Physiotherapy for soft tissue shoulder disordersBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7130.555 (Published 14 February 1998) Cite this as: BMJ 1998;316:555
Conclusion that therapeutic ultrasound is ineffective was based on weak evidence
- Thomas Brockow, Research fellowa,
- Annegret Franke, Biostatisticiana,
- Karl L Resch, Directora
- a Balneology and Rehabilitation Sciences Research Institute, 08645 Bad Elster, Germany
- b Physiotherapy Centre for Rehabilitation, Derby City General Hospital, Derby DE22 3NE
- c Institute for Rehabilitation Research, PO Box 192, 6430 AD Hoensbroek, Netherlands
- d Institute for Extramural Medical Research, Vrije Universiteit, Amsterdam
Editor—Van der Heijden et al claim therapeutic ultrasound to be ineffective in treating patients with soft tissue shoulder disorders and recommend that treatment guidelines should be updated or reimbursements for the use of therapeutic ultrasound in soft tissue shoulder disorders should be withheld.1 Their conclusion, however, seems to be hasty: lack of evidence of effect is not evidence of lack of effect.
Their conclusion is based on a total of three sham controlled trials with adequate internal validity, which failed to show superiority of therapeutic ultrasound over sham treatment.2–4 In the other three trials of therapeutic ultrasound included in the analysis therapeutic ultrasound was not shown to be superior to reference therapies (active treatment).
Because of small group sizes the ß error of the reviewed studies clearly exceeds 20%. Downing and Weinstein themselves estimated the power of their study to be 0.3.2 The reliability of the outcome measures used (pain, mobility, activities of daily living) is not perfect. This also reduces study power and thus the validity of the evidence whether or not therapeutic ultrasound is effective.
Homogeneity of those three studies with respect to diagnosis and administration of the treatment was not assured. Besides diagnosis and stage of disease, dose variables (power density, sonation time per treatment, number of treatments, etc), mode of delivery (continuous or pulsed), and frequency used …