Summary of findings 1. Comparison of two prophylaxis regimens.

Prophylaxis regimen compared with another prophylaxis regimen for previously treated individuals with haemophilia A or B
Patient or population: children or adults with hemophilia A or B Settings: outpatient Intervention: secondary prophylaxis Comparison: secondary prophylaxis
Outcomes	Illustrative comparative risks* (95% CI)		Relative effect (95% CI)	No of Participants (studies)	Certainty of the evidence (GRADE)	Comments
	Assumed risk	Corresponding risk
	Prophylaxis regimen	Prophylaxis regimen
Number of joint bleeding episodes per year (AJBR)    Follow‐up: 12 months	No difference was seen between prophylaxis regimens in any of the studies. Thrice‐weekly higher dose prophylaxis regimen compared to a twice‐weekly lower dose regimen, MD ‐1.70 (95% CI ‐5.06 to 1.66) (LEOPOLD II 2015).    PK‐guided prophylaxis targeting trough levels of 8% to 12% compared to targeting trough levels of 1% to 3%, MD ‐1.50 (95% CI ‐3.54 to 0.54) (n = 115 participants) (PROPEL III 2020).   Low frequency prophylaxis (100 IU / kg once a week) compared to standard frequency regimen (50 IU / kg twice a week, MD of 1.70 (95% CI ‐1.09 to 4.49) (Valentino 2014).		N/A	219 participants (3 trials)	⊕⊕⊝⊝ lowa	We were unable to combine results in a meta‐analysis due to the different prophylaxis regimens used in each trial.
Number of total bleeds per year (ABR)   Follow‐up: 12 months	There was no difference in total number of bleeds between prophylactic regimens in five trials (Aronstam 1977; LEOPOLD II 2015; PROPEL III 2020; Valentino 2012; Valentino 2014).   A twice‐a‐week regimen (7.5 IU/kg) was favoured over a once‐a‐week regimen (15 IU/kg), MD 11.20 (5.81 to 16.59) (Morfini 1976) and a prophylaxis group with dosing producing at least 0.25 IU/mL of factor VIII showed a significant reduction in overall bleeding frequency compared to a dosing regimen producing at least 0.01IU/mL once weekly, MD 3.44 (95% CI 2.42 to 4.46) (Aronstam 1976).		N/A	310 participants (7 trials)	⊕⊝⊝⊝ lowb,c	Due to heterogeneity of intervention and design, none of the trials we were unable to combine data from any of the trials (LEOPOLD II 2015).
Treatment‐related adverse events   Follow‐up: 32 weeks to 12 months	One trial reported no difference in total treatment‐emergent adverse events, MD 1.00 (95% CI 0.54 to 1.84) at 32 weeks (Valentino 2014). A further trial reported no difference between treatment regimens in mean rates of adverse events (Valentino 2012). In the study targeting different trough levels, no serious adverse event was treatment‐related in the arm targeting trough levels of of 1% to ‐3%, and in the arm targeting trough levels of 8% to ‐12%, one serious adverse event was estimated to be treatment‐related (PROPEL III 2020).		N/A	223  participants (3 trials)	⊕⊝⊝⊝ very lowa,d	Three trials did not report the rate of adverse events by treatment groups (Aronstam 1977; LEOPOLD II 2015; Morfini 1976). The LEOPOLD II trial reported three treatment related  adverse events but gave no further detail (LEOPOLD II 2015). There was no reported inhibitor development reported in six of the trials in this comparison (Aronstam 1976; Aronstam 1977; LEOPOLD II 2015; Morfini 1976; Valentino 2012; Valentino 2014).
*The basis for the assumed risk (e.g. the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). ABR: annualised bleed rate; AJBR: annualised joint bleed rate; CI: confidence interval; FIX: factor IX; RR: risk ratio; MD: mean difference.
GRADE Working Group grades of evidence High certainty: further research is very unlikely to change our confidence in the estimate of effect. Moderate certainty: further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate. Low certainty: further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate. Very low certainty: we are very uncertain about the estimate.

a. Downgraded twice due to risk of bias in the included trials, particularly across the domains of randomisation and allocation concealment. The trials were also considererd at high risk of bias due to lack of blinding 
b. Downgraded once due to imprecision as a result of small sample sizes. Although the total number of participants included in this outcome is 390, none of the studies could be combined and so we have based our assessment on the numbers in individual trials. The two trials that showed a difference between regimens included nine and 10 participants.

c. Downgraded twice due to an unclear or high risk of bias across many of the domains with particular concern around randomisation procedures, allocation concealment and blinding.

d. Downgraded once due to imprecision from small sample size and low event rates. Although the total number of participants is reasonable, none of the trials could be combined and so we have based our judgement on the numbers in the individual trials.