7.0 data release

Overview

The 7.0 data release (on 5/13/2026) includes new data from events and sessions newly completed since the 6.0 release; Baby Teeth, RECOVER, and SIP substudies data; as well as updates to the data and data dictionary to address known issues and other found errors since the 6.1 release. Below are diff files showing all metadata changes as well as an overview of the data changes made in the 7.0 release. Detailed information about data level changes made in the 7.0 data release are available in the internal release notes, which are available to authenticated users on the NBDC Data Hub.

Sample

The 7.0 data release (on 5/13/2026) includes data from 11,860 participants, representing the full ABCD cohort (N = 11,880), except for participants who withdrew consent to share their data, and 16 events. Data are considered complete through the 5-year follow-up and nearly complete for the 6-year follow-up. Data from the 7-year and 7.5-year follow-up events are also included; however, data collection for events at and beyond the 6-year follow-up was still ongoing at the time of the data freeze (cutoff date: August 1, 2025). As a result, these events include only participants who had assented by the cutoff date. No data are included from events after the 7.5-year follow-up to ensure sufficient event completion.

The following table shows the number of participants per core study event:

Session/event ID Session/event label n
ses-00S Screener 11859
ses-00A Baseline 11860
ses-00M 0.5 Year 11371
ses-01A 1 Year 11220
ses-01M 1.5 Year 11076
ses-02A 2 Year 10989
ses-02M 2.5 Year 10255
ses-03A 3 Year 10516
ses-03M 3.5 Year 9584
ses-04A 4 Year 9792
ses-04M 4.5 Year 7219
ses-05A 5 Year 9050
ses-05M 5.5 Year 7149
ses-06A 6 Year 9539
ses-07A 7 Year 8564
ses-07M 7.5 Year 4871

The ABCD 7.0 data release also includes data from associated substudies, including Social Development, COVID-19, and newly added RECOVER and SIPS substudies. Some of these substudy assessments are conducted during the same visits as the main study, while others have independent event structures.

RECOVER (ses-REC) and SIPS (ses-SIP) are newly added substudy events introduced in the 7.0 data release.

The following table shows the number of participants with data per substudy event:

Substudy Session/event ID Session/event label n
COVID-19 ses-C01 COVID Wave 1 10380
COVID-19 ses-C02 COVID Wave 2 10377
COVID-19 ses-C03 COVID Wave 3 10367
COVID-19 ses-C04 COVID Wave 4 10357
COVID-19 ses-C05 COVID Wave 5 10347
COVID-19 ses-C06 COVID Wave 6 10009
COVID-19 ses-C07 COVID Wave 7 10006
Social Development ses-S01 SDev Wave 1 2424
Social Development ses-S02 SDev Wave 2 2128
Social Development ses-S03 SDev Wave 3 1942
Social Development ses-S04 SDev Wave 4 1907
Social Development ses-S05 SDev Wave 5 1896
Social Development ses-S06 SDev Wave 6 1684
Social Development ses-S07 SDev Wave 7 822
RECOVER ses-REC RECOVER 6009
SIPS ses-SIP SIPS 559

Curation & Structure

Curation

Some variables were renamed for the 7.0 release to correct errors from prior releases and to better align keyword usage within and across domains. The full list of renamed variables is provided in the mapping table below.

Structure

Data in the 7.0 release are provided in both tabulated and file-based structures (see Tabulated Data and File-Based Data.

The file-based directory structure has been updated: the concat directory has been renamed concatenated. All corresponding file paths reflect this change.

Core Study

ABCD (General)

Demographics

NAA mean scores were added in the ab_p_demo where the Native American Acculturation items were asked to parents, the new item is ab_p_demo__ntvam_mean.

Friends, Family, & Community

New FC tables

There were new tables included in the 7.0 Data release

Acculturation Gap [Youth]
fc_y_ag
tabulated
Education & Vocation [Youth]
fc_y_ev
tabulated
UCLA Loneliness Scale [Youth]
fc_y_lone
tabulated

New NAA summary score

There is a new summary score fc_y_naa_mean for the Native-American Acculturation Scale (NAAS).

