Shared standards for data analysis can accelerate breakthroughs in ALS research

Researchers at the University Medical Centre Utrecht have found that results from studies on ALS treatments are analysed in many different ways. This variation makes it difficult to compare studies with one another and reduces confidence in conclusions about whether a treatment is effective. These findings have been published in the scientific journal Neurology. 

Statistical analysis allows researchers to process large volumes of study data and determine whether a treatment truly has an effect. An international working group has now been established to define shared standards for analysing clinical trial results. 

Outcome measures in clinical studies

Outcome measures that track disease progression play a central role in clinical studies. They allow researchers to assess the potential effect of a treatment. One widely used measure in ALS research is the ALSFRS-R (ALS Functional Rating Scale Revised). This questionnaire evaluates how a person with ALS functions over time, including speech, swallowing, breathing, and physical activity.

Tools such as the ALSFRS-R enable researchers to measure disease progression in a consistent way across patients. However, there are currently no shared standards for how these data should be statistically analysed.

The study

Researchers at the ALS Center reviewed completed ALS studies to examine how ALSFRS-R outcomes are analysed. In total, the researchers analysed 45 studies involving 7,338 patients.

Different analysis methods

The study found that ALSFRS-R data had been analysed in 39 different ways across the reviewed studies. One reason for this variation is missing data: in clinical trials, data are often incomplete because patients may withdraw or pass away. To address this, researchers apply different statistical methods. These differences in approach make it difficult to compare results across studies.

In addition, more than half of the studies reviewed (55.6%) did not use all available ALSFRS-R measurements. As a result, not all patient data were used to their full potential. This reduces statistical precision and leads to less certainty about study outcomes.

Furthermore, 38.9% of the analysis methods used were found to increase the risk of false-positive results. This means that some treatments may be incorrectly identified as effective when they are not.

International collaboration

To address these issues, the researchers have established an international working group to develop shared standards for the statistical analysis of outcome measures such as the ALSFRS-R. The goal is to make better use of patient data and accelerate the development of effective ALS treatments. The implications of this work extend beyond ALS research. Clear and consistent agreements on statistical analysis are essential across the entire field of clinical research.