Becker 2010

Methods	RCT conducted at three urban ANC clinics in Tanzania between May 2003 and October 2004.
Participants	1521 women were recruited: 760 in the intervention (CVCT) group; 761 in the control (IVCT) group. In addition 254 male partners of the pregnant women in the CVCT group agreed to participate. Participants were married pregnant women presenting for the first ANC visit during the first six months of their pregnancy and their male partner for the intervention group. Women were considered as being married if they were involved in a legal marriage, a traditional marriage or if they have been living with the same actual male partner, father of the unborn baby, for at least two years.
Interventions	Intervention: Couple Voluntary Counselling and Testing (CVCT). Pregnant women attending their first ANC visit received an invitation letter, written by the medical officer, addressed to their male partners. The husband was asked to accompany his wife to the clinic for discussing health issues related to the pregnancy and the health of the baby. The women did not receive counselling and testing during that first visit and were asked to come back even if the husband refused to accompany her. Couple that return to the clinic together could receive each step of counselling and testing process together. If only one of the couple consented to VCT he/she received IVCT. If the couple consented to receive VCT together but not to collect their results together they received IVCT. Control group: Individual Voluntary Counselling and Testing (IVCT). Pregnant women attending their first ANC visit received IVCT during their first ANC visit and received usual care. All HIV positive women were provided with ARV prophylaxis that consisted of Nevirapine at delivery for both, mother and infant.
Outcomes	The primary outcomes were the proportion of couple and women receiving each step of the VCT and, within the women included in the intervention group the percentage of women who accepted the CVCT and the proportion of women who cross over to IVCT. A couple was considered as receiving complete CVCT if they received counselling, testing and result together. Secondary outcomes were the proportion of women who received Nevirapine at delivery, the use of condom during sexual intercourse and the proportion of cases of domestic violence.
Notes
Risk of bias
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quoted: "All consented women were randomised to either CVCT or IVCT (every other one, sequentially)." Not Randomized
Allocation concealment (selection bias)	High risk	Quoted: "All consented women were randomised to either CVCT or IVCT (every other one, sequentially)."
Blinding of participants and personnel (performance bias) All outcomes	Low risk	Quoted: "Women randomised to the control arm were then consented to receive IVCT immediately" . However the knowledge of the randomisation was unlikely to affect the results
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Quantitative Outcomes: not stated. Qualitative data (not included in the outcomes of the present review) quoted "2004, after consent a follow‐up questionnaire was administered by an interviewer blinded to the woman’s study": Low risk of bias
Incomplete outcome data (attrition bias) All outcomes	Low risk	Missing baseline demographics data for 0,5% of all the participants.
Selective reporting (reporting bias)	Unclear risk	All the outcomes presented in the protocol were also presented in the paper. However the protocol has been published after the study was concluded