures used to guarantee the quality of the data.
MHAS Data Collection Description
The left side of Figure 1 presents the data collection timeline for the MHAS core surveys from the baseline interview in 2001–2021, including Wave 1 in 2001 and follow-ups in 2003 (Wave 2), 2012 (Wave 3), 2015 (Wave 4), 2018 (Wave 5), and 2021 (Wave 6). The MHAS baseline sample had 15 186 participants, which included target adults aged 50 and older and their spouses regardless of age. If applicable, new spouses are recruited to the study in follow-up interviews. In 2012 and 2018, the study added 2 refresher samples of 5 910 and 4 823 participants, respectively (including new spouses). In total, the MHAS study recruited 26 839 participants between 2001 and 2018. The left side of Figure 1 illustrates the longitudinal nature of the study with follow-up samples and new samples added as described earlier. The number of interviews at each wave ranges between 15 000 and 18 000. Another wave of the MHAS core survey was completed in 2021, which did not include biomarkers or genetic data collection.
Figure 1.
Description of MHAS data collection shows the study waves and sample sizes.
The MHAS study has also completed several ancillary studies using subsamples of the core MHAS participants, which are illustrated on the right side of Figure 1. Anthropometric biomarkers (anthropometric and performance measures) were collected in 2001, 2003, and 2012. Venous blood samples were collected for the first time in 2012 for blood-based biomarkers and genetic data. In 2016, as part of an ancillary study on cognitive aging, a subsample of 2 265 participants was selected from the MHAS 2015 (Wave 4). This ancillary study, known as the Mex-Cog, is designed to be comparable with the Harmonized Cognitive Aging Protocol coordinated by the HRS to measure cognitive function and dementia among older adults worldwide (17,18). As part of the Mex-Cog, anthropometric measures were obtained, and venous blood was collected for blood-based biomarkers and genetic data. In 2018 (Wave 5), a subsample provided saliva used to extract genetic data, and another subsample provided hair to examine exposure to heavy metals.
In total, MHAS collected biomarker and genetic data from 12 026 participants at different study times that include anthropometric, blood-based, and environmental exposure biomarkers and genetic data as described earlier. Because of the longitudinal nature of the MHAS study, participants might have more than one of these collected over time. For example, among those with anthropometric and biomarker information, 55% have data on one of these measures, and 45% have data on 2 or more.
Supplementary Table 1 and Supplementary Figure 1 provide additional information on MHAS participants, including information on more than 1 biomarker or 1 biomarker and genetic data. For example, there are 7 821 participants with genetic data extracted from saliva samples. Of these, 1 387 have a venous blood sample, and 1 381 have anthropometric data in at least 1 study wave (see Supplementary Table 1 and Figure 1).
Additionally, many participants who provided samples for biomarker data were interviewed in several study waves, resulting in a rich phenotype. For example, of the 12 026 participants with any biomarker or genetic data, 47% were interviewed in 5 waves, and 34% were interviewed in 4 or 3 waves of the core survey between 2001 and 2018 (data available upon request).
Summary of MHAS Biomarkers and Genetic Data Available for Research
In this section, we present the different biomarkers available in MHAS and the collection procedures. Table 1 provides the number of participants according to the sample type and the study wave in which samples were collected. There are 5 sources of biomarker and genetic data: venous blood, capillary blood, anthropometric and performance measurements, saliva, and hair.
Table 1.
Number of Cases for the Different Types of Biomarker Data in MHAS by Each Wave of Collection
| Wave (year) of Collection | Anthropometric Biomarkers | Blood-Based Biomakers* | Genetic Biomarkers* | Capillary Blood Sample (HbA1c) | Genetic Biomarkers (from saliva) | Exposure Biomarkers (from hair) |
|---|---|---|---|---|---|---|
| Wave 1 (2001) | 2 707 | — | — | — | — | — |
| Wave 2 (2003) | 2 361 | — | — | — | — | — |
| Wave 3 (2012) | 2 086 | 2 016 | 2 010 | 2 063 | — | — |
| Mex-Cog (2016) | 2 051 | 760 | 750 | 1 141 | — | — |
| Wave 5 (2018) | — | — | — | — | 7 821 | 2 671 |
Notes: See Supplementary Table 1 for detailed information from the number of cases eligible to the number of cases processed and in the MHAS Study. See Supplementary Table 2 for additional information on anthropometric measures. MHAS = Mexican Health and Aging Study.
*The number of participants with blood-based or genetic biomarkers is 2 016 in 2012 and 760 in 2016. The number of participants with both, blood-based and genetic biomarkers, is 1 997 in 2012 and 742 in 2016.
Anthropometric Biomarkers
In Waves 1 and 2, anthropometric and performance biomarkers were obtained in 2 707 and 2 361 participants, respectively. Standardized protocols were used, combining protocols from previous population studies in Mexico and the HRS. The data were collected by trained interviewers from the National Institute of Statistics and Geography (INEGI) in Mexico as part of the core survey interview. The first 6 anthropometric biomarkers in Supplementary Table 2 and balance were measured in the first 2 waves. Height was measured in centimeters using a flexible measuring tape provided to the interviewers. Similarly, interviewers used a digital Toledo scale (Mettler-Toledo International, Inc., Columbus, OH) to measure weight in kilograms. Waist and hip circumference was measured twice, and an average of both results was documented. Calf circumference was measured in centimeters, and it was only measured in 2001. Knee height was also measured in centimeters in the first 3 waves of the study. Balance was assessed using a 1-leg stand timed with a chronometer.
Additional anthropometric biomarkers were added in Wave 3 (2012) and as part of the Mex-Cog ancillary study in 2016. These data were collected by trained personnel from the National Institute of Public Health (INSP) in Mexico. Systolic and diastolic blood pressures were measured using a standard electronic OMRON cuff. Pulse was also measured using the same blood pressure cuff. Additional performance biomarkers, such as walking speed and hand grip strength, were captured in 2012 and 2016. To measure walking speed, the time in seconds to complete a 3-m walk was measured twice, and an average score was documented. Two measurements of hand grip strength in kilograms were obtained using a hydraulic handgrip dynamometer (Jamar Hydraulic Dynamometer, model 5030J1; JA Preston Corp., Clifton, NJ). Measurements from both hands were obtained when possible. Supplementary Table 2 summarizes the sample size and mean value for the 12 anthropometric and performance biomarkers measured between Waves 1 and 4.
