Table 2.

GRADE Evidence Profile.

Certainty assessment							№ of patients		Effect		Certainty	Importance
№ of studies	Study design	Risk of bias	Inconsistency	Indirectness	Imprecision	Other considerations	lubricants during or prior to sex	no lubricants	Relative (95% CI)	Absolute (95% CI)
1. Vaginal dryness or pain during vaginal/anal penetration:Experience of pain during last insertive partnered sexual event (assessed with: self-report)
117	Observational studies	Seriousa	Not seriousb	Not serious	Seriousc	None	11/61 (18.0%)	3/21 (14.3%)	RR 1.26 (0.39–4.09)	37 more per 1000 (from 87 fewer to 441 more)	⊕◯◯◯ VERY LOW	CRITICAL
1. Vaginal dryness or pain during vaginal/anal penetration:Experience of pain during last receptive partnered sexual event (assessed with: self-report)
117	Observational studies	Very seriousa,d	Not seriousb	Not serious	Not serious	None	45/71 (63.4%)	3/17 (17.6%)	RR 3.59 (1.27–10.18)	457 more per 1000 (from 48 more to 1000 more)	⊕◯◯◯ VERY LOW	CRITICAL
1. Vaginal dryness or pain during vaginal/anal penetration:Degree of pain during last insertive partnered sexual event (assessed with: self-report, higher score indicates greater pain)
117	Observational studies	Seriousa	Not seriousb	Not serious	Not serious	None	2.3	2.9	MD 0.6 lower		⊕◯◯◯ VERY LOW	CRITICAL
1. Vaginal dryness or pain during vaginal/anal penetration: Degree of pain during last receptive partnered sexual event (assessed with: self-report, higher score indicates greater pain)
117	Observational studies	Seriousa	Not seriousb	Not serious	Not serious	None	2.2	3	MD 0.8 lower		⊕◯◯◯ VERY LOW	CRITICAL
1. Vaginal dryness or pain during vaginal/anal penetration:Dyspareunia (assessed with: visual analogue score pain assessment of dyspareunia, lower score is better outcome; Scale from: 0–10)
14	Observational studies	Seriousa	Not seriousb	Not serious	Seriouse	None	2.7	7	MD 4.3 lower		⊕◯◯◯ VERY LOW	CRITICAL
1. Vaginal dryness or pain during vaginal/anal penetration:Sexual discomfort (assessed with: Sexual Activity Questionnaire – Discomfort subscale assessing vaginal dryness and dyspareunia, higher score is better outcome; Scale from: 0–6)
14	Observational studies	Seriousa	Not seriousb	Not serious	Seriouse	None	2.9	0.8	MD 2.1 higher		⊕◯◯◯ VERY LOW	CRITICAL
3. Sexual desire, arousal, lubrication, orgasm, satisfaction, and pleasure:Female sexual well-being (assessed with: FSWB scale overall score)
120	Randomised trials	Seriousf	Not seriousb	Not serious	Not serious	None	Least-squares mean change in score from baseline vs end of study: Couple lubricant (n = 80) vs no lubricant (n = 82): 6.35 vs 1.94; Female lubricant (n = 82) vs no lubricant (n = 82): 3.99 vs 1.94g				⊕⊕⊕⃝ MODERATE	IMPORTANT
6. STIs/HIV:HPV incidence (one or more new HPV type(s) detected, among participants with no HPV detected at baseline) (follow up: mean 12 months; assessed with: PCR for HPV consensus probe)
121	Randomised trials	Not serioush	Not seriousb	Not serious	Seriousc	None	120/593 (20.2%)	131/587 (22.3%)	RR 0.91 (0.73–1.13)	20 fewer per 1000 (from 60 fewer to 29 more)	⊕⊕⊕⃝ MODERATE	IMPORTANT
6. STIs/HIV:HPV incidence (one or more new oncogenic HPV type(s) detected, among participants with no HPV detected at baseline) (follow up: mean 12 months; assessed with: PCR for HPV consensus probe)
121	Randomised trials	Not serioush	Not seriousb	Not serious	Seriousc	None	56/593 (9.4%)	51/587 (8.7%)	RR 1.09 (0.76–1.56)	8 more per 1000 (from 21 fewer to 49 more)	⊕⊕⊕⃝ MODERATE	IMPORTANT

Notes: CI: Confidence interval; RR: Risk ratio; MD: Mean difference; FSWB: Female sexual well-being; PCR: Polymerase chain reaction; HPV: Human papillomavirus.

^aRisk of bias: Certainty of evidence downgraded for self-report of pain.

^bInconsistency: This could not be evaluated, as there is only a single study.

^cImprecision: Certainty of evidence downgraded because 95% CI for RR includes both 1 (no effect) AND either appreciable harm (0.75) or appreciable benefit (1.25).

^dRisk of bias: Participants who reported using lubricant during their last partnered event were asked to indicate their reasons for using lubricant. The most highly endorsed statement (89.3%) was that lubricant reduced their pain/discomfort. This may indicate reverse causation between lubricant use and experience of pain.

^eImprecision: Certainty of evidence downgraded due to small sample size (n = 25).

^fRisk of bias: Certainty of evidence downgraded for detection bias. Blinding was not possible given the nature of the intervention, and outcome may have been affected by lack of blinding.

^gFSWB scale includes subscales, whose results (least-squares mean change in score from baseline vs end of study) are presented in this footnote.

Interpersonal domain: Couple lubricant (n = 80) vs no lubricant (n = 82): 1.80 vs 0.13; Female lubricant (n = 82) vs no lubricant (n = 82): 1.32 vs 0.13.

Cognitive-emotional domain: Couple lubricant (n = 80) vs no lubricant (n = 82): 2.48 vs 1.08; Female lubricant (n = 82) vs no lubricant (n = 82): 1.67 vs 1.08.

Physical arousal domain: Couple lubricant (n = 80) vs no lubricant (n = 82): 0.81 vs 0.72; Female lubricant (n = 82) vs no lubricant (n = 82): 0.07 vs 0.72.

Orgasm satisfaction domain: Couple lubricant (n = 80) vs no lubricant (n = 82): 1.43 vs 0.01; Female lubricant (n = 82) vs no lubricant (n = 82): 1.04 vs 0.01.

^hRisk of bias: Certainty of evidence not downgraded for detection bias. Blinding was not possible given the nature of the intervention, but outcome unlikely to have been affected by lack of blinding.