Initial Development of a Multidimensional Computerized Adaptive Test for Intensive Longitudinal Assessment of Suicide...

Area of Science:

• Psychological Science
• Medical Research
• Behavioral Science

Background:

• Intensive longitudinal designs, like ecological momentary assessment (EMA), are crucial for studying dynamic processes in medical and behavioral science.
• However, participant burden from repetitive surveys limits EMA's scope, especially for complex phenomena like fluctuating suicide risk.
• The Computerized Adaptive Test for Suicide Risk Pathways (CAT-SRP) was developed to address these limitations by personalizing survey content.

Purpose of the Study:

• To develop, calibrate, and pilot the first multidimensional computerized adaptive test (CAT) for suicidal thoughts and related psychosocial risk factors.
• To adapt existing instruments for use in intensive longitudinal designs and create a comprehensive item bank.
• To evaluate the CAT-SRP's performance in reducing participant burden and improving the precision of suicide risk assessment.

Main Methods:

• A web-based platform administered the CAT-SRP, adapting items from validated instruments for intensive longitudinal designs.
• Item parameters were calibrated using a multidimensional graded response model in a large community sample (n=1759).
• The CAT-SRP was evaluated in an EMA study (n=29) using D-optimal item selection, a dual variable-length stopping criterion, and MAP scoring.

Main Results:

• The calibration identified 2 suicidal thought domains and 12 risk factor domains.
• The CAT-SRP demonstrated low to moderate item overlap and incurred low participant burden.
• The CAT-SRP showed potential for improved near-term prediction of suicidal thoughts compared to traditional EMA, though refinement of selection and stopping rules is needed.

Conclusions:

• The CAT-SRP effectively personalizes EMA content, reducing repetitiveness and burden in intensive longitudinal research.
• Multidimensional computerized adaptive testing (MCAT) provides nuanced measurement, addressing zero-inflation issues in EMA and strengthening predictive relationships.
• The CAT-SRP offers significant methodological advantages for intensive longitudinal data collection in suicide risk research.