Study designs of evaluations included in the review
Randomised controlled trials (RCTs) were included. The follow-up periods were reported to be 3 to 12 months in the table of study characteristics, but results for a 2-year follow-up were reported in the text.
Specific interventions included in the review
The types of intervention strategies included:
'step-up', defined as trials in which patients with insufficient benefit from one second-line agent continued the use of this drug and had another (or placebo) added to this;
'parallel', defined as starting a combination of new drugs; and
'step-down' defined as sequential withdrawal of simultaneously started drugs.
Step-up studies included the following comparisons: cyclosporin with prednisolone as anchor; monoclonal anti-CD4 antibody with methotrexate as anchor; cyclosporin with methotrexate as anchor; and each of cyclosporin, bucillamine and hydroxychloroquine with gold as anchor.
Parallel studies included the following comparisons: hydroxychloroquine with gold as anchor; sulphasalazine plus hydroxychloroquine versus each drug alone; D-penicillamine plus chloroquine versus each drug alone; dapsone plus hydroxychloroquine versus each drug alone; hydroxychloroquine plus methotrexate versus hydroxychloroquine; methotrexate plus chloroquine versus methotrexate; methotrexate plus sulphasalazine plus hydroxychloroquine versus sulphasalazine plus hydroxychloroquine versus methotrexate; methotrexate plus auranofin versus each drug alone; methotrexate plus azathioprine versus each drug alone; and methotrexate plus sulphasalazine versus each drug alone.
Step-down studies included the following: sulphasalazine plus methotrexate plus prednisolone versus sulphasalazine; gold plus prednisolone versus gold; gold plus methylprednisolone pulses versus gold; and sulphasalazine plus methylprednisolone versus sulphasalazine.
Participants included in the review
The participants were patients with rheumatoid arthritis, who had had the disease from less than one year to 11 years.
Outcomes assessed in the review
Efficacy was assessed using the core set measures of the World Health Organization and International League of Associations for Rheumatology. When less than 4 of these measures were assessed, first grip strength and second morning stiffness were selected as well.
How were decisions on the relevance of primary studies made?
The titles and abstracts (when available) were screened by one of the authors.