Properties of Measurements of Patient-Reported Outcome Measures Physical Activity Assessment in Chronic Kidney Disease: A Systematic Review.

Only 17 studies were found, and the authors describe the evidence as heterogeneous, with wide variation in which questionnaire was used, how it was tested, and how well it was done. Many of the included studies had small samples. The evidence could not be separated by stage of kidney disease, and the authors say questionnaire performance may differ between stages. Most studies looked at severe disease, so a questionnaire tested in dialysis patients may not behave the same way in your earlier-stage patient. No included study measured responsiveness, so there is no evidence here on whether any of these questionnaires can pick up change after treatment. Grey literature was not searched. The authors ran a secondary search of the reference lists of included studies to reduce the chance of missing papers, but reports outside indexed journals would not have been found. The findings are for adults only and the authors state they cannot be extended to children. The two quality-assessment tools disagreed with each other on 5 of the 11 questionnaires (45.5%), so which tool was used changed the verdict for nearly half of them. The authors say the choice of tool should follow the assessor's aims and expertise, which means a single rating should not be read as final. The comparison monitors used to test validity are themselves described by the authors as reporting low sensitivity in distinguishing activity levels in mostly inactive populations, which may make the questionnaires look worse than they are.

The authors declare no conflicts of interest. The declarations of funding, ethics and consent all state that the authors have nothing to report.

Do not read the poor validity results as proof that these questionnaires do not work. Most of them were checked against wearable activity monitors, which the paper says are poor at telling different activity levels apart in people who are mostly inactive, so a weak correlation may say as much about the monitor as about the questionnaire. The one questionnaire rated well by both quality tools, CKD-PAQ, rests on a single study that tested only its validity.