Archives: News

Add Health is excited to announce the following data are now available:

Wave V Epigenetic Clocks: Biological aging measures were derived from DNA methylation data for 4,626 unique study participants based on blood collected during the Wave V home exam. The measures include Horvath, PhenoAge, GrimAge, DunedinPACE, Principal Component clocks, and cell type proportion estimates.

For additional recently released datasets, please visit the File Inventory. These data are only available to users with SRW or Home Institution contracts. The files are already available on the SRW and will be available soon to Home Institution users.

For more information on the SRW and Home Institution contract options, please visit the Restricted-Use Contract Options page.

A recent study using nationally representative data from Waves IV and V of the National Longitudinal Study of Adolescent to Adult Health (Add Health) examines the effects of long-term exposure to two widely cited pollutants, ozone (O₃) and particulate matter 2.5 (PM₂.₅), on allergy and asthma diagnoses in adolescents and young adults. Ozone and particulate matter were measured using annual average concentrations generated from the EPA’s Fused Air Quality Surface using Downscaling (FAQSD) data and assigned to participants based on their tract of residence in each wave. Health outcomes were measured using self-reported survey questions asking whether the participant’s child had ever been diagnosed with asthma or allergies, with relevant covariates included in the analysis.

The study focused on 6,860 children between the ages of 10 and 25 of the original Add Health participants from Waves IV and V. Findings showed that exposure to ozone and particulate matter was associated with a higher likelihood of asthma in females aged 10 to 12 and a higher likelihood of allergies in males aged 13 and above. This study shows that air pollution can have varying health impacts, dependent upon timing and demographics, which emphasizes that environmental regulations and interventions must account for both age and sex.

To read the full article click the link below. For more important findings using Add Health data, visit the Add Health publications page.

Amialchuk, Aliaksandr, and Onur Sapci. “The Effect of Long-Term Exposure to O3 and PM2.5 on Allergies and Asthma in Adolescents and Young Adults.” International Journal of Environmental Research and Public Health, vol. 22, no. 8, Aug. 2025, p. 1262.  doi:10.3390/ijerph22081262.

There is a large pool of research focusing on how adolescents’ perceptions of survival predict important aspects of their young adult lives, especially their physical and mental health, risky behaviors, and socioeconomic status. Using nationally representative data from Wave I of the National Longitudinal Study of Adolescent to Adult Health (Add Health), Carlyn Graham et. al. investigated whether those perceptions extend beyond these aspects to actually predict premature mortality, controlling for demographic, socioeconomic, health, and behavioral factors. Perceived survival expectations were measured using a single Wave I survey question asking, “What do you think are the chances you will live to age 35?” Responses ranged from “almost no chance” to “almost certain” and were categorized into three groups: “50% chance or less,” “a good chance,” and “almost certain.” The outcome, all-cause mortality (measured binarily as alive or deceased), was tracked from Wave I (1994–95) through December 2021 using the Mortality Outcomes Surveillance Data files.

The study included 18,923 participants and findings showed that when adjusting for sex and race, adolescents who perceived a 50% chance or less of surviving to age 35 had a higher risk of mortality compared to those who were almost certain they would survive. The strength of this association was weakened after sequentially adjusting for socioeconomic status, physical and mental health, risky behaviors, and exposure to violence. When separating analysis by sex, lower perceived survival was strongly associated with higher premature mortality risk among female adolescents, but not males. This study highlights the importance of addressing adolescents’ survival perceptions, emphasizing that healthcare providers should pay special attention to females’ sense of risk at a young age.

To read the full article click the link below. For more important findings using Add Health data, visit the Add Health publications page.

Graham, Carlyn, Robert A. Hummer, and Carolyn T. Halpern. 2025. “Gazing into the Crystal Ball: Do Adolescent Survival Expectations Predict Premature Mortality Risk in the United States?” Social Science & Medicine 364:117548. doi:10.1016/j.socscimed.2024.117548.

Add Health is excited to announce the following data are now available:

Wave VI Anthropometrics

Wave VI Cardiovascular Measures

Wave VI Demographics Home Exam

Wave VI Glucose Homeostasis

Wave VI Hepatic Injury

Wave VI Lipids

Wave VI Medications Home Exam

Current Add Health investigators can log in to the CPC Data Portal and use the “Request More Data” button to order these datasets.