Youth Education & Vocation school type level mislabeling

There is a level labeling issue for item fc_y_ev__edu_002__v01 asking about the specific type of school program in which the youth is enrolled in the fc_y_ev table such that the label for response option 3 displays as “Certificate, credential, or licensing program (including occupational, technical, or vocational) 18, Military Service Academy (USMA, USNA, USAFA, USCGA, USMMA)”. This issue only exists in the release dictionary, participants were only shown the correct label (“Certificate, credential, or licensing program (including occupational, technical, or vocational)”) for option 3 and the label and option for 18 were displayed separately.

Genetics

ABCD 7.0 introduces whole genome sequencing data for 8,710 individuals (8,877 samples including technical replicates), generated via Illumina NovaSeq at 30x coverage. See the WGS documentation for methodological details and quality control procedures.

For microarray data, we have adapted the procedure for calculating the genetic relatedness matrix to reduce inflated values exceeding 1. This was achieved by using 7 (rather than 2) principal components and truncating minor allele frequencies to X when calculating ancestry corrected allele frequencies in PC-AiR. Imputation of microarray data is now performed on the r3 rather than r2 reference panel. A further 6 individuals were included as part of the microarray data resource.

The most significant update for the ABCD genetics 7.0 data release is the inclusion of 8,877 samples of Illumina short read 30x whole genome covering 8,710 individuals. A brief overview of the key methodological steps is provided below. Blood and saliva samples were collected across participants roughly equally, with blood preferred for sequencing where available. DNA extraction and biospecimen storage were carried out at RUCDR (later rebranded as SAMPLED). Sequencing was performed on Illumina NovaSeq instruments at 30x coverage, with reads aligned to the hg38 reference genome. Per-sample variant calling was performed following GATK best-practices, and population-level VCF files were generated by joint genotyping across all samples for single nucleotide variants and small indels. Variant quality was assessed using Variant Quality Score Recalibration (VQSR), and all released files include a FILTER field where ‘PASS’ denotes high-confidence variant calls. Sample-level quality control was anchored by cross-validating WGS data against existing microarray genotype data on the same individuals, with discordant samples flagged for resequencing. A total of 95 samples were removed on this basis, yielding the final set of 8,877 samples covering 8,710 individuals (including 168 technical replicates). Full methodological details — including library preparation, alignment parameters, joint genotyping strategy, and QC thresholds — will be provided in an upcoming data resource paper.

Linked External Data

SEDA Data

Beginning with the 7.0 ABCD release, SEDA data are provided longitudinally across waves, rather than at baseline only, and are available in multiple formats (e.g., by assessment wave, school year, and month) to support different analytic approaches. Data linked by assessment wave can be downloaded within the tabulated datasets, whereas data linked by school year and month must be downloaded as file-based data through the NBDC Data Hub.

Satellite-based O3 Measures

There was an error during curation of the le_l_lo3 table for the 6.0 release, in which the new variable names for three average ozone variables (le_l_03__addr1__03_mean__2016, le_l_03__addr2__03_mean__2016 and le_l_03__addr2__03_mean__2016) were incorrectly mapped. This naming issue has now been corrected for the 7.0 release.

Mental Health

Behavioral Inhibition / Behavioral Approach System Scales

There was a previous error with the summary score mh_y_bisbas__bis_sum computation as the variable mh_y_bisbas__bis_005 was not reverse-coded. The data and score algorithm have been corrected and updated for the 7.0 Release.

Early Adolescent Temperament Questionnaire

Summary scores for some subscales (Activation, Aggression, Attention, Depressive Mood, Inhibitory Control, Shyness, and Surgency) of the mh_p_eatq measure were mistakenly reverse coded in the score algorithm despite already being reverse coded at the item level. Users should compute their own means when using 6.0 data and the issue has been fully corrected for the 7.0 Release.

