Reproducibility of pressure pain threshold testing in physiotherapy students, a rater dependent skill.

Introduction: Although experienced raters show good reproducibility1 of Pressure Pain Thresholds (PPT) in most body locations2, it is unknown if inexperienced physiotherapy students perform comparably. Recruiting students as raters would increase opportunities for the student's development and the design of studies. Methods: A total of 175 participants were measured by ten undergraduate students using either a Wagner (n = 87) or Somedic (n = 88) digital algometer on four body locations; the lower leg, upper leg, arm and neck. Each student measured a subsample of ~9 participants. Agreement (reported as a coefficient of variance; standard error of measurement divided by the average threshold) and reliability (intraclass correlation (ICC)) were measured for both the individual rater and the total group. Results: Agreement ranged from 9-17% per body location, whereas individual raters differed from the group result from -7.9% up to +6.6%. Reliability was good to excellent, ICCs (1,1) ranging from 0.88 to 0.95. Individual ICCs (2,1) were between 0.00 to 0.99. Discussion: Reproducibility of the total group was good; however, it varies per student. Even though reliability is affected by group variance differences, agreement parameters also show individual differences. PPTs seem rater dependent, which may explain different findings in the literature. Consequently, using reproducibility values of other studies/raters should be done cautiously. Students may be recruited as research assistants to perform PPTs. Process evaluation: Organising raters and participants together is challenging. Using parallel rooms with observational tools helped. Dependency on raters was a threat, resulting in an excluded rater due to sickness. [ABSTRACT FROM AUTHOR]