For more information on the CPC Data Portal, please visit the Frequently Asked Questions page. 

The following Add CAPS data are now available:

Animal Naming Test

This test provides data on a participant’s verbal fluency/language by assessing their ability to generate words within a semantic category under time constraints. Participants are asked to name as many animals as possible within a fixed time period. The primary research outcome for this test is the total number of unique, valid animal names produced within 60 seconds. N=2,613, 20 variables

Physical Function

Handgrip strength is a reflection of the overall state of muscle strength and physical function. These test results serve as a practical indicator of age-related functional decline. N=2,613, 28 variables

Sensory Function

The Sensory Function dataset provides data on participants’ auditory functioning, assessed using the hearX hearTest mobile audiometry system. This assessment measures hearing thresholds across multiple frequencies in both ears to evaluate hearing sensitivity and identify potential hearing impairment. The primary research outcomes include pure-tone averages (PTA) for each ear. N=2,613, 49 variables

NIH Toolbox Cognition

The NIH Toolbox Cognition battery includes three cognitive assessments: Dimensional Change Card Sort Test, Pattern Comparison Processing Speed Test, and a Picture Vocabulary Test. These tests provide data on the cognitive domains of executive function, language, and processing speed. N=2,613, 55 variables

Word Recall and Backward Digit Span

The Word Recall tasks were included in the Add CAPS battery to assess episodic memory, a cognitive domain that is known to decline with age and is predictive of dementia risk. Backward Digit Span provides a measure of working memory, reflecting the ability to temporarily hold and manipulate information. Results from these assessments can be combined with results from the same tests in Waves IV and V to conduct longitudinal analyses of cognitive function. Both tasks were adapted from standard neurocognitive batteries. N=2,613, 72 variables

To access the Wave VI restricted-use data, please visit the Add Health Data Portal.

Add Health is pleased to announce the following Wave VI Restricted-Use Data are now available:

• Wave VI Mixed-Mode Survey (Samples 1 and 2)
• Wave VI Mixed-Mode Survey (Sample 2)
• Wave VI AddCAPS – Test My Brain
• Wave VI Survey Medications 
• Wave VI Final Disposition of Initially Eligible Cases
• Wave VI Mixed-Mode Survey Weights

Please visit the Add Health Data Portal to access the Wave VI restricted-use data. As a reminder, researchers must have one of two contract options to access Wave VI data.

1. Secure Research Workspace (SRW) Contract: This option provides access to Add Health data through the University of North Carolina’s Secure Research Workspace (SRW).

2. Home Institution Contract: This option to access Add Health data requires the Principal Investigator’s home institution to deploy a secure server that complies with federal data security standards and is managed by the institution’s IT unit.

Comprehensive information on both access options may be found on the Restricted-Use Contract Options page.

The new Contracts Management System (CMS) is now available. Any researcher interested in applying for a new restricted-use contract or renewing their existing contract may access the CMS via the Add Health Data Portal. 

Researchers now have two options for accessing Add Health restricted-use data:

1. Secure Research Workspace (SRW) Contract: This option provides access to Add Health data through the University of North Carolina’s Secure Research Workspace (SRW).

2. Home Institution Contract: This option to access Add Health data requires the Principal Investigator’s home institution to deploy a secure server that complies with federal data security standards and is managed by the institution’s IT unit. 

The CMS is designed to facilitate an online process for the application, renewal, and administration of restricted-use data agreements, thereby enhancing efficiency and user experience. Anyone who is currently working on the SRW does not need to do anything at this time.

With the release of Wave VI data, the UNC Data Portal will no longer support Add Health data downloads. This change applies to requests for both newly released and previously available datasets. As a reminder, researchers must have one of two contract options to access Wave VI data when it is released. Comprehensive information on both access options is available on the Add Health Restricted-Use Data page.

