Skip Navigation

Your Environment. Your Health.

University of Pittsburgh at Pittsburgh

Psychosocial Stress Exposure: Real-Time and Structured Interview Technologies

Thomas Wilson Kamarck
tkam@pitt.edu

 

Project Description

 

Over the past five years, we have developed two prototype technologies that use computer-based self-report methods to capture variations in temporal exposure to psychosocial stress: the Life Events Assessment Profile (LEAP) and the Self-report Mobile Activity Recording Tool (SMART).

 

The LEAP is a computer-assisted interview designed to quantify degree of recent exposure to adverse life circumstances. The LEAP is closely modeled after the Life Events and Difficulties Schedule (LEDs), originally developed by Brown and Harris (1989) and widely acknowledged to be the gold standard assessment tool for the measurement of stressful life circumstances. Among other features, the LEDS uses a life history calendar system to help the respondent identify the timing, onset, and duration of each reported stressor.

 

Although the LEDS has a number of useful features, the time demands involved in training and scoring with this method make it expensive and impractical in large scale epidemiological research. The LEAP was developed to embody the key features of the LEDS, while reducing time requirements associated with training, administering, and scoring. 

 

Our pilot work supports the reliability and validity of the LEAP. In particular, we have shown a high level of correspondence between the LEAP and LEDS (r=0.75-0.83), and excellent test-retest stability in the assessment of 12-month stressor exposure (r=.92). Our current version of the web-based LEAP interview includes:

 

  • Menus and functions guiding the course of the interview,
  • An interactive calendar/dating system, and
  • A real time scoring algorithm which is currently under development.

 

These 3 characteristics should reduce time demands and enhance the standardization and disseminability of life stress interview assessment.

 

The SMART is an Android mobile phone-based assessment technology designed to assess key dimensions of daily psychosocial stress in near real-time and in the natural environment using ecological momentary assessment (EMA). We have previously shown that daily life experiences of psychosocial demands, when assessed in this manner, are associated with important health outcomes. One of the barriers to the use of EMA methods for epidemiological research involves the dearth of standardized assessment tools in this area.

 

We have developed a user-friendly platform for administering real-time interviews by mobile phone, as well as a new item pool for EMA assessment of psychosocial stress for use with this instrument. Important features of the SMART include

 

  • Inclusion of conceptually important dimensions of stress appraisal, and
  • Items relevant to those of diverse educational backgrounds,
  • Application of computer adaptive testing (CAT) methods, which enhance reliability and reduce subject burden. 

 

We know of no previous efforts to make use of CAT in the development of EMA-based measurement tools.  In an item-analysis study, the 13 scales developed for the SMART were reliable (low standard errors of estimate) and valid (associated with divergent psychosocial features during daily life and with concurrent estimates of momentary perceived stress).

 

We are currently moving into a new phase of validation and field testing for each of these instruments and have been funded to administer the LEAP and the SMART offsite as part of an epidemiologic cohort in New York City (Masked Hypertension Study, Dr. Joe Schwartz, PI). This study will help to establish the validity of the SMART and LEAP as measures of psychosocial stress and health risk. It will also provide us with important field testing experience with these devices and ultimately facilitate their use as part of multi-site epidemiological work. We plan to continue to refine these tools, to enhance their portability, and to collect normative and validation data from larger population samples.

 

See this project's publications and patents 

Back to Top