The Brain Injury Functional Outcome Measure (BI-FOM) is a measure of global functioning after moderate or severe brain injury, developed through a collaborative project of the Traumatic Brain Injury Model Systems.

The BI-FOM is intended to address an important measurement gap in both acute neuroprotection trials and post-acute rehabilitation outcome and intervention studies: existing outcome measures either apply only to a subset of the recovery spectrum (i.e., have serious floor or ceiling effects) or span the full range with poor precision.

Thus, researchers and clinicians have struggled to find a single measure that demonstrates graded functional impairment at enrollment and graded functional outcome at follow up for use in studies that enroll patients acutely (when the range of possible outcomes is great) or for longitudinal studies of the post-acute period.

The BI-FOM is intended to allow measurement of global function throughout the range of functioning from unconsciousness to resumption of normal activities, with interval measurement properties and precision similar to FIM. 

The BI-FOM is a composite of 31 items drawn from other measures of functioning. It was developed in two phases using Rasch Analysis.

In the first retrospective phase, datasets from other studies, containing multiple functional outcome measures, were analyzed to identify items that fit the measurement construct assessed by FIM but extended the range either upward and downward.

In the second prospective phase, the 49 items so identified were administered to participants or their surrogates around the time of acute inpatient rehabilitation and again about one year post-injury, in approximate order of difficulty, allowing termination of the interview if participants failed a series of items.

All items except a small set assessing fully unconscious patients (seen almost exclusively during the inpatient phase) could be administered by telephone interview.

A number of items were dropped for misfitting the dimension or for redundancy with other items, and other items were re-coded. The final 31-item set has very high reliability, a considerably larger measurement range than FIM, no notable floor and a considerably lower ceiling than FIM.

However, the BI-FOM is not yet ready for general use as an outcome tool. The measurement properties need to be replicated in an independent data set. Moreover, a few of the 31 items were derived from FIM which is no longer routinely collected in rehabilitation institutions.

As a result, the ability to abstract numerical functional data from routine clinical sources for these items no longer exists. Thus, interview-based alternatives to these items will need to be assessed for their measurement properties. In addition, before final dissemination of the BI-FOM, discussions are needed with the original developers of some of the items.

Current Status: An article detailing the development of the BI-FOM has been peer-reviewed and published in the Archives of Physical Medicine and Rehabilitation.