Blood-Based Biomarkers
MHAS participants in 2012 and in the 2016 Mex-Cog ancillary study provided 2 venous blood samples. A total of 2 305 participants were selected to provide blood in 2012 and 1 429 in 2016. One venous blood sample was processed from a plain tube (red top) from 2 003 participants in 2012 and 752 participants in 2016. Blood in these tubes was centrifuged on-site within 30 minutes of extraction and stored in a nitrogen tank before transportation to the central laboratory of the INSP in Mexico, where samples were analyzed. Levels of total cholesterol and high-density lipoprotein were measured. Total cholesterol was measured using Cholesterol assay on the Architect c Systems and the AEROSET System, and HDL was measured using Ultra HDL assay for the Architect c Systems and the AEROSET System. C-reactive protein levels were also measured using the Architect Multigent CRP Vario assay. Vitamin D level was obtained using Architect 25-OH Vitamin D Assay; thyroid-stimulating hormone level was obtained using Architect TSH Reagent Kit, Architect TSH Calibrators, and Architect TSH Controls (information regarding the laboratory protocol are available upon request to admin@MHASWeb.org).
The second venous blood sample was collected in a tube with anticoagulant (lavender top, ethylenediaminetetraacetic acid [EDTA] tube) from 2 011 participants in 2012 and 750 participants in 2016. These samples were frozen and stored. DNA was extracted from these frozen samples at the National Institute of Genomic Medicine (INMEGEN) in collaboration with the National Institute of Geriatrics (INGER) in Mexico in 2017. This DNA was exported to the National Centralized Repository for Alzheimer’s Disease and Related Dementias (NCRAD) at Indiana University, which provides access to researchers for downstream analyses. A detailed description of the quality control is provided later.
In 2012 and 2016, 2 308 and 2 267 participants were also selected to provide two capillary blood samples. Samples were processed from 2 086 participants in 2012 and 1 141 in 2016. One capillary sample was used to measure glycosylated hemoglobin (HbA1c) levels using an A1CNOW device. The second capillary sample was used to measure hemoglobin levels with a HEMOCUE device.
Genetic Data
To supplement the blood-based genetic data obtained in 2012 and 2016, saliva samples for genetic data were obtained in Wave 5 (2018). These samples were obtained from 7 848 participants. Oragene 500 saliva kits were used for the sample collection. INEGI interviewers were trained to collect the saliva sample using an adapted version of the manufacturer’s training materials and prior teaching materials used by Columbia University for genetic studies with Hispanic populations. Each saliva sample was collected, individually packed, and sent to the headquarters of INEGI in Aguascalientes, Mexico, where they were stored and then transported to Mexico City for quality control. Quality control consisted of ensuring that each saliva tube had the minimum amount of saliva needed for DNA extraction, that there was no contamination of the package with saliva, and that all samples were correctly labeled. Saliva samples passing the quality control were exported to NCRAD in the United States, where they were processed, stored, and plated for shipment when requested for downstream analysis.
The most critical factor for DNA extraction is the additional recovery of bacterial DNA, which may affect downstream genetic analysis. However, it has been reported that most of the DNA in Oragene-saliva samples is of human origin (19). To further minimize bacterial DNA contamination, sensitive and specific qPCR assays were conducted to quantify the percentage of bacterial DNA in saliva samples following extraction and methods to deplete contaminating bacteria DNA from human DNA before library preparation.
Table 2 summarizes the genetic data available in MHAS. Of the 9 841 MHAS participants with genome-wide genetic data, 27% were derived from venous blood obtained in 2012 and 2016, and the rest were derived from saliva obtained in 2018. In addition, there is an overlap of participants with genome-wide genetic data derived from venous blood and from saliva. Of those with genome-wide genetic data derived from saliva (n = 7 218), 16% also obtained genome-wide genetic data from blood. For ancestry, 17.4% was derived from venous blood DNA and 82.6% from saliva DNA. For apolipoprotein-E (APOE) genotype data, 31.2% was derived from venous blood DNA and 68.8% from saliva DNA.
Table 2.
Number of Cases for Different Genetic Biomarker Data in MHAS by Wave of Collection
| MHAS Wave | Genome-Wide Genetic Data (blood) | Genome-Wide Genetic Data (saliva) | APOE Genotype Data | Ancestry |
|---|---|---|---|---|
| n = 2 623 | n = 7 218 | n = 8 600 | n = 7 156 | |
| In study* | ||||
| MHAS 2012, n (%) | 1 912 (73) | 1 951 (22.7) | 845 (11.8) | |
| Mex-Cog 2016, n (%) | 711 (27) | 733 (8.5) | 401 (5.6) | |
| MHAS 2018—saliva, n (%) | 7 218 (100) | 5 916(68.8) | 5 910(82.6) |
Notes: Please note that there is an overlap of genome-wide genetic data obtained from blood and saliva. APOE = apolipoprotein-E; MHAS = Mexican Health and Aging Study.
*For 2012 and 2016, genome-wide genetic data results come from EDTA tubes.
The Illumina GSA array was used for the genome-wide genetic data. The number of genotyped single nucleotide polymorphisms (SNPs) in the array was 759 993. The following quality control metrics were used: exclusion of individuals with sex discrepancies when comparing reported sex versus sex based on genetic data, and exclusion of SNPs with genotype missing rates>95%, with minimum allele frequencies <1%, with deviations from Hardy–Weinberg equilibrium (p ≤ 10−6). After applying the quality control filters, the total number of variants for downstream analysis purposes was 411 361. All the SNP data were used to estimate global genetic ancestry, using ADMIXTURE (v1.3.0) software and the Human Genome Diversity Project as the reference panel like previously detailed in publications from our group (20,21). Either genotyped or imputed genotypes at the APOE locus were used. In the saliva samples, genotypes at the APOE locus were directly derived from SNPs rs7412 and rs429358 using the KASPar PCR SNPs genotyping system (LGC Genomics, Hoddesdon, Herts, UK). The genotypes at APOE SNPs rs7412 and rs429358 for blood-derived samples were imputed using the TOPMed imputation server (https://imputation.biodatacatalyst.nhlbi.nih.gov/#) (22,23). Results comparing direct and imputed APOE genotyping revealed a highly accurate imputation process, with a concordance of 99%.
