Table 2 Checklist for the most relevant aspects of comparative effectiveness research for Chinese medicine clinical studies

Designing comparative effectiveness research (CER) studie

1. Stakeholder involvement

All relevant stakeholders are involved in identification of research topic, plan and design of CER, interpretation of results

2. Efficacy-effectiveness continuum

Location on the efficacy-effectiveness continuum is determined for participant selection/eligibility criteria, treatment protocol, practitioner expertise, outcomes, and setting in which the study is conducted

3. Study design

Designs for multi-component interventions should be considered

Study population

4. Eligibility criteria

Should be as broad as possible in the context of available resources - Study population includes both CM-naïve and CM-non- naïve patients

5. Diagnoses

Recruitment of patients should follow Western diagnoses - CM diagnoses should be done whenever possible

6. Patient recruitment

Patients are recruited from site(s) where the treatment is usually provided

Treatment, expertise, and setting

7. Defining treatments

If intervention involves multi-component treatment the combination should be plausible and feasible in usual care - non-CM best practice alternatives are based on guidelines or broad expert consensus

8. Acupuncture

See [9]

9. Qi gong/tai chi

Style and setting should reflect typical community-based programs

10. Herbal medicine

Local and national regulations should be taken into account

11. Treatment documentation

Documentation reflects which treatments were received by all groups (interventions and co-interventions)

Outcomes

12. Measures

Widely accepted or standardized outcome measure used - secondary outcomes capture relevant patient-centered dimensions for the condition under study

13. Timing

Assessment schedule is balanced allowing study to acquire relevant data without substantial disruption of treatment or setting

Study design and statistical analysis

14. Allocation

Allocation is concealed - stratification for subgroups and/or dynamic allocation for key characteristics are used

15. Blinding

Outcome data are kept inaccessible to practitioners - blinded outcome rater is used if possible

16. Preferences/expectation

Preferences and expectations are measured at baseline

17. Sample size

Sample size takes patient heterogeneity into account - required sample size is feasible - Study has enough power for planned subgroup analyses

18. Subgroups

Relevant subgroups are pre-planned - exploratory subgroup analysis is mentioned in study aims

19. Statistical analysis

Intention-to-treat analyses are planned - relevant subgroup analyses are planned - data analyses are adjusted for stratification variables, baseline differences and relevant confounders

Economic evaluations

20. Relevance

Setting reflects reality in clinical practice

21. Methodological approach

Standard methods for economic evaluations are used - sensitivity analysis is employed for all relevant stakeholder perspectives - relevant subgroups are identified

22. Observation time

Long-term observation (≥12 months) are planned if possible

Publications

23. Guidelines

Relevant guidelines (CONSORT) are consulted and followed

24. Content

Statements of how and why this is CER are included - study setting (including practitioner selection procedure) is described in detail - treatment group description from informed consent is provided - comparison groups are described in detail - data are provided on all interventions and co-interventions received - relevant subgroup analyses are reported