Function propensity_score_matching

Description

---

The propensity_score_matching function matches patients in two cohorts based on their propensity scores to create two balanced cohorts.

Inputs

---

• scores - user defined table with 3 mandatory columns:

  • patient_id

  • cohort

  • propensity_score

• optional arguments:

  • caliper - a user-specified caliper for matching; default is 0.10 SD

Returns

---

• Upon running the propensity_score function, patients are matched based on propensity scores to create a 1:1 ratio between cohorts using:

  • Nearest neighbor (greedy) without replacement

  • Default caliper of 0.10 SD or user-defined caliper if entered as an input

  • Cohorts are randomly shuffled prior to matching

• Returns a table of matched patients with the following columns

  • patient_id

  • cohort (returns whatever value was input)

  • propensity_score

Example

---

scores

| patient_id  | cohort   | propensity_score  |
|-------------|----------|-------------------|
| 1           | 1        | 0.30              |
| 2           | 1        | 0.55              |
| 3           | 1        | 0.92              |
| 1           | 2        | 0.89              |
| 2           | 2        | 0.74              |
| 3           | 2        | 0.21              |

matched = propensity_score_matching(scores)

| patient_id  | cohort   | propensity_score  |
|-------------|----------|-------------------|
| 1           | 1        | 0.30              |
| 7           | 2        | 0.30              |
| 2           | 1        | 0.55              |
| 4           | 2        | 0.56              |
| 3           | 1        | 0.92              |
| 1           | 2        | 0.89              |