Environmental Exposure Biomarkers
During Wave 5 (2018), INEGI interviewers were also trained to take hair samples. Hair was collected from 2 671 participants, individually packaged, and shipped from across Mexico to INEGI’s headquarters in Aguascalientes, Mexico. Once all hair samples were received in the central office of INEGI, they were shipped to the National Institute of Ecology and Climate Change (INECC) in Mexico City for processing and measurement of levels of heavy metals, in collaboration with the INSP in Mexico. Levels of heavy metals in hair samples were determined using Inductively Coupled Plasma—Mass Spectrometry (ICP-MS). Before the mass spectrometry analyses, hair samples were washed, dried, and then digested using the microwave-assisted acid digestion of siliceous and organically based matrices, also known as the EPA 3520 method. Results from the mass spectrometry analyses were reported initially in µg/kg and then transformed to µg/g for comparison purposes. For additional details, see the laboratory manual available on the MHAS website (https://www.mhasweb.org/DataProducts/AncillaryStudies.aspx). Table 3 summarizes the results of the concentration of heavy metals identified in the hair samples. Of the 2 671 hair samples analyzed, titanium, copper, and lead were identified in more than 98% of the samples. Conversely, antimony and arsenic were identified in less than 27% of the hair samples. Copper had the highest mean concentration, while cobalt had the lowest.
Table 3.
Information on Heavy Metal Measurements From Hair Samples in MHAS 2018 (n = 2 671)
| Heavy Metal | # Observations | Mean Concentration (mg/g) | SD | Min | Max | % Total Sample | n < LOD | %<LOD |
|---|---|---|---|---|---|---|---|---|
| Titanium | 2 645 | 5.13 | 13.38 | 0.03 | 385.42 | 99.0% | 26 | 1.0% |
| Vanadium | 2 114 | 0.22 | 0.52 | 0.01 | 18.67 | 79.1% | 557 | 20.9% |
| Chromium | 2 499 | 0.42 | 1.07 | 0.03 | 22.71 | 93.6% | 172 | 6.4% |
| Manganese | 2 584 | 1.97 | 4.42 | 0.03 | 110.62 | 96.7% | 87 | 3.3% |
| Cobalt | 1 497 | 0.08 | 0.38 | 0.00 | 8.75 | 56.0% | 1 174 | 44.0% |
| Nickel | 2 557 | 1.25 | 3.11 | 0.04 | 61.93 | 95.7% | 114 | 4.3% |
| Copper | 2 645 | 12.34 | 24.34 | 0.02 | 429.70 | 99.0% | 26 | 1.0% |
| Arsenic | 663 | 0.09 | 0.23 | 0.01 | 5.01 | 24.8% | 2 008 | 75.2% |
| Molybdenum | 1 555 | 0.11 | 0.40 | 0.01 | 8.77 | 58.2% | 1 116 | 41.8% |
| Silver | 1 826 | 0.76 | 6.63 | 0.00 | 232.72 | 68.4% | 845 | 31.6% |
| Cadmium | 1 185 | 0.12 | 0.56 | 0.01 | 12.68 | 44.4% | 1 486 | 55.6% |
| Antimony | 712 | 0.14 | 0.85 | 0.01 | 17.13 | 26.7% | 1 959 | 73.3% |
| Lead | 2 624 | 3.82 | 86.63 | 0.01 | 4 273.50 | 98.2% | 47 | 1.8% |
| Mercury | 2 302 | 0.81 | 13.86 | 0.01 | 643.16 | 86.2% | 369 | 13.8% |
Note: LOD = limit of detection; MHAS = Mexican Health and Aging Study; SD = standard deviation.
A detailed description of all the procedures followed in each protocol to select the eligible participants, obtain the samples and measurements, process the samples, and store the biomarker and genetic data is beyond the scope of this profile. More details are available under “Survey Design” in the MHAS website (www.mhasweb.org).
Description of Participants With Longitudinal Biomarker Data
Because of the longitudinal nature of the MHAS, there are participants who have information on more than one of the biomarkers and genetic data. Supplementary Figure 1 shows the number of participants with multiple biomarkers and genetic data from blood samples for those that provided saliva samples. Of the 7 821 participants with genetic data from saliva samples, 1 035 have anthropometric data in at least 1 study wave, but do not have biomarkers from venous blood. In total, 2 146 participants with genetic data from saliva samples also have anthropometric data in at least 1 study wave.
Table 4 summarizes participants’ sociodemographic characteristics according to type of biomarker data. The table reveals the diversity in the subsamples for biomarkers and genetic analyses. Panel A shows the information for those with blood-based, genetic, and environmental biomarkers. For example, those who provided saliva were, on average, 67 years old, had 5.1 years of education, 58% lived in rural areas, and 77% had at least 1 chronic disease (among hypertension, diabetes, cancer, respiratory problems, and arthritis). In contrast, those who provided hair were, on average, 63 years old, had 5.7 years of education, 59% lived in rural areas, and 84% had at least 1 chronic disease (among hypertension, diabetes, cancer, respiratory problems, and arthritis). The table includes confidence intervals for comparison purposes. For example, MHAS participants who provided saliva were 6 years older than those who provided venous blood, capillary blood, or hair samples.
Table 4.
