Structural and functional MRI dataset from the Adolescent Health and Development in Context (AHDC) study in Columbus, Ohio

# README Structural and functional MRI dataset from the Adolescent Health and Development in Context (AHDC) study in Columbus, Ohio with 2 waves of data

Authors: Baldwin M. Way (Principal Investigator), Christopher R. Browning, Dylan D. Wagner, Jodi L. Ford, Bethany Boettner, Ping Bai.

Study Overview
These data were collected as part of a longitudinal study of adolescent health and well-being collected in Columbus, Ohio (Adolescent Health and Development in Context Study). The larger goals of the project (R01DA042080) were to understand how geospatial exposures predicted substance use. More specifically, Specific Aims 1a & 1b were to longitudinally and cross-sectionally determine how neural function and structure is reshaped by EtV in the community (1a) and substance use (1b). Specific Aim 2 was to use baseline as well as longitudinal neural changes to predict subsequent substance use and identify neural mediators. Specific Aim 3 was to identify risk and resilience factors that alter the effects of community EtV on the neural embedding of EtV as well as the neural prediction of substance use outcomes. The participants in this longitudinal neuroimaging study were recruited from the Adolescent Health and Development in Context study (Boettner, B., Browning, C. R., & Calder, C. A. (2019). Feasibility and validity of geographically explicit ecological momentary assessment with recall‐aided space‐time budgets. Journal of Research on Adolescence, 29(3), 627-645.).

Inquiries about this dataset should be directed to: [e-mail]
This dataset is licensed under the Creative Commons Zero (CC0) v1.0 License.

Study Area
The study area is a contiguous space within the Interstate 270 loop outerbelt freeway, encompassing a majority of the city of Columbus as well as several suburban municipalities.

Sampling
The sampling frame was based on a combination of a vendor-provided list of households in the study area with high probability of meeting eligibility criteria and directory data from public school districts represented in the study area. Households were mailed a letter or postcard describing the study, followed by interviewer calls to the household to solicit participation in the study. Once eligibility was confirmed with the household, one randomly selected youth aged 11-17 and one primary caregiver (English speaking) were recruited to participate in the study.

The racial/ethnic identity of the first wave of the AHDC study was 1,405 youth with 47% white, 38% Black, 5% Hispanic/Latino, 8% multiracial, and 2% Asian. The sample closely approximates the population in the study area with respect to household income of families with children and youth racial/ethnic composition, with the exception that the AHDC sample has a somewhat higher percent of youth who identified as
Black compared with the 2009-2013 American Community Survey (ACS) estimates of the area.

For Wave 3 (the first neuroimaging wave), in addition to the participants recruited from the original sample, a refresher sample was recruited. These participants were recruited from within the families of the original sample (i.e. siblings) as well as using the same methods of recruiting the initial wave from low-income census tracts as well as tabling at schools in these tracts.

Study Design
The study employs a prospective cohort design in which the data on youth and caregivers were collected at multiple time points. The Wave 1 field period began in spring 2014 and was completed in summer 2016. Wave 2 was conducted between January and December 2016. Wave 3 (the first imaging wave) was conducted between July of 2018 and March of 2020, concluding with the cessation of in person activities due to COVID-19 related restrictions on in-person activities. Wave 4 was run between July and October of 2020. Wave 5 was run between March and July of 2021. Wave 6 (the 2nd imaging wave) was run between May of 2022 and January of 2024.

Study Procedures
Within each wave, participant data were collected over a weeklong period. An Entrance Survey with both a focal youth and his or her caregiver was followed by a seven-day smartphone-based Global Positioning System (GPS) tracking and Ecological Momentary Assessment (EMA) data collection period (EMA Week), and either a final Exit Survey at the end of the week (Waves 1 and 2) or a session at the Center for Cognitive and Behavioral Brain Imaging at the Ohio State University (Waves 3 and 6). Waves 4 and 5 were slightly different due to restrictions on in-person activity. Wave 4 consisted of a phone interview and online survey that was completed remotely with participants downloading an app on their phone for responding to EMAs and GPS tracking. Wave 5 only consisted of an online survey and the responding to EMAs with GPS tracking.

The Entrance Survey was collected at the initial in-home visit with adolescent participants and their caregivers. It included a wide range of measures across social, economic, psychological, health, and behavioral domains. Both adolescent and caregiver participants reported on geographic location of and experiences at routine activities (e.g. school, work, church, stores, relative’s house).

