Comparison of certainty of evidence between the net benefit approach and the traditional GRADE method using the data of Japanese Clinical Practice Guidelines for Management of Sepsis and Septic Shock 2020

Table 2 Characteristics of clinical questions

	Group I^a	Group II^b	Group III^c
N	1 (5%)	10 (50%)	9 (45%)
The number of CQs where the direction (benefit or harm) of the point estimate in outcomes^d is consistent	0 (0%)	4 (40%)	1 (11%)
The number of outcomes of each CQ	5	3.5 (3–4)	3 (3–5)
The number of CQs where classification of the precision of each outcome
Included “serious”	1 (100%)	8 (80%)	8 (89%)
Included “very serious”	0 (0%)	9 (90%)	8 (89%)
The proportion of outcomes with insufficient OIS in each CQ (%)	0	100 (80–100)	100 (100–100)
The number of CQs where the outcome that critical for net effect estimate^e could not be identified	0 (0%)	1 (10%)	1 (11%)

CQ clinical question, OIS optimal information size
^aCertainty of evidence evaluated using net benefit approach was lower than that evaluated using the traditional method
^bThe certainty of evidence evaluated by net benefit approach was equal to that of the traditional method
^cCertainty of evidence evaluated by net benefit approach is higher than that by the traditional method
^dIn this table, “outcomes” means outcomes to be combined for calculating net effect estimate
^eThese outcomes are as follows: 1) outcomes for which the removal of the outcome would change the classification of the precision of the net effect estimate. 2) Outcomes for which the addition of plausible increases to the effect estimate (effect estimates with lower certainty) would change the classification

ISSN: 2052-0492