Goal Attainment Scaling

Original Editor - Lauren Heydenrych

Top Contributors - Lauren Heydenrych  

Introduction

The Shirley Ryan Abilitylab defines Goal Attainment Scaling or GAS as "an individualized outcome measure involving goal selection and goal scaling that is standardized in order to calculate the extent to which a patient's goals are met."

GAS is a goal-setting technique which has a mathematical basis. GAS was first introduced in the 1960s by Kirusek and Sherman. This was originally for the mental health setting. Since then it has been adapted and modified to include a variety of areas and settings.

GAS is important, as it allows the various stakeholders in the rehabilitation process - and primarily the patient - involvement in the measurement and scoring of goals. GAS gives individuals the ability to construct their own outcome measures; this is in contrast to measures that are based on a standard set of tasks. Scoring, however, is performed in a standardised way which in turn allows for statistical analysis.

Advantages in using GAS

1. Encourages communication and collaboration between multidisciplinary team members.
2. Promotes patient involvement. In research cited by Turner-Stokes (2009) goals for rehabilitation are more likely achieved with the involvement of the patient and their family. In addition, more information sharing is noted at the beginning of a rehabilitation process when formalised goal setting is introduced. This in turn allows for negotiation of realistic goals.

Criticisms of GAS

1. Potential for bias: For example, goals being too easy to attain or measuring change which does not actually occur.
2. Lack of interrater reliability. This however improved with training into goal setting and rating. As well as the use of multiple raters.
3. A time consuming process. Cited as taking as long as 45 minutes per child in the paediatric setting.

Diagnosis/Conditions which can use GAS

• Brain injury
• Cerebral Palsy
• Pain management
• Parkinson's' Disease
• Neurologic Rehabilitation
• Sports and Musculoskeletal

Assessment type

Patient-reported outcomes

Description

GAS comprises of goals divided into a 5-point scale from -2 to +2.

• Achievement of goal = 0
• Achieved somewhat more than expected outcome = +1
• Achievement of much more than expected goal = +2
• Achievement of somewhat less than expected goal = -1
• Achievement of much less than expected goal = -2

As some goals may be more difficult to achieve than others and some may be more important than others, the option of weighing different goals is a possibility. A spreadsheet to help calculate goals scores, including their weighting can be found at various sites, including Assesschild.com (found in the resources list at the bottom of the page).

There is no set limit to how many or how few goals can be set at any one time. However, it is recommended that 3 is a good number and a maximum of 6 is set.

Procedure

1. Identify goals

• Interview patient and/or caregivers to identify a) problem areas and b) priority areas
• Decide on a date of achievement
• Note: Goals should follow the SMART principle.
  • Specific, Measurable, Attainable, Realistic and Time-related.

2.Weight goals

• Done by patient ranking goals by importance or using a weighting scale. Here the team may rank or weigh goals by the expected difficulty in achieving them.

Weight = importance × difficulty

3. Define the expected outcome

• Assign descriptors of results in each case of -2 up to +2. Each level should be as objective and observable as possible.

• This is an opportunity to address unrealistic expectations which may be held by the patient and/or caregivers.

4. Scoring baseline and interval scoring

• Usually baseline scores are rated at -1, unless there is no clinically plausible worse condition. In this case -2 is rated. Generally, and in the paediatric field -2 is the patient's baseline level. With intervals, it is important that the scaling between each level is as equal as possible. Ideally when change is measured, it is also important that there is only 1 variable of change per scale level. Only when the goal does not remain clinically meaningful, can 2 or more variables be changed within a scale. However, improvement should not be expected to occur simultaneously on these scales. For example:

Goal - Writing letters with assistance of caregiver.

• -2 = Unable to write name, even with assistance.
• -1 = Writes 5 letter with hands-on assistance and verbal cuing given by caregiver.
• 0 = Writes 10 letters with hands-on assistance and verbal cuing by caregiver.
• 1 = Writes 10 letters with verbal cuing
• 2 = Writes 10 letters independently.

In this example, the variables that changed were a. the number of letters written (5 to 10) and b. assistance given (Namely, hands-on, then verbal cueing, then none). While both variables changed in the scale, only 1 variable changed at a time between scale levels.

5. Goal attainment scaling

Goals are reviewed at appointed dates. Here, actual performance is judged at defined levels. The ideal is to have this is performed with the team - including the family.

Baseline scores are compared with currently performed tasks.

The use of weighted values are used within a formula in order to get a standardised score. An excel spreadsheet with the formula can be found on the Assesschild website. A link is found in the resource section.

Otherwise, baseline scores can simply be compared to scores achieved on follow-up, to monitor development or deterioration.

Reliability

In the analysis of systematic reviews within the last decade, it was found that inter-rater reliability (IRR) was good, but varied according to a number of factors including:

• The person scoring - IRR improved when the physician or therapist treating the patient as well as the raters observed the patient directly, instead of a recording being watched.
• Field in question
• Precision of levels described

One study performed by Mailloux et al. (2007) recommended training programs and testing of inter-rater reliability to improve this aspect.

A study performed by King et al (2000) found that IRR could be high (ICC=0.98) when performed at the same occasion

Intra-rater reliability has not been commented on.

Validity

One of the criticisms of GAS is that it doesn't measure one clear construct. This is because of the individualistic nature of the goal setting. This makes validity difficult to assess. Despite this, research done in 2016 found that content validity produced results that were "good" or "intermediate".

Resources

• Goal Attainment Scaling (GAS) in Rehabilitation. A practical guide.
• GAS manual (paediatric orientated)
• GAS at Assesschild.com
• Page at Shirly Ryan Ability Lab website: https://www.sralab.org/rehabilitation-measures/goal-attainment-scale