Relationship between lung function and Modified Shuttle Test performance in adult patients with cystic fibrosis: a methodological issue

We were interested to read the paper by Doeleman WR and colleagues that was published in the June 2016 issue of Physiotherapy. The authors aimed to predict Modified Shuttle Test (MST) performance using by a cystic fibrosis (CF) specific equation. They developed a tool to make an a priori prediction of MST walking distance [1].

Although the statistical method is correct and data are interesting but as a rule of thumb, using by cross sectional studies unbiased prediction would not be achieved. The developed CF-specific equation may have limited generalizability because it did not validated. For validation of prediction models, we need at least two different sets of cohort data that established prediction model from sample one be validated in sample two [2].

As the authors point out in their conclusion, we provided a model for prediction of MST performance. Such a conclusion should be interpreted with caution.