Sociodemographic and Health Characteristics of Participants With Biomarker Data by Source of Sample (Panel A), and by Genetic Biomarker (Panel B)
| Panel A | |||||
|---|---|---|---|---|---|
| Sociodemographic Characteristics at the Corresponding Wave | Venous Blood Sample | Capillary Blood Sample (HbA1c) | Genetic Biomarkers (saliva) | Environmental Biomarkers (hair) | |
| Blood-Based Biomarkers | Genetic Biomarkers (blood) | ||||
| (CI-95%) | (CI-95%) | (CI-95%) | (CI-95%) | (CI-95%) | |
| n | 2 755 | 2 760 | 2 959 | 7 821 | 2 671 |
| Sex (%) | |||||
| Females | 59.4 | 59.6 | 59.5 | 55.6 | 60.1 |
| (57.6–61.3) | (57.7–61.4) | (57.7–61.3) | (54.4–56.6) | (58.3–61.9) | |
| Age | 64.1 | 64.1 | 64.4 | 67.5 | 63.1 |
| (63.7–64.5) | (63.7–64.5) | (63.9–64.7) | (67.3–67.6) | (62.6–63.5) | |
| Years of formal education | 5.6 | 5.6 | 5.7 | 5.1 | 5.7 |
| (5.4–5.8) | (5.4–5.8) | (5.5–5.9) | (5.0–5.2) | (5.5–5.9) | |
| Rural* (%) | 59.0 | 58.9 | 59.7 | 57.9 | 58.7 |
| (57.1–60.8) | (57.1–60.8) | (57.9–61.5) | (56.8–59.0) | (56.7–60.5) | |
| Chronic diseases (%) | |||||
| Hypertension | 44.4 | 44.4 | 44.5 | 48.8 | 45.2 |
| (42.6–46.1) | (42.6–46.1) | (42.7–46.1) | (47.7–49.8) | (42.9–47.4) | |
| Diabetes | 24.4 | 24.6 | 24.4 | 24.7 | 26.3 |
| (22.8–26.0) | (22.9–26.2) | (22.8–26.0) | (23.8–25.7) | (24.2–28.4) | |
| Cancer | 2.1 | 2.1 | 2.1 | 1.9 | 1.9 |
| (1.6–2.6) | (1.6–2.6) | (1.6–2.6) | (1.6–2.2) | (1.3–2.5) | |
| Respiratory problems | 6.1 | 6.1 | 6.3 | 6.3 | 6.4 |
| (5.3–6.9) | (5.3–6.9) | (5.6–7.0) | (5.8–6.8) | (5.4–7.4) | |
| Arthritis | 14.9 | 15.0 | 15.3 | 16.1 | 15.4 |
| (13.7–16.1) | (13.8–16.2) | (14.0–16.4) | (15.3–16.8) | (13.8–16.9) | |
| Having ≥1 chronic disease (%) | 74.3 | 74.4 | 75.4 | 76.7 | 84.0 |
| (72.7–75.9) | (72.7–76.0) | (73.8–76.9) | (75.7–77.6) | (82.6–85.4) | |
| Medical insurance (%) | 93.7 | 93.7 | 93.7 | 95.7 | 94.8 |
| (92.7–94.6) | (92.7–94.6) | (92.8–94.6) | (95.2–96.2) | (93.7–95.9) | |
| Panel B | |||||
| Sociodemographic Characteristics | Genome-Wide Genetic Data—Blood (CI†-95%) | Genome-Wide Genetic Data—Saliva (CI-95%) | APOE Genotype Data (CI-95%) | Ancestry (CI-95%) | |
| n | 2 623 | 7 218 | 8 599 | 7 156 | |
| Sex (%) | |||||
| Females | 59.6 | 55.1 | 56.1 | 55.1 | |
| (57.6–61.4) | (53.9–56.2) | (55.0–57.2) | (53.9–56.2) | ||
| Age | 64.0 | 67.3 | 66.5 | 67.3 | |
| (63.6–64.4) | (67.1–67.5) | (66.3–66.6) | (67.1–67.5) | ||
| Years of formal education | 5.6 | 5.1 | 5.3 | 5.1 | |
| (5.4–5.8) | (5.0–5.2) | (5.2–5.4) | (5.0–5.2) | ||
| Rural* (%) | 58.9 | 57.8 | 58.2 | 57.8 | |
| (57.0–60.7) | (56.6–58.9) | (57.2–59.2) | (56.7–58.9) | ||
| Chronic diseases (%) | |||||
| Hypertension | 44.6 | 48.8 | 47.9 | 48.8 | |
| (42.8–46.4) | (47.7–49.9) | (46.8–48.9) | (47.7–49.9) | ||
| Diabetes | 24.4 | 24.5 | 24.5 | 24.6 | |
| (22.7–26.0) | (23.5–25.5) | (23.5–25.4) | (23.6–25.5) | ||
| Cancer | 2.2 | 1.9 | 2.0 | 1.9 | |
| (1.6–2.7) | (1.6–2.2) | (1.7–2.3) | (1.6–2.2) | ||
| Respiratory problems | 6.1 | 6.2 | 6.3 | 6.3 | |
| (5.2–6.8) | (5.7–6.7) | (5.9–6.8) | (5.8–6.8) | ||
| Arthritis | 14.8 | 15.9 | 15.5 | 15.9 | |
| (13.5–16.1) | (15.2–16.7) | (14.7–16.1) | (15.2–16.7) | ||
| Having ≥1 chronic disease (%) | 74.3 | 76.5 | 76.3 | 76.6 | |
| (72.6–76.0) | (75.5–77.5) | (75.4–77.2) | (75.6–77.5) | ||
| Medical insurance (%) | 93.9 | 95.7 | 95.3 | 95.7 | |
| (92.9–94.8) | (95.2–96.1) | (94.9–95.8) | (95.2–96.2) | ||
Notes: For venous blood sample and capillary blood sample (HbA1c), we use the average values of 2012 and 2015 for the mean age, chronic conditions, and medical insurance. For saliva and hair, we use the values of 2018 for all the variables. For genetic biomarkers we use the average values of 2012, 2015, and 2018 for the mean age, chronic conditions, and medical insurance.
*We defined rural as communities with less than 100 000 habitants.
†CI: confidence interval is calculated for a comparison between the 4 columns.
Panel B of Table 4 provides similar information for the genome-wide genetic data, genotype, and ancestry data. For example, 55% of those with ancestry data are women; their mean age is 68.8, and they have an average of 5.1 years of formal education. The prevalence of having at least one chronic disease (among hypertension, diabetes, cancer, respiratory problems, and arthritis) was higher among those with APOE genotype, ancestry, or genome-wide genetic data from saliva compared to those with blood-based genome-wide genetic data.
MHAS Biomarker Main Findings and Research Opportunities
Biomarkers in MHAS can be used in multiple ways to understand different aspects of aging in Mexico. Regarding anthropometric biomarkers, for example, using the mean weight and height for each MHAS wave, the mean body mass index (BMI) can be calculated (weight in kg/height in m2); see Supplementary Table 2. The mean BMI was 27.9 for 2001, 28.2 for 2003, 29.0 for 2012, and 28.5 for 2016. According to the World Health Organization guidelines, all these BMI values fall within the overweight category (24). BMI can also be correlated with waist and hip circumference, indicators of abdominal obesity considered for cardiovascular and metabolic risk studies. BMI can also be compared with knee height, considered an alternate nutritional status measure, to determine the concordance between these biomarkers and how they predict health over time.
Published work using anthropometric biomarkers in MHAS shows that obesity is associated with higher mortality and risk of disability and leads to higher incidence of diabetes, lower levels of vitamin D, cognitive impairment, and insomnia among older Mexican adults (25–33). This information can be used to inform healthcare leaders and policy-makers about health trends to design interventions that can help older Mexicans achieve better health outcomes. Anthropometric biomarkers have also been used to define syndromes in the MHAS population, such as frailty, and understand the risk factors leading to the onset of frailty and outcomes related to frailty (34–38).
