FOTO Measure Calculation Guide

FOTO provides step-by-step instructions for manually calculating their risk-adjusted functional outcome measures. The six required steps are reproduced here using Measure 217, the Functional Status Survey for the Knee, for a patient with the following details:

  • Male
  • Age 68
  • Osteoarthritis of the Knee
  • Pain onset: 10 days ago

Step 1: The patient completes FOTO’s Functional Status Survey for the Knee at Admission.  Use the patient's score to determine the patient’s Functional Status Score at Admission.

Example: The patient scores 26 points on the survey. This translates into the Functional Status Score of 42 on the Scoring Algorithm sheet (scores range 0-100 with higher scores indicating better physical function). 


Step 2: The patient completes FOTO’s Functional Status Survey at or near Discharge. Use the patient's score to determine the patient’s Functional Status Score at Discharge.

Example: The patient scores 48 points on the survey at discharge. This translates to the Functional Status Score of 78 on the Scoring Algorithm sheet.


Step 3: Calculate the patient’s Functional Status Change Score (raw, non-risk-adjusted).

This is the patient's score at discharge (Step 2) minus the patient's score at admission (Step 1).

Example: 78 – 42 = 36 points


Step 4: Calculate the patient's Risk-adjusted Predicted Functional Status Change Score using a regression equation.

Risk adjustment is a mathematical process for addressing factors that might influence a patient’s performance on an outcome measure. FOTO provides a spreadsheet that lists all of the factors they have determined influence outcomes on their surveys. Examples include characteristics such as patient gender, payer type, and surgical history.

All of the factors for each survey are listed on Coefficient Spreadsheets that are linked here:

217: Knee | 218: Hip | 219: Ankle/Foot | 220: Low Back | 221: Shoulder | 222: Elbow/Wrist/Hand | 478: Neck (Login to access survey)

What you're calculating is how your patient’s actual change compares with a change predicted by a statistical formula called a regression equation. Regression equations are handy because they help us make apples-to-apples comparisons among patients who may differ according to those influential factors. For example, patients who are male may have predictably different outcomes than patients who are female; or patients who have had multiple surgeries may have predictably different outcomes than patients who have had no surgeries.

In this scenario, the predicted outcome is the risk-adjusted change in function score for the patient who completed the Functional Status of the Knee survey.

The regression formula used to predict a Risk-adjusted Functional Status Change Score includes the following information:

  • Constant: This is a statistically determined value. The constant is listed at the top of each coefficient spreadsheet. In all cases, it is a positive number. Example: The constant for the Functional Status of the Knee is 50.49. As a result, the constant would be entered into the regression formula as +50.49.

You'll also need to include all patient relevant characteristics. A characteristic that positively influences the outcome score is entered in the formula with a + sign. A characteristic that negatively influences the outcome score is entered in the formula with a – sign. Characteristics in all regression equations are adjusted by a statistically determined coefficient. FOTO has provided the coefficients for all of their surveys. You'll determine which coefficient to use for each characteristic based on the specific details of the patient who completed the survey. [Note: Coefficients are specified to at least 2 decimal places. DO NOT round up or down or your results will be inaccurate.] 

  • Intake Score Regression. A critical characteristic to add in the FOTO risk adjustment calculation is the intake (i.e., first) survey score. Example: This patient’s initial score was 42 points. As a result, the intake score would be entered into the regression formula as (-0.53 x 42) or -22.26.
                                       
  • Age. Some characteristics, such as patient age, have a single coefficient to choose from. For the knee survey, the coefficient for age is -0.14. Example: The patient who completed the Functional Status of the Knee survey is 68. As a result, the characteristic for age would be entered in the regression formula as (-0.14 x 68) or -9.52.   
                         
                
  • Acuity. Some characteristics, such as acuity, have multiple coefficient options from which only one should be selected. Example: This patient reports the onset of knee pain about 10 days ago after going on a strenuous hike during vacation. As a result, the characteristic for acuity would be entered in the regression formula as (3.62 x 1) or +3.62. 
  • Arthritis/COPD. Some characteristics, such as comorbidities, have multiple coefficient options, all of which should be entered into regression formula. [NOTE: In situations like this, the value assigned to factor = 1 if it is present and = 0 if it is absent.]Example: This patient has arthritis but does not have COPD. As a result, the characteristic for arthritis would be entered into the regression formula as (-1.45 x 1) or -1.45 and the characteristic for COPD would be (-1.05 x 0) or +0.     
                       

Fianlly, a regression formula takes this form:

y = constant + [coefficient x (characteristic A)] – [coefficient x (characteristic B)] – [coefficient x (characteristic C)] + [coefficient x (characteristic D)] ...etc.

Don’t panic – this is math you can do!!!

  • Example: Using the prior examples, the Functional Status of the Knee regression formula would start to look like this [NOTE: To complete this formula, you must enter all of the relevant characteristics on the spreadsheet.]
  • Predicted risk-adjusted outcome score = 50.49 (Constant) – 22.26 (Intake Score Regression) – 9.52 (Age) + 3.62 (Acuity) – 1.45 (arthritis) + 0 (COPD)…. Note: this formula is not complete and you would need to add in every relevant patient characteristic. But for the sake of this example, we'll use this information for the calculation.
  • For this scenario, the hypothetical predicted Risk-adjusted Functional Change Score with all of the influential factors included = 20.88 points.

Step 5: Calculate the Risk-adjusted Functional Status Change Residual Score for the patient.

A “residual score” is a statistical term for the difference between the change score your patient achieved (step 3) and the predicted change score you calculated from your regression formula (step 4).

Example: 36 – 20.88 = 15.12 points

NOTE: Residual scores may be positive (+) or negative (-). A negative score does not necessarily mean your patient failed to achieve meaningful improvement. Individuals may respond to your interventions sufficiently to meet their goals without always hitting the predicted change score.


Step 6: Submit the appropriate quality data code based on the residual score.

MIPS outcome measure 217 specifies the quality data codes for the risk-adjusted FOTO functional status of the knee change score.

Example: In this scenario, the residual change for this patient is > 0. As a result, you would submit G8647: Risk-Adjusted Functional Status Change Residual Score for the knee impairment successfully calculated and the score was equal to zero (0) or greater than zero (> 0).

Remember, if you are including this score at discharge, use an FLR discharge code to signal the end of treatment for this patient.

Click here to learn more about reporting FOTO measures in the WebPT EMR.