Family History summary scores

New summary scores and derived items were added to the mh_p_famhx table. The _indicator items summarize over checkbox items in the alc and drg subscales to create binary items indicating if any problem in those subscales were endorsed for each family member. This is intended to allow easier data use across all subscales as most other scales use binary items in the original questions for each family member. Additionally, _indicator and _score items were added for each subscale to give summary information across both parents. Details on these scores can be found in the ABCDscores documentation

KSADS

KSADS Item-level Data

The item-level KSADS data have been processed and included in the 7.0 Release. Item level data refers to the answers to each question asked in the KSADs interview. In some cases questions map one-to-one with specific symptoms, but in other cases several questions might map to the same symptom or there may be a combination of questions that need to be answered in a certain way to indicate the presence of a symptom. Users should use these data with caution, recognizing that we have not been able to fully QC all of the item/question data. In releasing these data, we attempted to harmonize the items between KSADS 1.0 and 2.0, a considerable amount of work. We have done our best to match questions across versions, but there are still items about which we have questions and we will work with KSADS over the upcoming year to address these remaining questions. Please see here for a file that lists the question numbers that we matched across versions. We can only provide question numbers and not the questions themselves due to copyright protections, but interested people could reach out to KSADS directly to request further information.

Item level data was curated and made available for the KSADS tables starting at the 7.0 release. This combined KSADS1 and KSADS2 items into single variables in the release dataset wherever 1:1 relationships existed across the two versions. Mapping from release names to KSADS questions IDs is available below. Data users will need to request data dictionary information from KSADS directly for more information on question details for each question ID number.

KSADS Diagnosis and Symptom Data

The KSADS diagnosis and symptom score data were re-ingested and cleaned for the 7.0 release. Cases of accidental participant swaps and incorrect time point assignments were addressed, though some cases were not resolvable for 7.0 and thus are not included. Further corrections will be made for the 7.1 release update. Additionally, missingness coding was updated so score data in modules that were not given will be coded as 555 and any scores that a participant did not branch into based on item-level responses are coded as 0.

KSADS Symptom Mean Summary Scores

A new set of summary scores was added for all KSADS modules. The experts selected the relevant KSADS symptoms to include an overall symptom mean measure. Documentation of the algorithms for these scores can be found in the ABCDscores documentation.

KSADS – Suicidality

Previously, there was an issue with skipped item responses about past suicidal ideation/behavior reports (any items starting with the prefix mh_y_ksads__suic__past) from the mh_y_ksads__suic table. This was addressed for the 7.0 Release. Starting at the 5-year follow-up event, youth were no longer supposed to be asked about past suicidality and RAs were told to input “no” in order to allow the program to move forward, but there were some cases that had data entered. All data for these events and items have now been updated to 555 “not administered” to reflect this protocol change. Do not use past suicidality data from the 5-year visit and beyond.

KSADS – Bipolar Disorders

The duplicate variable mh_p/y_ksads__bpd__irrit__expl__pres_sx__v01 was removed as this item only indicated branching logic within the KSADS module and did not represent a true symptom score.

Life Events

Life Events Summary Scores

Previously, there was an issue with the Life Events summary scores due to the validation algorithm with missing data. This issue is fixed for the 7.0 Release.

Life Events Item Coding

The levels for the items mh_y_ple__severity_032 and mh_p_ple__severity_032 were previously incorrectly coded as values 1-4 instead of 0-3 causing an inflation of values in all summary scores that item is included in. The item coding has been corrected in the 7.0 Release.

Prodromal Psychosis Scale

Two new summary scores were added to the mh_y_pps table: mh_y_pps__dist__pers_score, mh_y_pps__dist__curr_score. These scores relate to the scores in the new table mh_y_ppsss table and are used in the inclusion criteria for the SIPS substudy.

