Dataset DOI: 10.5061/dryad.9p8cz8wzc

Description of the data and file structure

This dataset spans the period from August 16, 2016 to December 30, 2025, reflecting the full history of the UCI Consent-to-Contact (C2C) Registry from its launch through the current reporting period. The UCI C2C is a local, online potential participant registry established by the University of California, Irvine Institute for Memory Impairments and Neurological Disorders (UCI MIND) and the Institute for Clinical and Translational Science (UCI ICTS). The registry was created to enhance the efficiency of recruitment to clinical research studies at UCI, with a particular emphasis on "preclinical" Alzheimer's disease (AD) trials that enroll cognitively normal older adults meeting specific genetic or biomarker risk criteria, while also serving as a resource for any UCI investigator conducting IRB-approved human subjects research.

Following IRB-approved electronic informed consent, registrants first submit an email address and are subsequently invited to complete a secure REDCap-based enrollment survey. This survey captures self-reported demographic data (contact information, date of birth, sex, height, weight, race, ethnicity, education, and language), personal and family medical and psychiatric history, prescription medication use (cross-referenced against RxNorm for accuracy), lifestyle information (a physical activity scale and brief food frequency questionnaire), subjective cognitive function (Cognitive Function Inventory), and attitudes toward research participation (Research Attitudes Questionnaire). Registrants also indicate their willingness to be contacted about studies involving specific procedures, including cognitive testing, MRI, PET imaging, blood draws, lumbar puncture, genetic/saliva testing, and autopsy. Consistent with the registry's ongoing design, registrants are invited to renew their enrollment annually, at which point demographic and medication data are reconfirmed, and select measures (exercise, diet, CFI, and RAQ) are recollected longitudinally, allowing the dataset to capture repeated measures for many individuals over time.

In addition to individual-level registrant data, the dataset includes registry-level operational data collected continuously since launch, including enrollment activity over time (email sign-ups versus full survey completion), the type and relative effectiveness of recruitment strategies employed (earned media, community outreach, referrals, email campaigns, internet/social media, and other channels), and outcomes when registry data were used by independent investigator teams to recruit participants into specific IRB-approved studies. This includes detailed recruitment funnel data (numbers of registrants identified, reached, screened, deemed ineligible, declined, and enrolled) across the many studies the registry has supported over its operating history, including preclinical AD trials such as the Anti-Amyloid Treatment in Asymptomatic AD (A4) study.

In short, this dataset reflects a continuously operating recruitment-science effort at UCI, combining self-reported participant demographic, health, and attitudinal survey data with registry engagement and recruitment-effectiveness metrics, collected from the registry's inception through the current reporting period, to evaluate and refine strategies for improving enrollment into Alzheimer's disease clinical trials and other human subjects research.

Files and variables

File: c2c_20260903.csv

Missing data, blank cells, and non-response codes

The C2C is a web-based, self-report registry. All data are entered directly by registrants through a REDCap survey, and participation is voluntary at the item level. As a result, blank (empty) cells appear throughout the released files. A blank cell means no value was recorded for that variable in that survey instance; it does not mean zero, "no," or "not applicable." Blank cells arise for the following reasons:

1. Item skipped by choice. No survey item is required. Registrants may leave any item blank, including demographic, medical history, lifestyle, and willingness items. Some items also offer explicit non-response options, which are retained as codes rather than blanks (for example, sex = 77 "Prefer not to answer"; race = 77 "Refuse"; ethnicity = 7 "Refuse"; edu_band = 9 "Unknown"). Where an explicit refusal code exists, a blank indicates the item was skipped entirely rather than actively refused.
2. Incomplete or partially completed surveys. Registrants who began but did not finish the enrollment survey are retained in the dataset. Their records contain values for items completed before they stopped and blanks for all subsequent items. Blanks therefore tend to cluster in sections that appear later in the survey instrument.
3. Survey branching / conditional logic. Some items were only displayed when a prior response made them relevant, so they are legitimately blank ("not applicable") for other registrants. For example, ref_method_postcard is blank unless the registrant indicated a postcard referral in ref_method, and cancer- or neurology-related detail was only asked of registrants answering "yes" to cancer_dx or neuro_dx.
4. Longitudinal (renewal) instances. The file contains multiple rows per person where applicable (event_name distinguishes enrollment from annual renewal instances). Only a subset of measures is re-collected at renewal (exercise, diet, sleep, CFI, and RAQ; demographics and medications are reconfirmed). Variables not administered at a given event are blank for renewal rows. Users analyzing baseline characteristics should filter on the enrollment event; users analyzing repeated measures should expect blanks for non-repeated variables.
5. Instrument changes over the registry's history. The registry has operated continuously since August 16, 2016, and survey content has been revised over time. Items added after launch are blank for registrants who enrolled before the item existed; items later retired are blank for registrants who enrolled afterward.
6. Calculated and banded variables. Derived fields are blank when their source items were not sufficiently complete to compute a value. Specifically, calc_cfi_score and calc_raq_score are blank if the underlying CFI or RAQ items were not completed; calc_family_hx_ad is blank if the parental medical history items needed to derive the flag were not answered; and age_band and edu_band are blank when date of birth or years of education were not provided (edu_band = 9 is used where education was reported as unknown).
7. De-identification. Values were not suppressed to create blanks. Variables removed for privacy (including all geographic variables, raw age and education, and individual CFI, RAQ, and parental history items) were dropped as entire columns and are documented in the "Description of the data and file structure" section; they do not appear as blank fields.

Human subjects data

Data collected for this study were obtained through the UCI Consent-to-Contact (C2C) Registry, approved by the University of California, Irvine Institutional Review Board (IRB #20152494). All registrants provided electronic informed consent prior to submitting any personal information.
Prior to deposit, all data were de-identified by removing or recoding fields that could directly identify participants, including name, email address, phone number, and mailing address. Date of birth was recoded to age at time of enrollment, removal of free-text fields, and aggregation of rare demographic categories. All remaining data are self-reported by participants and do not constitute protected health information under HIPAA, as no data were obtained from medical records.
The dataset made available here reflects updated data while remaining consistent with the variables reported in Grill et al. (2018, J Alzheimers Dis 63:1055–1063), and contains only de-identified information. No data are included that could reasonably be used to re-identify an individual participant, either alone or in combination with other publicly available information.