Table 1.

GRADE evaluation of interventions for MRSA: treating people with infection

Important outcomes	Clinical or microbiological cure, length of hospital stay, mortality, adverse effects
Number of studies (participants)	Outcome	Comparison	Type of evidence	Quality	Consistency	Directness	Effect size	GRADE	Comment
What are the effects of treatment for MRSA infections at any body site?
5 (73) [15]	Mortality	Linezolid v vancomycin in bacteraemia	4	–3	0	0	0	Very low	Quality points deducted for sparse data, methodological weaknesses (no blinding), and subgroup analysis
7 (900) [13] [14]	Clinical or microbiological cure	Linezolid v vancomycin in infection at any body site	4	–2	0	–2	0	Very low	Quality points deducted for incomplete reporting and methodological weaknesses (no blinding). Directness points deducted for population issues (inclusion of people with non-MRSA infections in some studies), unclear/subjective outcome (clinical cure), and inclusion of co-intervention in one study (aztreonam)
more than 1 RCT (>62 people) [13] [14]	Clinical or microbiological cure	Linezolid v vancomycin in nosocomial pneumonia	4	–3	0	–2	0	Very low	Quality points deducted for incomplete reporting, methodological weaknesses (no blinding), and subgroup analysis. Directness points deducted for population issues (inclusion of people with non-MRSA infections in some studies), unclear/subjective outcome (clinical cure), and inclusion of co-intervention in one study (aztreonam)
more than 1 RCT (>80 people) [13] [14]	Clinical or microbiological cure	Linezolid v vancomycin in skin and soft-tissue infections	4	–3	0	–2	0	Very low	Quality points deducted for incomplete reporting, methodological weaknesses (no blinding) and subgroup analysis. Directness points deducted for population issues (inclusion of people with non-MRSA infections in some studies), unclear/subjective outcome (clinical cure), and inclusion of co-intervention in one study (aztreonam)
12 (223) [15] [16]	Clinical or microbiological cure	Linezolid v vancomycin in bacteraemia	4	–2	0	–2	0	Very low	Quality points deducted for incomplete reporting and methodological weaknesses (no blinding). Directness points deducted for population issues (inclusion of people with non-MRSA infections in some studies, inclusion of children in some studies), and unclear/subjective outcome (clinical cure)
1 (182)[17]	Clinical or microbiological cure	Linezolid v teicoplanin	4	–2	0	–2	0	Very low	Quality points deducted for sparse data and inclusion of people without MRSA. Directness points deducted for low follow-up and use of unclear/subjective outcome (clinical cure)
1 (182)[17]	Mortality	Linezolid v teicoplanin	4	–2	0	–1	0	Very low	Quality points deducted for sparse data and inclusion of people without MRSA. Directness point deducted for highly selected population (on intensive care)
1 (51)[25]	Clinical or microbiological cure	Quinupristin–dalfopristin v vancomycin	4	–2	0	–1	0	Very low	Quality points deducted for sparse data and subgroup analysis. Directness point deducted for use of unclear/subjective outcome (clinical cure)
1 (298)[25]	Mortality	Quinupristin–dalfopristin v vancomycin	4	–1	0	–1	0	Low	Quality point deducted for inclusion of people without MRSA. Directness point deducted for lack of subgroup analysis in people with MRSA only, hence, limited generalisability to this population group
1 (47)[26]	Clinical or microbiological cure	Trimethoprim–sulfamethoxazole (co-trimoxazole) v vancomycin	4	–2	0	–2	0	Very low	Quality points deducted for sparse data and subgroup analysis. Directness points deducted for highly selected population (injecting drug users) and for use of unclear/subjective outcome (clinical cure)
1 (157) [24]	Clinical or microbiological cure	Tigecycline v vancomycin	4	–2	0	0	0	Low	Quality points deducted for sparse data and no statistical assessment

Type of evidence: 4 = RCT; Consistency: similarity of results across studies; Directness: generalisability of population or outcomes; Effect size: based on relative risk or odds ratio.