Static Prodromal Psychosis Scale Summary Scores

There was a new table Static Prodromal Psychosis Scale Summary Scores (mh_y_ppsss) included in the 7.0 Release.

Youth Self Report Summary Scores

The ABCD Study removed 7 items (YSR items 18, 88, 91, 98, 106, 109, and 110) from the youth administration to reduce participant burden and alleviate redundancy in the overall protocol. Only 3 of the items removed (18, 91, and 110) are problem items included in the YSR scales Anxious/Depressed, Thought Problems, and Other Problems, respectively. Starting with the 7.0 release, ABCD has decided to compute summary scores for these scales as sum of the remaining items, which is in agreement with ASEBA guidance of ensuring no more than 8 problem items are missing to still be able to compute summary and T-scores. No imputation of the missing items was performed because these 3 items have typically very low base rates of response and imputation could be inaccurate. Since summary scores for these three scales may underestimated, caution is recommended in interpretation of group means, particularly in analyses involving a combination of ABCD data and other datasets with YSR data.

ABCD has also updated the overall approach to scoring with regards to missingness. Missingness across all problem items is evaluated and if more than 5 additional problem items, beyond the 3 intentionally skipped by ABCD, are missing, any subscale with 1 or more items missing will not be scored, but any complete subscales will still be scored. Additionally, even if less than 8 problem items are missing, any scales with more than 20% missingness of expected items within the scale will not have computed scores. Please be aware of this during analyses and comparisons to other studies using the YSR.

Neurocognition

Little Man Task

For nc_y_lmt, the previous data were incorrectly entered in percentage form and some data contain zero, which might seem implausible. These have been fixed in 7.0.

NIH Toolbox

During the generation of the file-based data for neurocognitive tasks, there was an issue that caused some file-based data not to be generated. This issue primarily affected the ses-06A timepoint and NIH Toolbox tasks. About 1500 participant sessions are thus missing the file-based data but do have tabulated data. It is planned to restore these participants’ file-based data for the 7.1 release.

Novel Technologies

Fitbit raw data files

The known issues from 6.0 have been fixed for 7.0 Release.

The qc_300min quality control flag in the sleep daily output was computed using min_total_slp > 299 rather than min_total_slp >= 300. Because min_total_slp is a floating-point value, the > 299 condition passes any value strictly greater than 299, meaning days with fewer than 300 valid sleep minutes (e.g., 299.5) may incorrectly pass QC. Users relying on this flag to enforce a 300-minute minimum should be aware that affected records may be present in their data.

Sleep daily and weekly summary outputs were not filtered to exclude pre-session records which can be identified by day = 0 and wk = 0. Users should apply filter(day != 0) before using daily data and apply filter(wk != 0) before using corresponding weekly summaries.

Records in fitbit_raw_metrics belonging to session_id ses-08A were incorrectly labeled with session_id = 08A.

Fitbit summary scores

Fitbit summary scores have been included in the 7.0 Release.

Physical Health

Blood Pressure

Previously, there were some errors with the summary scores counting invalid inputs like 0, and there were some outliers in the data. These issues have been addressed in 7.0 Release data

Blood Results

Data errors in the blood results were discovered in the 6.0 dataset. Most outliers or errors were due to duplicate values being stored for a single sample, leading to unusually high values. All values that were correctable for 7.0 were fixed, but some outliers still remain. ABCD will continue to work to correct the remaining issues in a future release.

COVID Survey

In the table ph_p_covid it was discovered that there were many date entries throughout the form that were implausible, such as dates from before or after the COVID pandemic. Experts worked to fix data where correctable, e.g. 2002 being corrected to 2022, but dates that were not resolvable were removed from the dataset.

Munich Chronotype Questionnaire Score update

The algorithm for the Munich Chronotype Questionnaire (MCTQ) free day and school day count items has been updated such that if participants endorse they no longer go to school or work, the free day count will = 7 and the school/work day count will = 0 rather than missing as they were in prior releases. This should help improve missingness in other scores that were dependent on these fields having values.

