Partsch 2004.

Methods	Study design: randomised clinical trial (3 treatment arms) Method of randomisation: sealed envelopes Concealment of allocation: yes, sealed envelopes were supplied by a statistician Blinding: no Follow‐up: 2 years Losses to follow‐up: 3 died, 1 had an ulcer, 7 were reported to be well without specific problems, and 5 could not be contacted
Participants	Country: Vienna, Austria No. of participants: 53 in primary study. For assessment of PTS incidence, 37 participants were reinvestigated 2 years later. Age: mean age 58 years (baseline data available from 45 participants in first study, Partsch 2000) Sex: males 29; females 16 (baseline data available from 45 participants in first study, Partsch 2000) Inclusion criteria: proximal DVT, confirmed by compression ultrasound scan or phlebography in mobile patients older than 18 years. Duration of symptoms had to be less than 14 days. Exclusion criteria: compression therapy or heparin therapy already started; massive symptomatic PE or need for thrombolysis or thrombectomy; ABI < 0.8; pregnancy or breastfeeding; and contraindication for compression, anticoagulants, or both Participants included 7 patients with previous DVT in the same leg and 11 with previous DVT in the contralateral leg.
Interventions	Group A (n = 18): inelastic compression bandages (Unna boots on the lower leg, adhesive bandages on the thigh) Group B (n = 18): thigh‐length compression stockings (class II, 30 mmHg) Group C (n = 17): bed rest for 9 days without compression After 9 days, all participants were encouraged to walk with compression class II stockings.
Outcomes	Reduced pain and leg circumference on days 0 and 9; on both days, PE and extent of DVT Clinical score calculated for the following symptoms: pain during walking; painful foot sole; painful calf palpation; subfascial oedema; increase in skin temperature; redness/cyanosis. Severity score comprised 4 stages: absent, mild, marked, and severe. Cumulative incidence of PTS
Notes	Information on cumulative incidence of PTS is available only for the mobile group (treatment arms A + B combined) vs the group with bed rest (arm C).
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No information available on allocation sequence generation
Allocation concealment (selection bias)	Unclear risk	Randomisation was performed with sealed envelopes. Methods of allocation and sequence generation are not described, and concealment of allocation cannot be judged.
Blinding (performance bias and detection bias) Objective outcome(s)	High risk	Participants and investigators were not blinded.
Blinding (performance bias and detection bias) Subjective outcome(s)	High risk	Participants and investigators were not blinded.
Incomplete outcome data (attrition bias) All outcomes	Low risk	Numbers and reasons for incomplete outcome data are described.
Selective reporting (reporting bias)	Low risk	All primary outcomes are reported. PTS was not a primary outcome. PTS scores are described as bed rest vs mobile. No information is available on the comparison of PTS scores for the 3 treatment groups, but it is judged unlikely that this introduces bias.
Other bias	Low risk	Study appears to be free of other sources of bias.