Monday, 27 February 2017

Outcomes reported in a cluster-randomized controlled trial with the socially assistive robot Paro

Jose M. Moran, PhD
Metabolic Bone Diseases Research Group. Nursing Department,
University of Extremadura, Spain


Commentary on Jøranson N., Pedersen I., Rokstad A.M.M. & Ihlebæk C. (2016) Change in quality of life in older people with dementia participating in Paro-activity: a cluster-randomized controlled trial. Journal of Advanced Nursing 72(12), 3020–3033


We read with interest the recent paper from Jøranson and colleagues (Joranson et al. 2016) about the changes in quality of life in older people with dementia participating in Paro-activity. As indicated by the authors the trial was in adherence with the CONSORT statement and was registered at ClinicalTrials.gov with the record NCT02008630. We would like to highlight that authors report outcomes and experimental design are both different to those initially registered (Clinicaltrials.gov, 2013)

The overall objective of the trial NCT02008630 was to investigate changes in measures related to the risk of falls among elderly with dementia participating in animal-assisted interventions. The prespecified intervention consisted of 30 minutes sessions with animal-assisted activity or animal-assisted therapy two times a week for 12 weeks in groups of 4-6 participants. The prespecified primary (Changes in Norwegian version of Berg balance test) was not reported and was changed to the Brief Agitation Rating Scale (BARS). Prespecified secondary outcomes changes in Norwegian version of Quality of Life in Late-Stage Dementia (QUALID)) and changes in optional and ordinary medication were reported. Other two prespecified outcomes were not reported (changes in physical activity measured by Actigraphy and video recordings of two group sessions at each location). Additionally prespecified design included three arms (not two) with animal-assisted (dog) activity/therapy and no intervention. No interventions with the Paro assistive robot were prespecified. Some prespecified outcomes were published by the authors in Geriatric Nursing (Olsen et al. 2016), a manuscript that reported data from the effect of animal-assisted activity on balance and quality of life in home-dwelling persons with dementia, and was also linked to the trial NCT02008630 but reports data from a sample that differs from the recently reported in Journal of Advanced Nursing.

The Journal of Advanced Nursing endorses the CONSORT (Consolidated Standards of Reporting Trials) guidelines on best practice in clinical trials reporting (Moher et al. 2010). Any changes to trial outcomes after the trial commenced should be declared (with reasons) in the methods section to avoid selective outcome reporting (Williamson and Gamble 2005) that may produce an overestimation of benefit. As the adherence to the CONSORT statement is mandatory by the authors that published in the Journal of Advanced Nursing such discrepancies should be noticed in the peer review process. We understand that the concerns expressed here might not affect to the overall conclusions presented by Jøranson and colleagues but certainly those results are not in adherence with the CONSORT statement as the authors declared in their manuscript.


Jose M. Moran, PhD
Metabolic Bone Diseases Research Group. Nursing Department, University of Extremadura. Avd. Universidad S/N, Cáceres, Spain.
e-mail: jmmorang@unex.es


References

ClinicalTrials.gov: NCT02008630. 2013. Animal-assisted Interventions in Health Promotion for Elderly With Dementia. Accessed at https://clinicaltrials.gov/show/NCT02008630 on 22 January 2017.

Joranson N., Pedersen I., Rokstad A.M. & Ihlebaek C. (2016). Change in quality oflife in older people with dementia participating in Paro-activity: acluster-randomized controlled trial. Journal of Advanced Nursing 72(12), 3020-3033. doi: 10.1111/jan.13076

Moher D., Hopewell S., Schulz K.F., Montori V., Gotzsche P.C., Devereaux P.J., Elbourne D., Egger M., & Altman D.G. (2010). CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials. British Medical Journal 340, c869. doi: 10.1136/bmj.c869.

Olsen C., Pedersen I., Bergland A., Enders-Slegers M.J., & Ihlebaek C. (2016). Effect of animal-assisted activity on balance and quality of life in home-dwelling persons with dementia. Geriatric Nursing 37(4), 284-291. doi: 10.1016/j.gerinurse.2016.04.002.

Williamson P.R. & Gamble C. (2005). Identification and impact of outcome selection bias in meta-analysis. Statistics in Medicine 24(10), 1547-1561. doi: 10.1002/sim.2025.


No comments:

Post a Comment