Most U.S. Alzheimer’s disease (AD) research focuses on older adults, leaving little understanding of risk factors earlier in life across representative populations. Using nationally representative data from Waves IV and V of the National Longitudinal Study of Adolescent to Adult Health (Add Health), researchers investigated whether widely cited AD risk factors and blood biomarkers are associated with cognitive function before midlife. Risk factors were assessed using the Cardiovascular Risk Factors, Aging, and Incidence of Dementia (CAIDE) Score, a comprehensive algorithm integrating social, behavioral, and biological factors such as education, sex, age, cholesterol, blood pressure, body mass index, and physical activity. Biomarkers included APOE ε4 status, two neuropathological ATN markers (total Tau and neurofilament light chain [NfL]), and several inflammatory molecules and interleukins, including high-sensitivity C-reactive protein (hsCRP), IL-1β, IL-6, IL-8, IL-10, and TNF-α. Cognitive function was assessed using three tests: immediate word recall, delayed word recall and backward digit span.

The study included 4,507–11,449 participants in Wave IV (median age 28) and 529–1,121 in Wave V (median age 38). Roughly half were women, and most were White, with smaller proportions of Black and Hispanic participants. Findings showed that higher CAIDE scores were linked to lower performance on all three cognitive tests in Wave IV. In Wave V, higher levels of total Tau were linked to worse immediate word recall results. In both Wave IV and V, blood markers of inflammation (such as hsCRP, IL-6, IL-1β, IL-8, and IL-10) were also linked to lower cognitive scores across all tests. This study shows that known Alzheimer’s risk factors are linked to cognitive function in adults aged 24–44, not just older populations, emphasizing that prevention efforts in the U.S. should begin earlier.

To read the full article, please click the link below. For more important findings using Add Health data, visit the Add Health publications page.

Risk factors for Alzheimer’s disease and cognitive function before middle age in a U.S. representative population-based study. Aiello, Allison E. et al.The Lancet Regional Health – Americas, Volume 45, 101087

The following data are now available to Add Health contract researchers. 

The Rural-Urban Commuting Area (RUCA) Codes
Rural-urban commuting area (RUCA) codes classify U.S. census tracts using measures of population density, urbanization, and daily commuting. The data file including them is based on RUCA codes for census years 1990, 2000, and 2010. The rationale for and utility of acquiring RUCA codes, assigning them to census geographies in which Add Health respondents have resided over three decades. N=97,700

Add Health Sample Member Birth Records Database
Birth record data was collected from participating states for AHSM birth years, 1974-83. When these states provided birth data for all recorded births occurring during that time interval, an AHSM-specific subset was created using Link Plus, a statistical linkage software developed by the U.S. Centers for Disease Control and Prevention (CDC), Cancer Division. One participating state performed its own AHSM linkages and provided Add Health with the linked subset of births. Add Health then performed transformations on all of the original data from the participating states to create the categorical variables present in this release. N=2,750

Aircraft Noise Measures
Day-Night Level (DNL) Noise Exposures from 90 Major Airports data contains the estimation of aircraft noise measures around ninety major airports and aircraft noise proxies for approximately 900 additional airports. Merged with geopositioned/geocoded Add Health respondent locations over Waves I-VI, it also documents how the aircraft noise source data were acquired, as well as the protocol for quality controlling their assignment across waves. N=83,357

Equivalent Sound Level for a 15-Hour Day (LAEQD) Noise Exposures from 90 Major Airports data contains the estimation of aircraft noise measures around ninety major airports and aircraft noise proxies for approximately 900 additional airports. Merged with geopositioned/geocoded Add Health respondent locations over Waves I-VI, it also documents how the aircraft noise source data were acquired, as well as the protocol for quality controlling their assignment across waves. N=73,174

Equivalent Sound Level for a 9-Hour Night (LAEQN) Noise Exposures from 90 Major Airports data contains the estimation of aircraft noise measures around ninety major airports and aircraft noise proxies for approximately 900 additional airports. Merged with geopositioned/geocoded Add Health respondent locations over Waves I-VI, it also documents how the aircraft noise source data were acquired, as well as the protocol for quality controlling their assignment across waves. N=57,886

Proxies for Aircraft Noise from Other Airports: Airport Counts data contains the estimation of aircraft noise measures around ninety major airports and aircraft noise proxies for approximately 900 additional airports. Merged with geopositioned/geocoded Add Health respondent locations over Waves I-VI, it also documents how the aircraft noise source data were acquired, as well as the protocol for quality controlling their assignment across waves. N=166,195

Proxies for Aircraft Noise from Other Airports: Mean Distances data contains the estimation of aircraft noise measures around ninety major airports and aircraft noise proxies for approximately 900 additional airports. Merged with geopositioned/geocoded Add Health respondent locations over Waves I-VI, it also documents how the aircraft noise source data were acquired, as well as the protocol for quality controlling their assignment across waves. N=256,318