Medications Inventory & Developmental History Summary Scores

Summary scores have been developed for the Medications Inventory (ph_p_meds & ph_y_meds) and Developmental History (ph_p_dhx) tables. These include National Library of Medicine (NLM) API mapped fields from RxNorm codes, estimated use category mappings (estuse), and summarized true/false indicators to support ease of analysis. Estimated use category mappings derived from the parent-reported data were applied to youth-reported medications, with approximately 2% of medications currently unmapped. Future updates to the mapping framework may be incorporated in future releases.

Pain Questionnaire

Previously, variable names were swapped for four items in the ph_y_pq table. The left and right variable names, but not labels, were reversed for the palm and top of foot items in this measure. This has been fixed in 7.0 Release data.

New derived items

New derived birthweight item

A new derived item was added to ph_p_dhx to combine pounds and ounces into a single birthweight item.

New derived height scores

New derived items were added to the ph_p_anthr table to combine feet and inches into a single height item for both mother and father.

Substance Use

Hair Toxicology Results Coding Fix

In the 6.0 and 6.1 release, some hair toxicology results were incorrectly coded as 1/2 instead of 0/1 for negative/positive, respectively. All results coding has been updated in the 7.0 release and users of 6.0 data should be aware of this issue.

KSADS

KSADS Items

The KSADS Alcohol Use Disorder (su_p_ksads__aud) and Drug Use Disorder (su_p_ksads__dud) item level data and symptom means have been included in the 7.0 Release.

Item level data was curated and made available for the KSADS tables starting at the 7.0 release. This combined KSADS1 and KSADS2 items into single variables in the release dataset wherever 1:1 relationships existed across the two versions. Mapping from release names to KSADS questions IDs is available below. Data users will need to request data dictionary information from KSADS directly for more information on question details for each question ID number.

KSADS Diagnosis and Symptom Data

The KSADS diagnosis and symptom score data was re-ingested and cleaned for the 7.0 release. Cases of accidental participant swaps and incorrect time point assignments were addressed, though some cases were not resolvable for 7.0 and thus are not included. Further corrections will be made for the 7.1 release update. Additionally, missingness coding was updated so score data in modules that were not given will be coded as 555 and any scores that a participant did not branch into based on item-level responses are coded as 0.

KSADS Symptom Mean Summary Scores

A new set of summary scores was added for all KSADS modules. The experts selected the relevant KSADS symptoms to include an overall symptom mean measure. Documentation of the algorithms for these scores can be found in the ABCDscores documentation.

Substance Use Density, Storage, and Exposure

Substance use density items were previously open response which allowed for entries that fell outside of the expected range (i.e., greater than 7 days a week or more than 24 hours a day). The response option was changed to dropdown response on 9/1/2023 and responses that fell above that range were removed.

In the table su_p_des there were outliers found in the follow-up questions ending in __01 and __02 asking “How many hours per day” or “How many days a week” that were above 24 or 7, respectively. These outliers have been removed from the dataset for the 7.0 release.

Substance Use Summary Scores

New dynamic and static substance use (SU) summary scores have been introduced in the 7.0 Release to harmonize substance use indicators across multiple measures (Substance Use Interview, TLFB, and Mid-Year Phone Interview; Lisdahl et al. (2018)). Harmonizing across multiple measures has been shown to more comprehensively capture substance rates across the ABCD sample (Sullivan et al. (2022); Sullivan et al., Under Review). Dynamic scores (su_y_dyn__) summarize participant-by-event indicators of lifetime use across substances, while static scores (su_y_stc__) summarize onset information, including age and visit of first reported use. These scores integrate information across instruments to generate consistent indicators of reported use and onset. Alcohol, Cannabis, and Nicotine use variables include composite variables of low-level use (i.e., sipping or puff), full standard use, or combined low and full use for each substance. Full substance use variables including low level use are indicated within the variable name (i.e., “lowuse”). Further, composite variables across all substance use are computed (i.e., __su__) as well as substance use other than alcohol, nicotine, or cannabis (i.e., __su__oth__).

