Plan analyses using CTDA data

Understanding CTDA data: Standard variable names and values

PACT/R datasets include individual-level data from prospective studies of children exposed to acute, potentially traumatic events.

CTPT datasets include individual-level data from studies that assess prevention or treatment of the impact of child trauma exposure.

When data are contributed to either Archive, the Child Trauma Data Archives team works with contributing investigators to apply standard variable names and values for many common variables.

CTDA has established standard names / naming conventions for variables including: 

  • Study-level information - e.g., study location, study group
  • Demographic characteristics
  • Trauma & event characteristics
  • Medical care and physiological measures
  • Trauma history variables
  • (for CTPT datasets) Variables related to intervention content and delivery
  • Item-level variables from measures of traumatic stress and other relevant outcomes

Download the Child Trauma Data Archives Project Data Manual for variable names and descriptions

Harmonizing item-level data across measures

Child trauma studies use a variety of different measures to assess traumatic stress and other symptoms, as well as risk and protective factors and other covariates. To facilitate cross-study analyses, the Child Trauma Data Archives team develops harmonization algorithms based on expert consensus. We then apply these algorithms to data from each study, to derive harmonized common variables at the item and scale level.

Examples of harmonized variables available from CTDA:

  • child acute traumatic stress symptom items assessed within the first 24 hours post-event
  • child acute traumatic stress symptom items assessed within the first month post-event
  • child posttraumatic stress outcomes assessed at least 3 months post-event
  • parent/caregiver posttraumatic stress outcomes
  • (coming soon) child and parent/caregiver anxiety and depression symptom items

The table below demonstrates the mapping of item content corresponding to DSM IV and DSM 5 criteria for ASD and PTSD: Each row shows a symptom criterion, with corresponding items from a variety of measures included in CTDA datasets. 

Exemplar syntax / algorithms

Please contact the CTDA team with any questions.

Exemplar syntax (SPSS) used to apply standard names, labels, values, and value labels:

Exemplar syntax (SPSS) used to derive harmonized common variables at the item level

In general, we follow the following three steps to create harmonized variables for the presence / absence of each symptom for each participant, linked to that participant's actual time of assessment (in days post-trauma). 

  • Compute the "true time" of assessment for each participant using that participant's days since an index trauma (days_t[x]) for each study assessment.
  • Dichotomize each individual item for symptom presence / absence
  • Create common-cross measure harmonized variables