Gatty 2004.

Methods	RCT. The participants were randomly assigned to 1 of 2 groups.
Participants	"All participants were female and met the inclusion criteria by being employed as full‐time clerical/office workers at a small western Pennsylvania college, and having no newly (within the last three months) diagnosed MSD". 15 workers participated in the study.
Interventions	The study compared individualised WIPPs vs no intervention Individualised WIPPs (group A) The WIPP were designed by the WIPP team (3 master of occupational therapy students and the principle investigator) was based on the work site analyses. Treatment sessions spanned weeks 1 through to 4. Each participant received 1 hour of treatment per week. During these 4 sessions the workers were actively engaged in education, workstation redesign, and task modification. Education: occupational therapy students and clerical workers discussed current work conditions as they related to experienced symptoms; for example, improperly bending to lift boxes may contribute to low back pain or excessive wrist extension may contribute to wrist pain. workstation redesign: based on work site analyses and input from the workers. Task modification: demonstrated by the occupational therapy student, practiced by the worker, and feedback was provided. No intervention (control) (group B) This group received no intervention. All participants (intervention and control group) received the symptom evaluation measure (measured the reported frequency and intensity of symptoms), stress and energy scale (10‐cm VAS to measure perceived stress energy levels), and follow‐up survey (to identify changes in work status)
Outcomes	Primary outcomes Frequency of symptoms: neck ache/pain, shoulder ache/pain, elbow‐forearm ache/pain, wrist‐hand ache/pain, upper back ache/pain, and lower back ache/pain defined as the number of days, 0 to 5, they experienced symptoms during the week while at work (data were collected at weeks 0, 5, and 16). Symptom intensity: rated using a 4‐point Likert scale 1 to 4: 1 = none, 2 = mild, 3 = moderate, or 4 = severe. Secondary outcome Compliance survey – for group A (intervention) only ‐ about: how often they used the issued ergonomic equipment, how often they performed recommended stretches and whether or not they performed their job duties differently based on recommendations. Responses were elicited on a 4‐point Likert scale with choices of 1 = never, 2 = sometimes, 3 = usually, 4 = always when I should.
Notes	The study was reported in 2 papers (see Gatty 2004): Martin SA, Work 2003;21:185‐96, reported results for weeks 0 and 5; Gatty CM, Work 2004;23:131‐7, reported results for weeks 0, 5, and 16. Worksite analyses were conducted for group B (control) workers during week 17, they received individualised WIPPs during weeks 18 to 21 and measures were repeated at week 22 (suspension of randomisation process).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	There was no mention of sequence generation. The only information given was: "This was a two‐phased randomized control pilot study with between and within subject comparisons … Participants were randomly assigned to one of two groups, A (intervention) or B (control)".
Allocation concealment (selection bias)	Unclear risk	No information provided on allocation concealment.
Blinding of participants and personnel (performance bias) All outcomes	High risk	There was no information on blinding and since the intervention consists of education, workstation redesign, and task modification, there was high risk for bias.
Blinding of outcome assessment (detection bias) Musculoskeletal disorders	High risk	The outcome was subjective reporting of symptoms frequency and intensity.
Incomplete outcome data (attrition bias) All outcomes	High risk	In group A (intervention), "one non‐compliant worker at week zero remained non‐compliant at week five and was dropped from the study.  One person was no longer employed by week 16 and membership decreased to six". In group B (control) "…Although there were originally eight participants, two different workers were non‐compliant with surveys, one at week zero and one at week five. By week 16, one person had left employment". Owing to the small number of participants, i.e. 16, the attrition of 3 participants was considered to induce a high risk of bias.
Selective reporting (reporting bias)	Low risk	All the outcomes were reported.
Other bias	High risk	Difference in baseline data: group A (intervention) reported lower average wrist‐hand ache/pain and upper back ache/pain intensities than group B (control). There was no mention of differences between participants and non‐participants.