  • Lisdahl, K. M., Sher, K. J., Conway, K. P., Gonzalez, R., Feldstein Ewing, S. W., Nixon, S. J., Tapert, S., Bartsch, H., Goldstein, R. Z., & Heitzeg, M. (2018). Developmental Cognitive Neuroscience, 32, 80–96. https://doi.org/10.1016/j.dcn.2018.02.007
    Lisdahl et al. (2018)
  • Sullivan, R. M., Wade, N. E., Wallace, A. L., Tapert, S. F., Pelham, W. E., Brown, S. A., Cloak, C. C., Ewing, S. W. F., Madden, P. A., Martz, M. E., Ross, J. M., Kaiver, C. M., Wirtz, H. G., Heitzeg, M. M., & Lisdahl, K. M. (2022). Drug and Alcohol Dependence Reports, 5, 100120. https://doi.org/10.1016/j.dadr.2022.100120
    Sullivan et al. (2022)
  • Sullivan, R. M., Wallace, A. L., Shankula, C., Celhay, O., Ziemer, L., Smith, C., Berman, S., Linkersdörfer, J., Martz, M., Courtney, K., Jacobus, J., Tapert, S., Heitzeg, M., Lisdahl, K. M., & Wade, N. E. (2026). https://doi.org/10.31234/osf.io/eh8ba_v1
    Sullivan et al. (2026)

Imaging Data

MRI info

Device Serial Number

  • New hash values to replace original DeviceSerialNumber values to be consistent for BIDS and tabulated data
  • Previously used a different replacement values for BIDS and tabulated data
  • In the ABCD 6.0 release, there was a mismatch between the tabulated and BIDS (sourcedata and rawdata) in the anonymized MRI scanner serial numbers. In the BIDS data, the sidecar json files contain a field “DeviceSerialNumber” of the form ‘anon12da’, which differs from the corresponding variable in the tabulated data, ab_g_dyn__design_mr__serial, which has the form “HASHd7cb4c6d”. Within each datatype (BIDS and tabulated), the values used are internally consistent. In the ABCD 7.0 release, the form of the serial number anonymization has been standardized across these data types.

Software Version

  • New categorical variable mr_y_adm__info__dev__sftw__maj_ver reduced number of categories limited to major version changes and minor bug fixes that may affect the data

All imaging

  • ~50 visits that differ for across all modalities related to recovery of previous processing failures and some re-evaluation of raw QC

rsfMRI

  • Tabulated data now based on concatenated data
    • Most important difference:
      • Previously for tabulated data, minimally processed voxelwise data were additionally processed (i.e., motion and global signal regression, interpolation of censored frames, bandpass filter) followed by calculation of average ROI timeseries and ROI-ROI correlations
      • For concatenated data, average ROI timeseries were calculated from minimally processed data, followed by additional timeseries processing (regression, interpolation, bandpass), and calculation of ROI-ROI correlations
    • Relatively minor difference:
      • Previously for tabulated data, correlations were calculated for each scan and then averaged
      • For concatenated data, processed timeseries data were concatenated across scans prior to calculation of ROI-ROI correlations
  • Preprocessing for concatenated timeseries data
    • Eliminated additional application of respiratory filter to motion time series affects motion metrics (e.g., mean FD) and censoring resulted in relatively small differences to ROI-ROI correlations
    • In 6.0, respiratory filter was applied to motion tseries before concatenation of time series data, and the respiratory filter was being applied again as part of the pre-processing of the time series before calculation of ROI-ROI correlations. This second application of the respiratory filter tended to reduce the estimated mean motion and reduce the number of censored frames. In 7.0, this was changed so that there was no second application of the respiratory filter. This has a small effect on the calculated correlations.
    • Calculation of mean FD (uncensored) now includes all scans even those that do not have enough valid time points after censoring to be included in the calculation of ROI-ROI correlations matches how uncensored mean motion was previously calculated for tabulated data
  • Concatenated data now includes additional motion-related variables e.g., mean FD censored, mean dx, dy, dz, rx, ry, rz

