# Responsiveness and minimum important change of the Pharmacotherapeutic Symptom Evaluation-20–Australian version: a tool for measuring changes in medicine-related symptoms over time

## TL;DR

This study evaluates a tool for tracking medicine-related symptoms over time and determines how well it detects meaningful changes.

## Contribution

The study establishes the responsiveness and minimum important change of the PHASE-20–Australian version for monitoring medicine-related symptoms.

## Key findings

- Strong correlations (rho = 0.815) were found between score changes and perceived symptom changes.
- The tool showed good discriminative ability (AUC = 0.739–0.975) for most symptoms.
- MIC values exceeded the Smallest Detectable Change for 84.2% of symptoms.

## Abstract

Monitoring patient-reported symptoms over time may support early detection of medicine-related harms. The Pharmacotherapeutic Symptom Evaluation-20–Australian version (PHASE-20–Australian version) is an 11-point rating scale designed for longitudinal monitoring of medicine-related symptoms; however, its responsiveness and minimum important change (MIC) have not been established.

To evaluate the responsiveness and MIC of the PHASE-20–Australian version for monitoring changes in medicine-related symptoms over time.

A prospective cohort study was conducted among Australian adults (≥ 18 years) taking medications. Participants completed the PHASE-20–Australian version at baseline and follow-up and reported perceived symptom changes using the Global Rating Scale (GRS) at follow-up. Responsiveness was assessed by correlating score changes with the GRS and calculating the area under the receiver operating characteristic (ROC) curve (AUC). MIC was estimated using ROC anchor-based and 0.5 standard deviation (SD) distribution-based methods and then compared with the Smallest Detectable Change (SDC).

Among 102 participants, 52% were aged ≥ 60 years and 65.7% were female. Strong correlations were observed for overall score changes (rho = 0.815) with the GRS as well as for 84.2% of individual symptoms (rho = 0.701–0.897). The tool demonstrated good to strong discriminative ability (AUC = 0.739–0.975 for improvement; 0.764–0.977 for deterioration), with sensitivity and specificity ≥ 0.75 for 16 symptoms. The MIC values ranged from 0.5 to 1.5 using the ROC method and 0.9–1.8 using the 0.5 SD approach. The estimated MIC values exceeded the SDC for 84.2% of symptoms. Limited responsiveness (rho < 0.7, AUC < 0.7) and MIC values below the SDC were noted for forgetful, swollen legs/ankles and frequent urination/incontinent of urine.

The PHASE-20–Australian version is responsive for most symptoms, with a clinically meaningful change of approximately 2.0 points on a 0–10 scale. The estimated MIC is applicable at the individual level, although caution is needed for symptoms with an MIC below the SDC.

The online version contains supplementary material available at 10.1007/s11096-025-02045-4.