Regarding blood biomarkers, for example, the normal range of C-reactive protein (CRP) is between 0.3 and 1.0 mg/dL. About 43% of the participants with CRP in 2012 and 2016 have a value >3.0 (Supplementary Table 3). These results raise the need to study further the factors contributing to inflammation in older Mexican adults. These levels can be associated with environmental risk factors, such as exposure to heavy metals, or can be used to examine the impact of inflammation on health outcomes, such as hospitalizations and quality of life. Relatedly, about 53% of participants with HDL, have a value for HDL >40 mg/dL, the recommended value to reduce cardiovascular risk; however, about 47% had total cholesterol above 200 in 2012 and 41% in 2016. These results can be studied in the context of policy and health insurance changes in Mexico (39) to examine screening behavior or correlated with self-reports of heart disease or mortality rates to study changes in cardiovascular health among older Mexican adults.
Published work using blood biomarkers in MHAS shows that low levels of vitamin D are associated with an increased risk of disability among older Mexican adults with arthritis (40). Other work has linked low vitamin D levels to risky health behaviors, such as smoking and physical inactivity, low socioeconomic status, and uncontrolled diabetes (25). These data allow us to advance the field of geroscience by providing data to understand biological mechanisms that drive aging among a well-defined cohort of older Hispanic adults that has been followed for 20 years (2,41).
Other biomarkers in MHAS have been used to compare with self-reports of diseases, such as hypertension and diabetes, to estimate or confirm prevalence rates, characterize populations with undiagnosed conditions, such as diabetes and dementia, or study mortality risk (27,42–44). This rich body of literature combines abundant sociodemographic, health, and economic information collected longitudinally from Mexican adults over 50 since 2001 with biomarker data, which allows a better characterization of aging in this Latin American cohort. In turn, this enables examination of how the life-course experiences of more than 27 000 Mexican adults are associated with their biological processes while they age.
Implications of MHAS Biomarker and Genetic Data
The MHAS is one of only a few population-based longitudinal studies in Latin America. The Costa Rican Longevity and Healthy Aging Study (CRELES—Costa Rica: Estudio de Longevidad y Envejecimiento Saludable) was conducted between 2005 and 2009 and included anthropometric, urine, blood biomarkers, and genetic data. The sample size for CRELES is 2 827 adults aged 60 and older (45). The Brazilian Longitudinal Study of Health, Ageing & Well Being (ELSI-Brasil) started in 2015 and collects data every 3 years. The overall sample size for ELSI-Brasil is close to 10 000 participants aged 50 and older (45). ELSI-Brasil includes most of the biomarkers measured in MHAS, including anthropometric biomarkers, blood biomarkers, and genetic data in a subsample of participants. Of these 3 studies, MHAS has the longest follow-up and the largest sample size. This makes MHAS an important study to advance aging research in Latin America and the Caribbean with comprehensive sociodemographic, economic, and health data enriched with biomarker data.
Multiple studies have used biomarker and genetic data to advance research on aging using MHAS data. This wealth of data has contributed to reducing the gap in the literature related to biological and healthy aging among Hispanic older adults; however, much remains to be done. MHAS data can be used to study aging in a comprehensive longitudinal cohort of older adults with Hispanic background and ancestry. In this biomarker and genetic data profile, we have described available data to motivate scientists to explore research questions that can be answered using this data and help produce scholarly work to characterize the aging process in Mexico and other low- and middle-income countries.
Three new dimensions of aging research may be explored with the addition of genetic data to MHAS: (i) describing how behavioral and environmental risk factors modify genes and affect health, (ii) identifying biological pathways and proteins that can be used to develop therapeutic targets, and (iii) providing genetic data linked to rich phenotype data on older Hispanic adults, an underrepresented group in genetic studies.
Ethnic-specific genetic risk factors have been identified for many complex diseases, and the effects of these genetic differences vary across populations (46). The association of APOE genotype with Alzheimer’s disease (AD) risk is an example of how genetic variants may affect populations differently. Although the APOE ε4 allele is more frequent among African Americans and Caribbean Hispanics, it is associated with a lower risk of AD for Hispanic adults compared to non-Hispanic White adults (47,48). This suggests that additional genetic or environmental factors influence disease risk in these populations. Furthermore, the percentage of Native American ancestry among Mexicans is close to 50%, a percentage that is much higher than the one reported for Caribbean Hispanics, which is close to 10% (49,50).
Even though data to reach a clinical diagnosis of dementia are not collected as part of the MHAS study, we use available data to get to a dementia category, following what comparable studies have done and using the expertise of some of our members and consultants. The algorithm used is described on the MHAS website (https://www.mhasweb.org/resources/DOCUMENTS/Constructed_Imputed/MHAS_Cognitive_Function_Measures_Scoring_and_Classification.pdf), and several manuscripts have used this variable to examine risk factors for cognitive decline and cognitive impairment and the impact of dementia on older adults in Mexico (18,51–55).
With the MHAS data, it is also possible to use novel methods, such as admixture mapping, to identify genetic characteristics associated with diseases like AD in ethnic groups with different ancestral genetic compositions, such as Mexicans. Other research may identify rare genetic variants that occur in less than 1% of the world’s population and examine their role in the development of AD among Hispanics.
The possibility of identifying new genetic variants contributing to the risk of AD might guide the development of therapeutics targeting these genes. Moreover, sample sizes of non-European ancestry populations in AD genome-wide association studies are generally small. Diversifying participants in AD genetic studies will improve the effectiveness of genomic medicine by expanding the scope of known genetic variation (ancestry-specific or ancestry-enriched risk variants), facilitating locus discovery, improving functional mapping, and ultimately bolstering our understanding of disease etiology, and increasing equity (56).
