Table 4.

Summary of findings: nonutility results for burden of treatments

Health state/outcome (categories of values and preferences)	Estimates	Certainty in evidence
	No. of participants/studies
Prophylaxis
Treatment burden of mechanical methods (Anand and Asumu,39 Brady et al,44 Chan et al80)	More patients wearing thigh-length SCDs and TEDs complained of discomfort compared with patients wearing knee-length SCDs and TEDs. Many patients using foot pumps reported sleep disturbance (range, 28%-57%), “heat intolerance” (43%), and found them soothing (54%).	⊕⊕○○ Low certainty due to very serious RoB* and indirectness†
	210 participants from 3 cross-sectional studies
Treatment
Treatment burden of DOACs (Zolfaghari et al65)	Some patients (4/33; 12%) using DOACs may switch to VKAs due to complications, such as hair loss, after 9 mo of treatment or general discomfort, due to fear of side effects on DOACs, or because of no reimbursement from the health insurer.	⊕⊕○○ Low certainty due to RoB‡ and imprecision§
	19 participants (subgroup of 1001 participants study sample) from 1 cross-sectional study
Treatment burden of VKAs (Attaya et al,40 Brekelman et al,45 Elewa et al,47 Lutsey et al,52 Zolfaghari et al65)	Although routine monitoring does not represent a limitation for patients accepting the VKA, many patients would like to switch to an equally effective anticoagulant, primarily due to the burden associated with monitoring and dietary change (58%-64% of patients; mean willingness to change, 3.3 [on a scale of 1-5]).	⊕⊕○○ Low certainty due to serious RoB¶ and inconsistency||
	2070 participants from 5 cross-sectional studies
Treatment burden of ECSs (Bouman et al43)	Significant determinants of preference were PTS risk reduction, putting on ECSs, duration of ECS therapy, reduction in current complaints, comfort of wearing ECSs, and ease of washing ECSs. Cost and appearance of ECSs did not significantly influence preference.	⊕⊕⊕○ Moderate due to serious RoB#
	300 participants from 1 RCT
Trade-off between treatment burden and benefits with VKAs (Locadia et al32)	Men were willing to take greater risks regarding recurrent VTE during cessation of treatment than women. Patients with a low educational level were more willing to opt for continuation of treatment, regardless of the risk for VTE, compared with patients with a medium or high educational level.	⊕⊕⊕○ Moderate certainty due to serious imprecision**
	124 participants from 1 case-control study
Cancer
Treatment burden of injection (Lemke et al,51 Maxwell et al,53 Noble et al,54 Sousou and Khorana60)	Many patients (range across studies, 46%-55.7%) had a positive attitude toward receiving shots, and very few (4%) reported swelling, pain, or anxiety related to the shot.	⊕⊕⊕○ Moderate certainty due to RoB††
	601 participants from 4 cross-sectional studies
Treatment burden of EPCs (Maxwell et al53)	The majority of patients were satisfied with treatment with EPC, although 26% of patients experienced discomfort, inconvenience, problems, and/or side effects related to EPC.	⊕⊕○○ Low certainty due to RoB† and imprecision‡‡
	211 participants from 1 cross-sectional study
Trade-off between treatment burden and benefits with LMWH (Cajfinger et al,46 Noble et al54)	Cancer patients place highest value on “the interference with cancer treatment,” followed by “efficacy of the VTE treatment” and “risk for major bleeding.” They place low value on monitoring through blood tests, frequency of administration, mistakes, and costs.	⊕⊕⊕○ Moderate certainty due to RoBa
	509 participants from 2 cross-sectional studies
Pregnancy
Willingness to be treated with LMWH (Bates et al24)	The majority of women with a previous VTE (76%) were willing to take LMWH prophylaxis throughout the antepartum period: 86% of women at high risk and 60% at low risk for recurrent VTE. The threshold reduction in VTE risk at which they would accept the use of LMWH was 2% higher for women with <2 wk of previous experience with LMWH during pregnancy (vs with >2 wk of LMWH experience) and 1.6% higher for those who were pregnant or planning pregnancy (vs neither pregnant nor planning a pregnancy).	⊕⊕⊕○ Moderate certainty due to imprecision**
	123 participants from 1 cross-sectional study

Moderate certainty in evidence: We are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty in evidence: Our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.

DOAC, direct oral anticoagulant; ECS, elastic compression stocking; EPC, external pneumatic compression; RCT, randomized controlled trial; SCD, sequential compression device; TED, thromboembolic deterrent stocking.

^*Brady et al⁴⁴ assessed SCDs and stockings, whereas the other 2 studies assessed foot pump devices.

^†The studies directly asked about the administration route the participants would prefer, without describing the possible consequence of the treatment. High RoB for health state presentation and measurement instrument.

^‡Patients undergoing anticoagulation treatment, very likely in long-term treatment.

^§Although the overall sample size is large, this information comes from a subgroup of 19 patients who changed from DOACs to VKAs, not included in the final analysis.

^¶Two studies directly asked which administration route the participants would prefer, without describing the possible consequence of the treatment, with high RoB for measurement instrument and unknown sampling methods and high risk in health state presentation.

^||Studies’ findings showing variability in the value of the monitoring burden.

^#Bouman et al⁴³ discrete choice analysis of results with RoB bias due to unknown sampling methods and high risk in health state presentation.

^**Imprecision due to small sample size.

^††Sousou and Khorana⁶⁰ with high RoB for health state presentation and, together with Maxwell et al⁵³ and Lemke et al,⁵¹ measurement instrument; the sampling methods are unclear in 2 studies, and the response rate is unknown in 2 studies.

^‡‡Maxwell et al⁵³ used forced choice method to elicit values and preferences, and the response rate was not reported.

^aCajfinger et al⁴⁶ directly asked which administration route the participants would prefer, without describing the possible consequence of the treatment; the response rate is unknown in both studies.