The real-time Ecological Momentary Assessment (EMA) surveys were collected via self-administered survey on project-provided smartphones. The study phones also passively collected GPS spatial coordinates during the seven-day EMA collection period. Youth respondents were prompted up to five times a day, and asked to report on their location, network partner presence, risk behaviors such as substance use, mood, surrounding social climate, and sleep patterns.

Waves 1 and 2: A second visit, the Exit Survey, gathered follow-up information about the EMA week. The youth completed an interactive Space-Time Budget with the interviewer to collect detailed activity data on five days – the three most recent weekdays and two weekend days. The processed GPS data results in summarized stationary and travel periods during those five days, along with activity types and network partner presence. Concurrently, caregivers completed a self-administered survey about perceptions of
social climate and safety in their neighborhood and at other routine locations.

Waves 3 and 6: The second visit at the conclusion of the week of GPS tracking and EMA sampling was conducted at the Ohio State Center for Cognitive and Behavioral Brain Imaging. Participants completed an initial battery of questionnaires before scanning as well as had the option of providing a hair sample for cortisol or substance use measurement and blood sample for measurement of immune related markers. Participants also completed questionnaires after the scan.

Participants
309 youths participated in the initial home interviews in Wave 3. 290 of these youths came to the imaging center and 271 adolescents were successfully scanned. Of these 271, 158 were in Wave 1 of the AHDC study, while 113 were part of the refresher sample and were thus new to Wave 3.

For Wave 6, there were 144 individuals who came to the imaging center and 120 were successfully scanned. Of these, 110 were also scanned at wave 3, while 10 of these were individuals who were scanned for the first time.

MRI Tasks
In the first wave of imaging data (2018 to 2020; Wave 3 of the AHDC parent study), the task sequence was the same for all youths. The time of each run is listed after each and then in parentheses is the number of subjects after quality control checks (e.g. motion).
1. MPRAGE: 6:58 min (n = 249)
2. T2: 3:36 min
3. Resting State Scan (eyes open, rest): 5 min
4. Emotional Faces Task (Surprise, Angry, Fear, Neutral): 4:30 min x 2 runs (n = 214)
5. Cue Reactivity Task (Food, Marijuana, Flavored E-Cigs, Alcohol, and Outdoor images): 5:40 min x 2 runs (n = 215)
6. DTI: 6:55 min
7. Resting State Scan (eyes open, rest): 5 min
8. Field Map: 1:33 min
9. Monetary Incentive Delay Task: 5:23 min x 2 runs (n = 207)
10. Working Memory Task: 4:51 min x 2 runs (n = 183)
These latter two tasks used the same Eprime script as used in the ABCD study.

In the second imaging wave run between 2022 and 2024 (Wave 6 overall), there was a slight change to the task order for all participants in order to reduce the probability of youths falling asleep during the first resting state scan. The scan order for the second wave of imaging data was T1, T2, Emotional Faces Task, Cue Reactivity, Resting State 1, MID, Resting State 2, Field Map, Nback task.

***Caution*** This dataset is for research purposes only. The data have been anonymized, and users must not perform analyses aimed at re-identifying individual subjects.

Acknowledgements. We are grateful to all of the youth and their caregivers who participated in the study. The Adolescent Health and Development in Context study (Waves 1 and 2) was funded by the National Institutes for Drug Abuse (R01DA032371; Browning, PI) as well as the Eunice Kennedy Shriver National Institute on Child Health and Human Development (Boettner, R03HD096182; Calder, R01HD088545; Hayford, the Ohio State University Institute for Population Research, 2P2CHD058484), and the William T. Grant Foundation). Participants for the imaging data (Waves 3 and 6) were recruited from this sample, which was generously supported by a grant from the National Institutes of Drug Abuse (R01DA042080; Way, PI). There were two waves of data collected during COVID (Waves 4 and 5) that were funded by a supplemental grant from the National Institutes of Drug Abuse (DA042080-03S1; Way, PI). Assay of head hair samples for cortisol during the imaging waves (Waves 3 and 6) was funded by a grant from the John Templeton Foundation (ID: 61803; Way, PI). Head hair cortisol and salivary cortisol collection and assays for Waves 1 and 2 were funded by R21DA034960 (Ford, PI).