Accessing the Data
The MHAS website serves as a dynamic repository for the study, providing free access to databases, codebooks, and documentation. The platform is in English (www.MHASweb.org) and in Spanish (www.ENASEM.org). Registered users can download raw data, imputed and constructed variables, and harmonized variables. The website also includes information about survey data that has been linked to external data capturing the context of residence of the MHAS participants at the community level, such as from population censuses, vital registration records, and censuses of health services. These data are available to the research community through restricted-use agreements to preserve the confidentiality of the participants. MHAS biomarker data, including anthropometric and performance measures, is available through the study website. Genome-wide genetic data from venous blood can be accessed through the NIAGADS repository (https://adsp.niagads.org/data/contributing-cohorts/ and https://dss.niagads.org/cohorts/mexican-health-and-aging-study-mhas/). Genome-wide genetic data from saliva samples, APOE genotype data, and ancestry data are available through a restricted-use agreement directly with the MHAS group; more information on how to request access or download the forms can be obtained from the study website (https://www.mhasweb.org/DataProducts/Linkages.aspx).
Summary and Future Work
This profile summarizes the biomarkers and genetic data available in MHAS. The long follow-up of this longitudinal study provides a rich phenotype that, combined with biomarkers and genetic data, can contribute significantly to the field of aging. This contribution hinges on a unique feature of this population for the study of aging: Mexico is a low- and middle-income country aging at a rapid pace, aging much faster than high-income countries that aged earlier. This feature is common with other low- and middle-income countries, but not many have the richness of data offered by MHAS, including a national sample and long follow-up. Furthermore, the ability to compare the data with other countries in the HRS network enhances the research impact of this data collection endeavor.
The proliferation of biomarkers and genetic data over the last decade has expanded the ability of population studies, such as MHAS, to advance the field of aging research, with the possibility of using large data on national, urban, and rural areas of countries like Mexico. These studies have shared a common approach of emphasizing the life course, including self-reported childhood socioeconomic and health conditions, work experiences, and history of migration, as well as the larger context of social and economic programs over their lifetime. The availability of these combined data sets means that researchers from multiple disciplines can collaborate and use the data, incorporating perspectives beyond what was possible with traditional data sources in the past.
Providing research communities access to this set of rich data from multiple sources (self-reports, biomarker results, performance measures, community historical records, and others) is expected to enhance the scholarly work matching this data proliferation and lead to the development of more advanced methods and theories in the decades to come.
Supplementary Material
Contributor Information
Rafael Samper-Ternent, Institute on Aging, UTHealth Houston, Houston, Texas, USA; School of Public Health, UTHealth Houston, Houston, Texas, USA.
Jesús Daniel Zazueta-Borboa, Netherlands Interdisciplinary Demographic Institute KNAW/University of Groningen, 2511 CV The Hague, The Netherlands.
Alejandra Michaels-Obregon, Department of Population Health Sciences, UTHealth San Antonio, Texas, USA; Barshop Institute for Longevity and Aging Studies, UTHealth San Antonio, Texas, USA.
Dolly Reyes-Dumeyer, Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, Department of Neurology, Columbia University College of Physicians and Surgeons, New York, USA.
Sandra Barral, Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, Department of Neurology, Columbia University College of Physicians and Surgeons, New York, USA.
Giuseppe Tosto, Taub Institute for Research on Alzheimer’s Disease and the Aging Brain, Department of Neurology, Columbia University College of Physicians and Surgeons, New York, USA.
Rebeca Wong, Department of Population Health Sciences, UTHealth San Antonio, Texas, USA; Barshop Institute for Longevity and Aging Studies, UTHealth San Antonio, Texas, USA.
Funding
The Mexican Health and Aging Study is partly sponsored by the National Institutes of Health (NIH)/National Institute on Aging (NIA) (grant number NIH R01AG018016) in the United States and the Instituto Nacional de Estadística y Geografía in Mexico. Data collection for ancillary studies was supported by R01AG051158 and R56AG059756.
Conflict of Interest
None.
Author Contributions
All authors contributed to this manuscript as follows: R.S.T., J.D.Z.B., A.M.O., and R.W. contributed to the study design, writing of the first draft, and conducting reviews and edits of subsequent versions of the manuscript; J.D.Z.B. and A.M.O. conducted all statistical analyses and created the tables; R.S.T., J.D.Z.B., A.M.O., D.R.D., S.B., G.T., and R.W. reviewed all versions of the manuscript and helped edit the final version of the manuscript; R.S.T., A.M.O., D.R.D., S.B., G.T., and R.W. verified all genetic data from collection to extraction and analyses; AMO and RW verified all anthropometric and blood biomarker data. The sponsor had no role in the elaboration of this manuscript.
References
- 1. Lopez-Ortega M, Aranco N.. Aging and Care of Persons with Disability in Mexico [Envejecimiento y atención a la dependencia en México]. Inter-American Development Bank (IDB); 2019:63. [Google Scholar]
- 2. Wong R, Michaels-Obregon A, Palloni A.. Cohort profile: the Mexican Health and Aging Study (MHAS). Int J Epidemiol. 2017;46:e2. https://doi.org/ 10.1093/ije/dyu263 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Wong R, Michaels-Obregon A, Palloni A, et al. Progression of aging in Mexico: the Mexican Health and Aging Study (MHAS) 2012. Salud Publica Mex. 2015;57(Suppl 1):S79–S89. https://doi.org/ 10.21149/spm.v57s1.7593 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Sonnega A, Faul JD, Ofstedal MB, Langa KM, Phillips JW, Weir DR.. Cohort profile: the Health and Retirement Study (HRS). Int J Epidemiol. 2014;43:576–585. https://doi.org/ 10.1093/ije/dyu067 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lee J, Phillips D, Wilkens J; Gateway to Global Aging Data Team. Gateway to global aging data: resources for cross-national comparisons of family, social environment, and healthy aging. J Gerontol B Psychol Sci Soc Sci. 2021;76:S5–S16. https://doi.org/ 10.1093/geronb/gbab050 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Study MHaA. MHAS methodological document. 2018.
