## Scoring **collabo**RATE

This page describes two methods of scoring

While preliminary data suggests acceptability of both scoring methods (Barr et al., 2013), we strongly recommend use of the 'Top Score' approach for its interpretability and ability to detect variation in scores across clinicians and clinics.

A

We have noted in preliminary testing that ensuring respondent confidentiality is critical to observing variation in scores.

**collabo**RATE.While preliminary data suggests acceptability of both scoring methods (Barr et al., 2013), we strongly recommend use of the 'Top Score' approach for its interpretability and ability to detect variation in scores across clinicians and clinics.

A

**collabo**RATE Score should only be calculated when all three**collabo**RATE items have been completed for at least 25 clinical encounters for the particular clinician, clinic, or other group of interest.We have noted in preliminary testing that ensuring respondent confidentiality is critical to observing variation in scores.

**collabo**RATE Top Score

**Categorical or continuous?**Continuous

**Possible range?**0-100

**How to calculate?**Exclude cases where a response to one or more of the

**collabo**RATE

__questions is missing. Code each encounter as either '1', if the response to all three__

**collabo**RATE

__items was 9, or '0' if the response to any of the three__

**collabo**RATE

__items was less than 9. Then, calculate the percentage of all encounters that were coded as '1'. This number is the__

**collabo**RATE

__Score.__

**How to interpret?**Higher scores represent more shared decision making. This number also corresponds to the proportion of patients for whom there was 'gold standard' shared decision making.

**collabo**RATE Mean Score

**Categorical or continuous?**Continuous

**Possible range?**0-9

**How to calculate?**Exclude cases where a response to one or more of the

**collabo**RATE questions is missing. Calculate the mean of the three

**collabo**RATE responses for each encounter. Then, calculate the mean of all encounters of interest. This number is the

**collabo**RATE Score.

**How to interpret?**Higher scores represent more shared decision making.