skip to Main Content
Doctor explaining diagnosis to her female patient

A Quick AM-PAC Overview

The AM-PAC™ instrument was developed at Boston University under a grant from NIH/NICHD/NCMRR and NIDRR. Mediware has licensed this instrument and incorporated it as a key feature of the PAC-Metrix Outcomes platform. Today, PAC-Metrix is the only commercially available source for rehabilitation therapy providers to license and use the AM-PAC™ instrument.

What is the AM-PAC?

The AM-PAC™ is a functional outcomes measurement system that can be used across care settings. At its core, the AM-PAC™ consists of 269 functional activities that have been calibrated across the range of human function using the contemporary measurement technique, Item Response Theory. This means that each of the 269 functional activities are calibrated in a hierarchy of function such that “climbing a flight of stairs using a handrail” is more functional than “climbing 3-5 steps using a handrail.”

These functional activities, called the Item Bank, can be used in two formats:

  1. Computer-based assessments using an application such as PAC-Metrix
  2. Paper-based assessments using pen/paper and a scoring conversion table

The PAC-Metrix web application supports both forms of assessment. It can be used to access the preformatted Short Form questionnaires or can administer the electronic computer-based assessment.


Much research has validated the approach to outcomes measurement used by PAC-Metrix. Here is a sample…


What does the AM-PAC Measure?

The AM-PAC™ assesses activity limitations. According to the World Health Organization, activity limitations are a subcomponent of disabilities and are defined as: “…difficulty encountered by an individual in executing a task or action.” Therefore, the AM-PAC™ is used to measure the amount of functional limitation a patient would experience while performing  everyday tasks.  It assesses activity limitations by asking two types of questions:

  • How much DIFFICULTY does the patient experience while performing a task?   
  • How much ASSISTANCE does the patient require to perform a task?

What sets the AM-PAC™ apart from other measures is that it measures across nearly the entire range of human function in one integrated scale.  Other measurement systems establish floor and ceiling boundaries and cannot measure outside of those limits. The AM-PAC™ involves questions with as much impairment as “How much DIFFICULTY do you currently have turning over in bed,” and as little impairment as “How much DIFFICULTY do you currently have running 5 or more miles?”

No other patient measurement instrument can address such a wide range of functional impairment levels. This is critical if you treat patients across a wide range of functional impairment or if you want to track patients across a series of care settings where functional impairment can vary widely.

Focus on Patients

As an activity limitations-centric instrument, the AM-PAC™ measures outcomes from the perspective of the patient. The patient is asked to consider his or her own functional performance and answer questions in the assessment. When a patient cannot effectively self-assess due to cognitive impairment, or other limitations, the instrument has proven reliable when taken by a proxy. Both family members and therapists have proven to be acceptable proxies, providing relevant assessments.

The use of functional activities as the subject of the assessment moves the focus away from specific body part of the medical or rehab diagnosis. Instead the measurement assesses the impact of the functional limitation within the life of the patient.  In this way, the single AM-PAC™ instrument is useful across a very broad range of impairments, limitations, and diagnoses:

physical therapy documentation software
  • Orthopedic (either post surgery or apart from any surgery)
  • Neurological (including but not limited to stroke, TBI, and SCI)
  • Major Medical Conditions (including cardiovascular)

The AM-PAC™ can be assessed pre-surgery to obtain a baseline or post-onset to track the rehabilitation process.  It can also be used to assess the functional limitations of individuals within a population (such as the aging in a community home setting).

The AM-PAC™ is not, however, validated for use in pediatrics (patients younger than 17 years old), swallowing, or pelvic floor impairments.



AM-PAC Functional Domains

The AM-PAC™ Item Bank contains 269 questions in total. These questions, or items, are divided across three functional and cognitive domains.

  • Basic Mobility. 131 items address basic mobility and activities such as bending, walking, carrying and climbing stairs
  • Daily Activity. 88 items cover self care and instrumental activities of daily living
  • Applied Cognitive. 50 items assess higher level cognitive functions which are needed to live independently

Assessing Multiple Domains

Depending on each patient’s functional or cognitive impairment, different domains may be relevant to be assessed. The AM-PAC™ is designed to automatically adapt to present the appropriate questions for the necessary domains. During the PAC-Metrix patient registration process, each patient must be identified with one of the following conditions:

  • Orthopedic
  • Neurologic
  • Cardiopulmonary
  • Major Medical
  • Other

For each of the selections other than orthopedic, the AM-PAC™ is designed to assess all three domains automatically. In the case of an orthopedic selection, the patient must be further classified by the body part being treated. A list of 16 body part areas are presented, and the user can select as many as are relevant to the patient’s treatment.

