Outcomes Framework - Frequently Asked Questions (FAQ)

1. How is the Treatment Episode start date defined?

For customers with an appointment integration with the EMR:

If we receive Treatment Episode information from you, we define the Treatment Episode start date as the enrollment date.
If we do not receive Treatment Episode information but do receive appointment information, we define Treatment Episode start date as the date of the first appointment for that organization/service. The date can be viewed on the Treatment Episode details page for a patient.

For customers without an appointment integration with the EMR:

We define Treatment Episode start date as the date of the first appointment for that organization/service. The date can be viewed on the treatment episode details page for a patient.

2. Why can’t I set my Success Criteria to <X> ?

Firstly - please feel comfortable reaching out! We may be able to support your criteria. In general, we have simplified the criteria that we support to make it easy for organizations to match the data analysis we provide, particularly in the excel download.

Metrics that involve complicated triggering criteria require the ability to only select one measure per Treatment Episode, so as not to double count a particular patient. This is challenging in Excel.
We therefore provide additional columns in our data export to help choose only one measure per patient.

3. What happens if a patient does not complete a full measure?

Each scale has a defined way of determining whether missed questions will render the result invalid. Examples include:

The PHQ-9 is only valid if all the questions are filled out.
Measures such as the SFSS allow you to miss up to 15% of the items, and the score is prorated to adjust for missed items.
Other schemes also exist.

The Outcomes Framework only deals in valid responses. If the patient has missed enough items that the measure is considered invalid, it will not be considered in the analysis.

4. How does validity of response affect outcomes?

Only valid responses are considered in the Outcomes Framework. If a patient has not filled out a completed measure and the result is considered invalid, it will not be present in any of the Outcomes analyses.

5. Why is the number of Included patients so low?

There are multiple reasons why the number of patients appearing the analysis could be lower than you expect:

Only patients who have been measured at least twice on the scale in question appear.

Things you can do to get more patients in the analysis:

Extend the date range from treatment episode start to a wider period.
Relax the Inclusion Criteria to allow more results to be counted.
For Efficiency Metrics, change the time period to include more results. If you are using the Last measurement of week mode, consider switching to the Last measurement of month mode instead.
Try to use a different scale for the analysis.
Consider as a last resort changing your measurement protocols to measure more often. You will not see the impact of this for several months as you must wait for the new data collection to take effect

6. The clinical severities do not match the scale’s documentation. What is happening?

In an attempt to make cross-scale comparison easy, the Mirah system has abstracted the unique per-scale diagnoses such as ‘Moderate Depression’ or ‘Severe Anxiety’ into four simple severity categories: ‘None’, ‘Mild’, ‘Moderate’ and ‘Severe’. If you wish to target a precise category you may still use criteria based on the raw score.

More on Clinical Severity Categories here.

7. My patient got better the day after the Time Period ended, can it be included as a success?

The Mirah Outcomes Framework is designed as a population-level analysis. Inherent in this is that the number of patients where edge cases like this occur will be fairly small. The intent is to make judgments holistically rather than on the basis of individual cases.

If you find that a large percentage of your patients are edge cases, consider changing the metric criteria to match your clinical intuition.

8. How many metrics should I track?

There is no simple answer to how many metrics you should track. We at minimum would recommend two:

An Effectiveness Metric measuring Treatment Episode Response and Remission for your most important scale.
An Efficiency Metric measuring the average time to achieve this Treatment Episode Response.

Beyond that, there are several circumstances where more metrics may be warranted:

If you are treating different age groups, each with a different scale to monitor the Treatment Episode, you will need a different set of metrics for each.
If you provide services for a variety of levels of care, for example both outpatient and inpatient, it is usually good practice to have a separate set of metrics for each service line, as the objectives of each tend to be very different.
If your programs are specialized towards a set of particular diagnoses where Treatment Episode varies for each, consider having a separate metric for each.

More on Effectiveness and Efficiency Metrics here.