dMRI

  • In 6.0, concatenated white matter (wm) right hemisphere (rh) matrices had wrong number of rows mismatched to vol_info, so visit IDs scrambled
  • A subset of the concatenated data files, specifically vertexwise DTI measures in peri-cortical white matter, have a different number of observations (i.e., participant-session rows) for the left hemisphere and the right hemisphere (30392 rows in left hemisphere and 30393 rows in right hemisphere). Because of the inadvertently mismatch rows, these data files should not be used. Concatenated vertexwise DTI data files with the following file stems were affected: “fa_wm”, “ld_wm”, “md_wm”, “td_wm”. This has been fixed in 7.0

SST

  • “Stop too early” trials now reclassified as go affects a large subset of visits
  • SST runs contained “Stop too early” (STE) trials are defined as trials in which the participant made their button press response during a stop trial before the stop signal would have been presented. In the version of the SST task used since August 2021, when STE trials occurred, the stop signal was then not presented on that trial, and the subsequent trial was replaced by another “Stop” trial, thus reducing the number of Go trials. For time series modeling, STE trials are now modeled as Go trials.

MID

  • There were minor changes to the ROI beta weights in tabulated data, especially dsk average of runs.
  • These are the changes between the tabulated data prepared for 6.0 vs 7.0: corrected event timing offset for GE scanners, modified event timing variables, corrected or excluded E-Prime timing errors for GE scanners.
  • There was previously a discrepancy in how stimulus times were modeled relative to the end of the calibration/dummy volumes in the image acquisition that affects GE scanners. The issue is related to how the E-prime tasks were programmed for GE scanners that resulted in an unexpected timing offset of 800 msec or less.
  • We modified our code for extracting event timing information from ABCD E-prime files (https://github.com/ABCD-STUDY/abcd_extract_eprime.git). Rather than rely on the first fixation event (e.g., CueFix.OnsetTime) the modified code now uses the timing data from the initial and or final trigger events (e.g., GetReady.RTTime) to determine the reference time that represents the start of the first non-dummy image volume. We also modified the number of initial volumes discarded prior to task fMRI time series analysis for GE scanners, with 4 volumes removed for GE DV26 and 15 volumes removed for GE DV26 and later.

Post QC

  • ~100-900 visits with differences, depending on modality
  • Revisions to previous scores, mostly for FreeSurfer QC

Missing data on Siemens XA30

In the ABCD 7.0 data release, anatomical scans (T1w and T2w) and field map scans (for both dMRI and fMRI) were inadvertently excluded from the sourcedata and rawdata BIDS datasets for the 889 scan sessions collected on Siemens scanners with software version XA30. Derivatives (mmps_mproc and freesurfer), concatenated data, and tabulated data are not affected. The missing scans will be added to the sourcedata and rawdata for the 7.1 patch release.

Substudies

Baby Teeth

Beginning at the initial (baseline) appointment of ABCD, researchers collected naturally shed deciduous (“baby”) teeth from participants, the rationale, procedure, and analysis of which have been described in detail (Uban et al. (2018)). Accordingly, our goals were to analyze various environmental neurotoxicant exposures in a sample of these teeth via laser ablation–inductively coupled plasma mass spectrometry (LA-ICPMS). The data from 448 teeth are provided in the 7.0 Release data and documentation is provided here.

RECOVER

The RECOVER data have been processed and included in the 7.0 Release. All survey questions were completed by the parent or caregiver. This includes surveys marked as “youth”. As such, the table information in the data documentation is all listed under “parent”/_p_, as the caregiver completed both the surveys about themselves and the surveys about the youth. For measures that are asked about both the youth and caregiver, the caregiver directed table ends with the suffix cg in the table name. The antibody test was completed by the youth directly, which is why it is marked as a youth table.

