NEW By 2030, CMS will tie reimbursement to outcomes. MyTherapyWizard is built for what is coming. Read the VBC brief →
The measurement system

GPI and SPI, how we measure therapy.

Goal Performance Index (GPI) combines how well a client performed against the target with how much support was required to perform at that level. Session Performance Index (SPI) averages GPI across the goals addressed in a session. Together, they turn every session into structured, longitudinal, defensible outcome data, calculated as a byproduct of doing therapy.

Book a demo → Clinical intelligence →
GPI · GOAL LEVEL

Goal Performance Index

GPI = f(performance against target, support level required)

Per-goal, per-session. Calculated at session-save time from the structured capture: correct out of attempted (or frequency, rate, duration, completion, whichever measurement type the goal uses), cues level, level of assistance, and any goal-specific challenges that surfaced.

A client scoring 70% accuracy independently produces a different GPI than a client scoring 70% accuracy with maximum assistance, because those are clinically different events. The math reflects that.

SPI · SESSION LEVEL

Session Performance Index

SPI = mean(GPI) across goals in session

Per-session aggregate of GPI across the goals addressed. Produces a single session-level snapshot, one number that captures how the session performed across the whole goal set you worked on.

Rolls up to longitudinal per-client views. Every session you run adds structured data points to every goal's trajectory, and to the session-level trajectory that rolls them up.

Worked example

Same accuracy. Different GPI. Because the clinical event is different.

Two clients, same goal, "sustain attention on structured tabletop task for 5 minutes, 70% of trials", same accuracy, different support. The math shows what a narrative note would hide.

Client A

70% accuracy, independently performed.

Performance
70% correct
Cues
None
Assistance
Independent
Challenges
None noted
GPI
High
Client B

70% accuracy, with maximum assistance.

Performance
70% correct
Cues
Continuous verbal
Assistance
Max assist
Challenges
Attention, regulation
GPI
Lower

Why this matters. In a narrative note, both sessions read the same, "client achieved 70% accuracy on the attention task." In GPI, Client A's score shows independent performance at target, while Client B's score shows target achieved only with maximum support. Three months later, when progress reports are generated, the trajectory of each client's GPI tells a completely different clinical story. Client A is approaching mastery. Client B is showing target-level accuracy only under intensive support, which is a data point the report should surface, and will, because the structure is there.

What this replaces

Narrative prose that says less than the data underneath.

Most therapy documentation captures accuracy. It doesn't capture support. And when progress reports are written months later, the reports say "improved from X% to Y%", without accounting for whether the support level changed in ways that make the comparison meaningful, or make it misleading.

GPI is designed to capture what clinicians already know intuitively, that 70% with max assist is a different clinical event than 70% independently. By encoding both dimensions into a single structured score, the system produces data that's comparable across sessions in ways narrative prose simply can't be.

Rasch-ready architecture

Interval-scaleable. Calibratable as the dataset grows.

Goals in MyTherapyWizard are stored as structured components, skill, measurement type, target value, support level, mastery rule, baseline. Performance data is captured as interval-scaleable values. That means as the cross-client dataset grows, the scoring framework supports statistical calibration to interval-level precision, the standard used in the most rigorous outcome measurement research in rehabilitation science.

You don't have to care about Rasch measurement theory to use the product. But if you're a practice owner or researcher who pays attention to where therapy measurement is going, it's worth knowing the infrastructure is already in place. The ICF framework alignment, the structured goal decomposition, the measurement-type-aware capture, all of it compounds as the dataset grows.

Value-based care readiness

The outcome data payers increasingly require.

Value-based care is coming to therapy. Payers are already beginning to require structured outcome reporting for prior authorization and continued authorization. MyTherapyWizard is built for that reality, every session produces the structured measurement data that value-based contracts will require, without a separate reporting workflow.

The Denial Analytics and Authorization Analytics views already produce the payer-dimension and therapist-dimension aggregates value-based contracts need: approval rate by payer, denial rate by payer, visit-reduction patterns, outcomes by CPT, outcomes by discipline. The data you need for contract negotiation, and for VBC risk adjustment, is in the system today.

What GPI / SPI feed

One capture. Six downstream outputs.

The structured session capture that calculates GPI and SPI drives every downstream clinical and reporting workflow, without a second pass, without a parallel data entry, without a separate outcomes system.

01

Session notes

Note composes from the structured capture. Goals addressed, performance, support, challenges, GPI per goal, SPI overall.

02

Progress reports

Longitudinal GPI / SPI trajectories become the "improved from X to Y" language in the report, grounded in measured change, not memory.

03

Plans of care

Goal trajectories inform continuation, modification, or discharge. Mastered goals close. Flat goals surface for clinical re-evaluation.

04

RTI summaries

Tiered response queries pull directly from GPI trajectories. Responding / partial / needs escalation, queryable across the caseload.

05

VBC outcome reporting

Per-payer, per-CPT, per-therapist outcome aggregates. The structured data value-based contracts require, already captured.

06

Research & Rasch calibration

As the dataset grows, interval-scaleable data supports statistical calibration to the precision rehabilitation outcome research requires.

See GPI and SPI inside MyTherapyWizard.

Book a 30-minute walkthrough. We'll show the exact session capture workflow, how GPI calculates, and what the trajectories look like across weeks and months.