- 7. Daniel G, McClellan E, Richardson E, Nosair W.. Facilitating Biomarker Development: Strategies for Scientific Communication, Pathway Prioritization, Data-Sharing, and Stakeholder Collaboration. Duke University; 2016. [Google Scholar]
- 8. Thyagarajan B, Shippee N, Parsons H, et al. How does subjective age get “under the skin?” The association between biomarkers and feeling older or younger than one’s age: the Health and Retirement Study. Innov Aging. 2019;3:igz035. https://doi.org/ 10.1093/geroni/igz035 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Ferrucci L, Cavazzini C, Corsi A, et al. Biomarkers of frailty in older persons. J Endocrinol Invest. 2002;25:10–15. [PubMed] [Google Scholar]
- 10. Crimmins EM, Vasunilashorn S, Kim JK, Alley D.. Biomarkers related to aging in human populations. Adv Clin Chem. 2008;46:161–216. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Chen X, Crimmins E, Hu PP, et al. Venous blood-based biomarkers in the China Health and Retirement Longitudinal Study: rationale, design, and results from the 2015 wave. Am J Epidemiol. 2019;188:1871–1877. https://doi.org/ 10.1093/aje/kwz170 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Ailshire JA, Crimmins EM.. Fine particulate matter air pollution and cognitive function among older US adults. Am J Epidemiol. 2014;180:359–366. https://doi.org/ 10.1093/aje/kwu155 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Crimmins EM, Soldo BJ, Kim JK, Alley DE.. Using anthropometric indicators for Mexicans in the United States and Mexico to understand the selection of migrants and the “Hispanic paradox”. Soc Biol. 2005;52:164–177. https://doi.org/ 10.1080/19485565.2005.9989107 [DOI] [PubMed] [Google Scholar]
- 14. Mitchell UA, Ailshire JA, Crimmins EM.. Change in cardiometabolic risk among Blacks, Whites, and Hispanics: findings from the Health and Retirement Study. J Gerontol A Biol Sci Med Sci. 2019;74:240–246. https://doi.org/ 10.1093/gerona/gly026 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. McCrory C, McLoughlin S, Layte R, et al. Towards a consensus definition of allostatic load: a multi-cohort, multi-system, multi-biomarker individual participant data (IPD) meta-analysis. Psychoneuroendocrinology. 2023;153:106117. https://doi.org/ 10.1016/j.psyneuen.2023.106117 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Velly AM, Mohit S, Schipper HM, Gornitsky M.. Biomarkers in epidemiologic research: Definition, classification, and implication. In: Goulet J-P, Velly AM, eds. Orofacial Pain Biomarkers. Springer Berlin Heidelberg; 2017:135–139. [Google Scholar]
- 17. Langa KM, Ryan LH, McCammon RJ, et al. The Health and Retirement Study harmonized cognitive assessment protocol project: study design and methods. Neuroepidemiology. 2020;54:64–74. https://doi.org/ 10.1159/000503004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Mejia-Arango S, Nevarez R, Michaels-Obregon A, et al. The Mexican Cognitive Aging Ancillary Study (Mex-Cog): study design and methods. Arch Gerontol Geriatr. 2020;91:104210. https://doi.org/ 10.1016/j.archger.2020.104210 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Nunes AP, Oliveira IO, Santos BR, et al. Quality of DNA extracted from saliva samples collected with the Oragene DNA self-collection kit. BMC Med Res Methodol. 2012;12:65. https://doi.org/ 10.1186/1471-2288-12-65 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Alexander DH, Lange K.. Enhancements to the ADMIXTURE algorithm for individual ancestry estimation. BMC Bioinf. 2011;12:246. https://doi.org/ 10.1186/1471-2105-12-246 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Kizil C, Sariya S, Kim YA, et al. Admixture mapping of Alzheimer’s disease in Caribbean Hispanics identifies a new locus on 22q13.1. Mol Psychiatry. 2022;27:2813–2820. https://doi.org/ 10.1038/s41380-022-01526-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Fuchsberger C, Abecasis GR, Hinds DA.. minimac2: faster genotype imputation. Bioinformatics. 2015;31:782–784. https://doi.org/ 10.1093/bioinformatics/btu704 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Taliun D, Harris DN, Kessler MD, et al. ; NHLBI Trans-Omics for Precision Medicine (TOPMed) Consortium. Sequencing of 53,831 diverse genomes from the NHLBI TOPMed program. Nature. 2021;590:290–299. https://doi.org/ 10.1038/s41586-021-03205-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. World Health Organization. Global Database on Body Mass Index. WHO Press; 2006. [Google Scholar]
- 25. Carrillo-Vega MF, Garcia-Pena C, Gutierrez-Robledo LM, Perez-Zepeda MU.. Vitamin D deficiency in older adults and its associated factors: a cross-sectional analysis of the Mexican Health and Aging Study. Arch Osteoporos. 2017;12:8. https://doi.org/ 10.1007/s11657-016-0297-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Kumar A, Karmarkar AM, Tan A, et al. The effect of obesity on incidence of disability and mortality in Mexicans aged 50 years and older. Salud Publica Mex. 2015;57(Suppl 1):S31–S38. https://doi.org/ 10.21149/spm.v57s1.7587 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Kumar A, Wong R, Ottenbacher KJ, Al Snih S.. Prediabetes, undiagnosed diabetes, and diabetes among Mexican adults: findings from the Mexican Health and Aging Study. Ann Epidemiol. 2016;26:163–170. https://doi.org/ 10.1016/j.annepidem.2015.12.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Milani SA, Lopez DS, Downer B, Samper-Ternent R, Wong R.. Effects of diabetes and obesity on cognitive impairment and mortality in older Mexicans. Arch Gerontol Geriatr. 2022;99:104581. https://doi.org/ 10.1016/j.archger.2021.104581 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Momin KN, Sheehan C, Samper-Ternent R, Lopez DS, Wong R, Milani SA.. The impact of insomnia symptoms on obesity among Mexicans aged 50 and older. Salud Publica Mex. 2023;65:530–541. https://doi.org/ 10.21149/14759 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Monteverde M, Noronha K, Palloni A, Novak B.. Obesity and excess mortality among the elderly in the United States and Mexico. Demography. 2010;47:79–96. https://doi.org/ 10.1353/dem.0.0085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Monteverde M, Novak B.. Obesidad y Esperanza de Vida en Mexico Obesity and life expectancy in Mexico. Poblac Salud Mesoam. 2008;6:4. [PMC free article] [PubMed] [Google Scholar]