Social Development

New tables

Social Development substudy has the following new tables:

Parental Incarceration [Parent]
sdev_p_incar
tabulated
Gang Membership and Activity [Youth]
sdev_y_gma
tabulated
Parental Incaarceration [Youth]
sdev_y_incar
tabulated
Police Interaction [Youth]
sdev_y_pol
tabulated

Age item outliers

Age items in the Social Development substudy previously contained implausible values, such as participants entering the year an event occurred rather than the participant’s age at the time, that have not been filtered out. This issue affected all questions asking “How old” in the following tables (sdev_p_rd, sdev_y_fa, sdev_y_rd, sdev_y_vict). These issues were corrected for the 7.0 release by creating new __dk fields to capture all 98 and 99 non-missing responses, such as “Don’t know” and “Decline to answer”, and all other impossible age values were removed.

Structured Interview for Psychosis Risk Syndromes (SIPS)

The SIPS data have been processed and included in the 7.0 Release. This single-timepoint study collected separate interviews from youth and caregivers. The substudy includes a clinician-administered interview, the Structured Interview for Psychosis Risk Syndromes (SIPS), administered independently to caregivers and youth. Ratings reflect clinician determinations, finalized through consensus during supervision. The substudy contains the following modules: Unusual Thought Content, Suspiciousness, Grandiose Ideas, Perceptual Abnormalities, Disorganized Communication, Dysphoric Mood, Global Assessment of Functioning, as well as modules assessing whether youth met criteria for Schizotypal Personality Disorder and whether the youth met criteria for any psychotic disorder or psychosis risk syndromes. Users should consult the data dictionary for details on coding conventions. Data were collected with funding from NIMH grants K23MH121792 and R01MH139880.

Key references:

References

Aas, I. M. (2011). Annals of General Psychiatry, 10(1), 2. https://doi.org/10.1186/1744-859X-10-2
Lisdahl, K. M., Sher, K. J., Conway, K. P., Gonzalez, R., Feldstein Ewing, S. W., Nixon, S. J., Tapert, S., Bartsch, H., Goldstein, R. Z., & Heitzeg, M. (2018). Developmental Cognitive Neuroscience, 32, 80–96. https://doi.org/10.1016/j.dcn.2018.02.007
Miller, T. J., McGlashan, T. H., Rosen, J. L., Cadenhead, K., Ventura, J., McFarlane, W., Perkins, D. O., Pearlson, G. D., & Woods, S. W. (2003). Schizophrenia Bulletin, 29(4), 703–715. https://doi.org/10.1093/oxfordjournals.schbul.a007040
Sullivan, R. M., Wade, N. E., Wallace, A. L., Tapert, S. F., Pelham, W. E., Brown, S. A., Cloak, C. C., Ewing, S. W. F., Madden, P. A. F., Martz, M. E., Ross, J. M., Kaiver, C. M., Wirtz, H. G., Heitzeg, M. M., & Lisdahl, K. M. (2022). Drug and Alcohol Dependence Reports, 5, 100120. https://doi.org/10.1016/j.dadr.2022.100120
Sullivan, R. M., Wallace, A. L., Shankula, C., Celhay, O., Ziemer, L., Smith, C., Berman, S., Linkersdörfer, J., Martz, M., Courtney, K., Jacobus, J., Tapert, S., Heitzeg, M., Lisdahl, K. M., & Wade, N. E. (2026). PsyArXiv. https://doi.org/10.31234/osf.io/eh8ba_v1
Uban, K. A., Horton, M. K., Jacobus, J., Heyser, C., Thompson, W. K., Tapert, S. F., Madden, P. A. F., & Sowell, E. R. (2018). Developmental Cognitive Neuroscience, 32, 97–106. https://doi.org/10.1016/j.dcn.2018.03.005