Table 12.

Gradepro assessment

Question: What is the incidence of neurosensory disturbance in patients with mandibular implants
Number of studies	Certainty assessment						Effect			Certainty	Importance
	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	Number of events	Number of individuals	Rate (95% CI)
Incidence of NSD
9	Observational studies	Not serious	Very seriousa	Not serious	Seriousb	Strong association dose-response gradient	105	2112	Event rate 13.50/100 person-year(s) (10.98-16.03)	⨁⨁⨁◯ Moderate	Critical
Rate of recovery
7	Observational studies	Not serious	Very seriousc	Not serious	Seriousb	Strong association dose-response gradient	60	95	Event rate 51.30/100 person-year(s) (31.21-71.4)	⨁⨁⨁◯ Moderate	Critical
Incidence of permanent NSD
6	Observational studies	Not serious	Very seriousd	Not serious	Seriousb	Strong association dose-response gradient	34	90	Event rate 18.67/100 person-year(s) (14.54-22.79)	⨁⨁⨁◯ Moderate	Important
Short-term recovery rate
2	Observational studies	Not serious	Very seriouse	Not serious	Seriousb	Strong association dose-response gradient	10	11	Event rate 46.96/100 person-year(s) (−10.77-104.69)	⨁⨁⨁◯ Moderate	Important
Long-term recovery rate
3	Observational studies	Not serious	Very seriousf	Not serious	Seriousb	Strong association dose-response gradient	26	36	Event rate 10.97/100 person-year(s) (3.51-18.43)	⨁⨁◯◯ Low	Important
Risk of NSD in anterior versus posterior implants
2	Observational studies	Not serious	Seriousg	Not serious	Very seriousb	Very strong association dose-response gradient	15	177	Event rate −0.02/1 person-year(s) (−0.21-0.16)	⨁⨁⨁⨁ High	Critical

⨁◯The certaintly of evidence is graded as very low to high with upto four crosshairs. One crosshair representing one point to mean very low while four crosshairs represent high certainty of evidence ^aThe lack of overlap in the CIs of the studies included and the high estimated heterogeneity (I²=99.8%) result in a high risk of inconsistency, ^bSmall sample size resulting in a serious risk for imprecision, ^cThe lack of overlap in the CIs of the studies included and the high estimated heterogeneity (I²=99.90%) result in a high risk of inconsistency, ^dThe lack of CI overlap in the studies included and the high estimated heterogeneity (I²=99.77%) result in a high risk of inconsistency, ^eThe lack of CI overlap in the studies included and the high estimated heterogeneity (I²=98.28%) result in a high risk of inconsistency, ^fThe lack of CI overlap in the included studies and the high estimated heterogeneity (I²=99.77%) result in a serious risk of inconsistency, ^gThe lack of CI overlap in the included studies and the high estimated heterogeneity (I²=71.94%) result in a serious risk of inconsistency. NSD: Neurosensory disturbance, CIs: Confidence intervals