The body parts selected will determine which domains are assessed:

  • Any Lower Extremity (Hip-Foot) will cause the Basic Mobility Domain to be assessed
  • Any Upper Extremity (Shoulder-Hand) will cause the Daily Activity Domain to be assessed
  • Any Spinal selection or the Other selection will cause both the Basic Mobility and Daily Activity Domains to be assessed

According to this method, selecting Shoulder (UE) & Hip  (LE) will cause both  Basic Mobility and Daily Activity Domains to be assessed.  By contrast, selecting only Shoulder would result in only Daily Activity and conversely, selecting only Hip would result in Basic Mobility.

Domains and Functional Stages

In order to define a context for understanding the scores generated by the  AM-PAC™ Assessments, the  AM-PAC™ scale scores have been divided into ranges called functional stages.  A functional stage represents a profile of the types of activities a subject might be able to accomplish at different scale score levels. These functional stages are hierarchical and represent functional activities that are increasingly more difficult as one proceeds from a lower to a higher functional stage. The level of difficulty exhibited by most patients with scores in the specified ranges is described for each functional stage.

The following documents explain the different functional stages for each domain as well as provide a visual guide to a series of representative activities and the difficulty of performing them at different AM-PAC™ scale scores and functional stages.

AM-PAC Assessment Report – Basic Mobility

AM-PAC Assessment Report – Daily Activity

AM-PAC Assessment Report – Applied Cognitive

Definition: The AM-PAC scores are based on a scale where the score of 50 represents the mean score, which is based on a 1000 patient calibration study. The scores are distributed around the mean where plus or minus 10 scale points represents 1 standard deviation.

If desired, these reports can be included in the post-assessment output of the PAC-Metrix AM-PAC™ instrument.  They illustrate the relative level of function of a given score by showing how commonly understood tasks map against this score and against the different functional stages.  These can be used to explain the assessment scores to patients, or even to facilitate a discussion of goal setting with the patient.

Even without the full report, the output of the AM-PAC™ instrument summarizes both the functional stage and CBOR G-code modifier value as illustrated in the figure below.


Basic Mobility
Range: -11.95 – 104.9
5 – 84-105: Strenuous Sports
4 – 66-83: Moving Around Outdoors
3 – 52-65: Moving Around Indoors
2 – 34-51: Limited Moving Indoors
1 – -12-33: Limited Movement

Daily Activity
Range: -2.73 – 115.4
5 – 84-110: On Your Own
4 – 62-83: Getting Things Done
3 – 53-61: Difficult Activities
2 – 34-51: Daily Tasks are a Struggle
1 – -3-33: No Independent Task

Applied Cognitive
Range: -6.84 – 68.2
5 – 56-68: On Your Own
4 – 42-55: On the Move
3 – 34-41: Minor Difficulties
2 – 29-33: Communication Limitations
1 – -7-28: Limited Life Skills

Computer Adaptive Testing

The AM-PAC™ was developed to be delivered via a computer and Computer Adaptive Testing (CAT). This means that the instrument does not follow a set course in terms of selecting questions from the Item Bank to present to the patient. The selection of questions is determined instead by the answers that are provided to the questions presented. In general terms, answering a question with “No Difficulty” will prompt the instrument to select a subsequent question that represents a lower level of functional impairment. Conversely, answering a question with “Unable” will prompt the instrument to select a subsequent question that represents a higher level of functional impairment.

As a Computer Adaptive Testing instrument, the AM-PAC™ has certain advantages over traditional survey instruments.

Fewer Questions

Because the AM-PAC™ doesn’t use a fixed schedule of questions, the instrument will only present the number of questions needed to meet the score precision target. In the case of the AM-PAC™, the following rules govern the number of questions:

  • A minimum of 5 questions per domain
  • A maximum of 10 questions per domain
  • Within this range, stop asking questions when the Standard Error (precision) is less than 2.0

Greater Functional Range

While maintaining a reasonable number of questions and a high level of precision, the AM-PAC™ can assess a much broader range of functional impairments than a traditional fixed survey instrument. Since the question selection is responsive to the patients answers, patients will quickly find questions that accurately assess their levels of function or impairment simply by answering the questions presented. This means that the AM-PAC™ is able to span the range of the full spectrum of post-acute care and provide meaningful outcomes data for patients, whether they are severely limited or only suffering minor impairment.

Greater Precision

Since the AM-PAC™ selects questions based on the patient’s prior responses, questions are selected that will improve the accuracy of the score (likelihood that the calculated score represents the patient’s actual level of functional disability). In the case of the AM-PAC™, the majority of the domains assessed have a standard error less than 2.0. This means that the calculated score is within 2.0 of the patient’s actual score. This has several benefits:

  • Change in functional impairment is easier to detect
  • Users can have greater confidence in the representative power of the instrument

Haley SM, Ni P, Hambleton RK, Slavin MD, Jette AM.  Computer adaptive testing improves accuracy and precision of scores over random item selection in a physical functioning item bank.  J Clin Epidemiol. 2006; 59: 1174-1182.

Back to top