- 32. Rontoyanni VG, Avila JC, Kaul S, Wong R, Veeranki SP.. Association between obesity and serum 25(OH)D concentrations in older Mexican adults. Nutrients. 2017;9:97. https://doi.org/ 10.3390/nu9020097 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Ruiz-Arregui L, Castillo-Martinez L, Orea-Tejeda A, Mejia-Arango S, Miguel-Jaimes A.. Prevalence of self-reported overweight-obesity and its association with socioeconomic and health factors among older Mexican adults. Salud Publica Mex. 2007;49(Suppl 4):S482–S487. https://doi.org/ 10.1590/s0036-36342007001000007 [DOI] [PubMed] [Google Scholar]
- 34. Garcia-Gonzalez JJ, Garcia-Pena C, Franco-Marina F, Gutierrez-Robledo LM.. A frailty index to predict the mortality risk in a population of senior Mexican adults. BMC Geriatr. 2009;9:47. https://doi.org/ 10.1186/1471-2318-9-47 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Sourial N, Bergman H, Karunananthan S, et al. Contribution of frailty markers in explaining differences among individuals in five samples of older persons. J Gerontol A Biol Sci Med Sci. 2012;67:1197–1204. https://doi.org/ 10.1093/gerona/gls084 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Aguilar-Navarro SG, Amieva H, Gutierrez-Robledo LM, Avila-Funes JA.. Frailty among Mexican community-dwelling elderly: a story told 11 years later. The Mexican Health and Aging Study. Salud Publica Mex. 2015;57(Suppl 1):S62–S69. https://doi.org/ 10.21149/spm.v57s1.7591 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Perez-Zepeda MU, Gonzalez-Chavero JG, Salinas-Martinez R, Gutierrez-Robledo LM.. Risk factors for slow gait speed: a nested case-control secondary analysis of the Mexican Health and Aging Study. J Frailty Aging. 2015;4:139–143. https://doi.org/ 10.14283/jfa.2015.63 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Perez-Zepeda MU, Garcia-Pena C, Carrillo-Vega MF.. Individual and cumulative association of commonly used biomarkers on frailty: a cross-sectional analysis of the Mexican Health and Aging Study. Aging Clin Exp Res. 2019;31:1429–1434. https://doi.org/ 10.1007/s40520-019-01127-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Servan-Mori E, Gomez-Dantes O, Contreras D, et al. Increase of catastrophic and impoverishing health expenditures in Mexico associated to policy changes and the COVID-19 pandemic. J Glob Health. 2023;13:06044. https://doi.org/ 10.7189/jogh.13.06044 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. Valderrama-Hinds LM, Al Snih S, Rodriguez MA, Wong R.. Association of arthritis and vitamin D insufficiency with physical disability in Mexican older adults: findings from the Mexican Health and Aging Study. Rheumatol Int. 2017;37:607–616. https://doi.org/ 10.1007/s00296-016-3622-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41. de Souto Barreto P, Rolland Y, Ferrucci L, et al. Looking at frailty and intrinsic capacity through a geroscience lens: the ICFSR & Geroscience Task Force. Nat Aging. 2023;3:1474–1479. https://doi.org/ 10.1038/s43587-023-00531-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42. Palloni AS, Beth; Wong, Rebeca.. The Accuracy of Self-Reported Anthropometric Measures and Self-Reported Diabetes in Nationally Representative Samples of Older Adults in Mexico. Population Association of America; 2003. [Google Scholar]
- 43. Melano-Carranza E, Lasses Ojeda LA, Avila-Funes JA.. [Factors associated with untreated hypertension among older adults: results of the Mexican Health and Aging Study, 2001]. Rev Panam Salud Publica. 2008;23:295–302. https://doi.org/ 10.1590/s1020-49892008000500001 [DOI] [PubMed] [Google Scholar]
- 44. Salinas JJ, Eschbach KA, Markides KS.. The prevalence of hypertension in older Mexicans and Mexican Americans. Ethn Dis. 2008;18:294–298. https://pmc.ncbi.nlm.nih.gov/articles/PMC3086015/ [PMC free article] [PubMed] [Google Scholar]
- 45. The Biomarker Network. Population studies with biomarkers. 2023.
- 46. National Research Council (US) Panel on Race E, and Health in Later Life. Genetic influences. In: Bulatao RA, Anderson NB, eds. Understanding Racial and Ethnic Differences in Health in Late Life: A Research Agenda. National Academies Press; 2004. [PubMed] [Google Scholar]
- 47. Granot-Hershkovitz E, Tarraf W, Kurniansyah N, et al. APOE alleles’ association with cognitive function differs across Hispanic/Latino groups and genetic ancestry in the study of Latinos—investigation of neurocognitive aging (HCHS/SOL). Alzheimers Dement. 2021;17:466–474. https://doi.org/ 10.1002/alz.12205 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48. Naslavsky MS, Suemoto CK, Brito LA, et al. Global and local ancestry modulate APOE association with Alzheimer’s neuropathology and cognitive outcomes in an admixed sample. Mol Psychiatry. 2022;27:4800–4808. https://doi.org/ 10.1038/s41380-022-01729-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49. Moreno-Estrada A, Gignoux CR, Fernández-López JC, et al. The genetics of Mexico recapitulates Native American substructure and affects biomedical traits. Science. 2014;344:1280–1285. https://doi.org/ 10.1126/science.1251688 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Bryc K, Durand EY, Macpherson JM, Reich D, Mountain JL.. The genetic ancestry of African Americans, Latinos, and European Americans across the United States. Am J Hum Genet. 2015;96:37–53. https://doi.org/ 10.1016/j.ajhg.2014.11.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51. Arce Renteria M, Briceno EM, Chen D, et al. Memory and language cognitive data harmonization across the United States and Mexico. Alzheimers Dement. 2023;15:e12478. https://doi.org/ 10.1002/dad2.12478 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52. Barral S, Gao Y, Mejia-Arango S, et al. Genome-wide gene-based analysis of episodic memory trajectories in the Mexican Health and Aging Study (MHAS). Alzhemier’s Association International Conference; 2020.
- 53. Michaels-Obregon A, Arango SM, Wong R.. The Mexican Health and Aging Study: Cognitive Function Measures Scoring and Classification Across Waves 2001-2015. The Mexican Health and Aging Study; 2022. [Google Scholar]
- 54. Saenz JL, Beam CR, Zelinski EM.. The association between spousal education and cognitive ability among older Mexican adults. J Gerontol B Psychol Sci Soc Sci. 2020;75:e129–e140. https://doi.org/ 10.1093/geronb/gbaa002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55. Saenz JL, Wong R, Ailshire JA.. Indoor air pollution and cognitive function among older Mexican adults. J Epidemiol Community Health. 2018;72:21–26. https://doi.org/ 10.1136/jech-2017-209704 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 56. Andrews SJ, Renton AE, Fulton-Howard B, Podlesny-Drabiniok A, Marcora E, Goate AM.. The complex genetic architecture of Alzheimer’s disease: novel insights and future directions. EBioMedicine. 2023;90:104511. https://doi.org/ 10.1016/j.ebiom.2023.104511 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.