Computer-Adaptive Functioning Testing
Also known as: Functional CAT, Adaptive Functioning Assessment, CAT for Disability Outcomes, Adaptive Function Measurement
Computer-adaptive functioning testing applies the logic of computerized adaptive testing to the measurement of functioning and disability outcomes, such as physical function, mobility, and daily activity. Instead of administering a fixed list of items to everyone, it draws on a precalibrated item response theory item bank and an algorithm that tailors the test to each respondent in real time. After each answer, the algorithm updates its estimate of the person's underlying function level and selects the next item that will be most informative at that estimate, continuing until a target precision is reached or a maximum number of items is administered. The approach was central to the United States PROMIS initiative, whose physical-function item bank Rose and colleagues evaluated in 2008 and whose first wave of adaptive instruments Cella and colleagues described in 2010. The result, framed here specifically for disability and rehabilitation outcomes, is measurement that achieves high precision with far fewer items than a fixed questionnaire, reducing respondent burden for people who may already face fatigue or communication barriers.
Key highlights
- Achieves high measurement precision with far fewer items than fixed questionnaires by spending each item where it is most informative.
- Reduces respondent burden, which is particularly important for people with disabilities who may face fatigue or communication barriers.
- Produces scores on the item bank's common metric, so adaptive results are comparable across respondents and with short forms.
- Tailors difficulty to each person, avoiding floor and ceiling effects that fixed instruments suffer at the extremes of function.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use computer-adaptive functioning testing when you need precise, comparable measurement of functioning or disability outcomes while keeping the number of administered items as low as possible. It is especially valuable in rehabilitation, chronic-illness, and disability research and care, where respondents may tire easily or face communication barriers, and where repeated measurement over time benefits from low burden. The approach requires a precalibrated, well-targeted item bank and a computerized platform to run the selection and scoring in real time, so it is not appropriate when no calibrated bank exists, when administration must be on paper, or when every respondent must answer identical items for procedural reasons. It is also less suitable for very small banks or constructs that are not well represented across the full range of function, where a fixed short form may perform just as well.
Strengths & limitations
- Achieves high measurement precision with far fewer items than fixed questionnaires by spending each item where it is most informative.
- Reduces respondent burden, which is particularly important for people with disabilities who may face fatigue or communication barriers.
- Produces scores on the item bank's common metric, so adaptive results are comparable across respondents and with short forms.
- Tailors difficulty to each person, avoiding floor and ceiling effects that fixed instruments suffer at the extremes of function.
- Requires a precalibrated item bank and a computerized administration platform, raising the setup cost relative to a paper short form.
- Item selection can over-expose the most informative items and under-use others unless exposure-control constraints are added.
- Precision near the extremes of function depends on having enough informative items there, which sparse banks may lack.
- The validity of the tailored score rests on the calibration holding in the target population, so differential item functioning must be monitored.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How does CAT measure function with fewer items than a fixed questionnaire?
By choosing each item to be maximally informative for the specific respondent. After every answer, the algorithm updates its estimate of the person's function level and then selects the next item whose information is greatest at that estimate. Items that are far too easy or too hard for a given person carry almost no information, so skipping them loses nothing while saving questions. The result is that a confident estimate is reached in a handful of well-targeted items rather than dozens of one-size-fits-all ones, which is why Rose and colleagues found the PROMIS bank precise with few items.
What stops the adaptive test from running indefinitely?
A stopping rule, usually based on precision. The most common rule ends the test once the standard error of the function estimate drops below a target threshold, meaning the estimate is precise enough. A maximum-items cap is also set so the test terminates even when a respondent's function is hard to localize, for example near the extremes of the bank. Together these rules ensure each person answers only as many items as needed to reach the desired precision, which is the central efficiency promise of adaptive functioning testing.
Are adaptive functioning scores comparable across people who answered different items?
Yes, provided the items come from the same calibrated bank. Because every item is calibrated to a common metric through item response theory, scores estimated from different item sets land on the same scale, just as they do for fixed short forms drawn from the bank. A frail patient answering easy items and an active person answering hard items receive directly comparable scores. The key caveat is that the calibration must hold in the population being tested, so differential item functioning should be checked before comparing groups.
Sources
- 1.Rose, M., Bjorner, J. B., Becker, J., Fries, J. F., & Ware, J. E. (2008). Evaluation of a preliminary physical function item bank supported the expected advantages of the Patient-Reported Outcomes Measurement Information System (PROMIS). Journal of Clinical Epidemiology, 61(1), 17-33.
- 2.Cella, D., Riley, W., Stone, A., et al. (2010). The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005-2008. Journal of Clinical Epidemiology, 63(11), 1179-1194.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 23). Computer-Adaptive Functioning Testing. ScholarGate. https://scholargate.app/disability-studies/computer-adaptive-function-testing