Proxies for Aircraft Noise from Other Airports: Mean Total Enplanements data contains the estimation of aircraft noise measures around ninety major airports and aircraft noise proxies for approximately 900 additional airports. Merged with geopositioned/geocoded Add Health respondent locations over Waves I-VI, it also documents how the aircraft noise source data were acquired, as well as the protocol for quality controlling their assignment across waves. N=250,740

Mortality Surveillance
Individual Vital Status and Underlying Cause of Death File, 2022
This file contains one record for each of the 20,745 Add Health sample members from Wave I. It provides the vital status of each sample member through 2022 as well as the National Death Index-provided underlying cause of death code in ICD-10 format for each decedent. The month and year of the most recent Add Health interview are provided for living sample members, while the month and year of death are provided for decedents. N=20,745

Ordered Cause of Death File, 2022
This file contains entity- and record-axis codes reported by the National Death Index (NDI) for each decedent in the Add Health sample through 2022. The file is arranged hierarchically, by axis code; therefore, each decedent may have multiple records depending on the maximum number of entity- and record-axis codes recorded by NDI. The sequence of the decedent’s records reflects the order in which the entity- and record-axis codes were reported in the NDI record. N=2,377

All Coded Causes of Death File, Including Entity-Axis Codes, 2022
This file contains all underlying cause of death and entity-axis codes appearing in the National Death Index (NDI) source file through 2022. Functioning as dummy variables, zero represents the absence of a code on the decedent’s death certificate, while one denotes the presence of one. N=706

All Coded Causes of Death File, Including Record-Axis Codes, 2022
This file contains all underlying cause of death and record-axis codes appearing in the National Death Index (NDI) source file through 2022. Functioning as dummy variables, zero represents the absence of a code on the decedent’s death certificate, while one denotes the presence of one.  N=706

Contextual Heterosexism Database-Phase 1
Contextual Heterosexism Database-Phase 1 (CHD1), further expands the collection of contextual data available to users of The National Longitudinal Study of Adolescent to Adult Health (Add Health) through the provision of state, county, and tract level measures from the Decennial Census of Population and Housing, American Community Survey (ACS), the Movement Advancement Project (MAP), Lax and Phillips (2009), Public Religion Research Institute (PRRI), Cooperative Election Study (CES), U.S. Religion Census, and Massachusetts Institute of Technology (MIT) Election Lab. These data include indicators of social policies, social climate, and confounding factors related to the study/measurement of structural heterosexism that correspond to Waves 3, 4, and 5. Some of these indicators are new to the Add Health contextual database and others were previously not available at all three of these waves. N=18,352

Current Add Health investigators can log in to the CPC Data Portal and use the “Request More Data” button to order these datasets. 

For more information on the CPC Data Portal, please visit the Frequently Asked Questions page. 

In honor of National Hispanic Heritage month, September 15 to October 15, Add Health would like to highlight some of the important discoveries our researchers have found thanks to our Hispanic American participants.

High educational aspirations, high expectations about attending college, high parent-child relationship quality, and high friends’ GPAs were found to be the most important predictors of both high school and college graduation among Hispanic adolescents. Friends’ lack of substance use was also a significant predictor for college graduation. (Chapin 2019).

Close social bonds with parents were found to be a very important protective factor against violent crime victimization for Latino adolescents and young adults. The especially close monitoring offered by immigrant Latino parents was especially protective for their adolescent and young adult children. (Lopez and Miller 2021).

Both Latino and Black Americans are more likely to experience unfair police treatment compared with Whites. Further, Latinos with medium brown, dark brown, or black skin tones are more likely to experience unfair treatment by police than those with lighter skin. The authors provide ideas for reducing the excess unfair treatment experienced by Latino and Black persons in American society, especially those with darker skin tones. (Finkeldey et al. 2022).

In their late 30s, Hispanic adults exhibit favorable patterns of physical functioning compared with White adults in the United States, with the favorable patterns being especially pronounced among Hispanic immigrants and second-generation Hispanics. These findings point to the generally favorable health of Hispanic immigrants and the children of Hispanic immigrants in the United States. (Touma and Hummer 2022). 

For more important findings, visit the Add Health publications page.