Ref (type)	Population	Outcome, Interventions	Results and statistical analysis	Effect size	Favours
Pain relief
Systematic review	Women with primary dysmenorrhoea 32 RCTs in this analysis	Moderate or excellent pain relief (duration of treatment in the included studies varied from 1 cycle per treatment to 5 cycles per treatment) with NSAIDs with placebo Absolute results not reported	OR 4.50 95% CI 3.85 to 5.27 P <0.00001 Significant heterogeneity: I² = 53%, P = 0.00011 Analysis of individual NSAIDs versus placebo also significant See Further information about studies	Moderate effect size	NSAIDs
Systematic review	Women with primary dysmenorrhoea 2 RCTs in this analysis	Pain relief continuous data: percentage improvement in Visual Analogue Scale (VAS) pain score (scale 1–100) with NSAIDs (diclofenac) with placebo Absolute results not reported	Mean difference 65.96 95% CI 55.70 to 76.22 P <0.0001	Effect size not calculated	NSAIDs
Systematic review	Women with primary dysmenorrhoea 2 RCTs in this analysis	Pain relief continuous data: final pain relief score difference (time-weighted TOTPAR-8 scale) with NSAIDs with placebo Absolute results not reported	Mean difference 7.21 95% CI 4.65 to 9.76 P <0.00001	Effect size not calculated	NSAIDs
Systematic review	Women with primary dysmenorrhoea 2 RCTs in this analysis	Pain relief continuous data: final pain relief score difference (repeated 0–3 scale) with NSAIDs with placebo Absolute results not reported	Mean difference 4.83 95% CI 3.61 to 6.06 P <0.00001	Effect size not calculated	NSAIDs
RCT Crossover design 3-armed trial	149 women, aged 18–44 years, with primary dysmenorrhoea	Pain intensity, assessed by mean TOTPAR-8 scores over the first 8 hours 18.28 with celecoxib 12.82 with placebo	P <0.001	Effect size not calculated	celecoxib
RCT Crossover design 3-armed trial	149 women, aged 18–44 years, with primary dysmenorrhoea	Pain intensity, assessed by mean TOTPAR-8 scores over the first 8 hours 20.59 with naproxen sodium 12.82 with placebo	P <0.001	Effect size not calculated	naproxen sodium
RCT Crossover design 3-armed trial	149 women, aged 18–44 years, with primary dysmenorrhoea	Pain intensity, assessed by mean SPID-8 values over the first 8 hours 10.06 with celecoxib 5.96 with placebo	P <0.001	Effect size not calculated	celecoxib
RCT Crossover design 3-armed trial	149 women, aged 18–44 years, with primary dysmenorrhoea	Pain intensity, assessed by mean SPID-8 values over the first 8 hours 11.48 with naproxen sodium 5.96 with placebo	P <0.001	Effect size not calculated	naproxen sodium
RCT 3-armed trial	180 women with primary dysmenorrhoea	Pain scores, assessed by VAS (scale 0–10, higher scores indicating more severe pain) 2 months 3.6 with mefenamic acid 5 with placebo	P <0.01	Effect size not calculated	mefenamic acid
RCT 3-armed trial	180 women with primary dysmenorrhoea	Pain scores, assessed by VAS (scale 0–10, higher scores indicating more severe pain) 3 months 2.4 with mefenamic acid 6 with placebo	P <0.01	Effect size not calculated	mefenamic acid
RCT 3-armed trial	180 women with primary dysmenorrhoea	Pain duration 2 months 3 hours with mefenamic acid 16.2 hours with placebo	P <0.01	Effect size not calculated	mefenamic acid
RCT 3-armed trial	180 women with primary dysmenorrhoea	Pain duration 3 months 3 hours with mefenamic acid 15.4 hours with placebo	P <0.001	Effect size not calculated	mefenamic acid
Need for additional medication
Systematic review	990 women 13 RCTs in this analysis	Additional analgesics required with NSAIDs with placebo Absolute results not reported	OR 0.33 95% CI 0.26 to 0.42 P <0.00001 Significant heterogeneity: I² = 51%, P = 0.01 (see Further information about studies)	Moderate effect size	NSAIDs