Included study characteristics
| Study | Location | Sample size | Age, years mean (range/SD) | Study design | Meta-analysis (yes/no) | Outcome measures | Main finding |
| Farr et al26 | USA | 106 children (56 ([61.9%] LG vs 50 ([38.1%] heterosexual) | Children: 3.01 (1.08-6/1.31); parents: 41.69 (30-60/5.51) | Cross-sectional | Yes | Children’s psychological adjustment; children’s gender role behaviour; couple relationship satisfaction; family functioning*† | No significant differences were found between same-sex parent families and heterosexual parent families in children’s psychological adjustment, gender role behaviour, couple relationship satisfaction, family functioning. |
| Goldberg and Smith27 | USA | 174 parents (107 ([61.5%] LG vs 67 ([38.5%] heterosexual) | Children: 5.42; parents: NA | Cohort | Yes | Children’s psychological adjustment; parents’ mental health; other outcomes: parent-school relationships* | Same-sex parents reported significantly lower levels of child internalising symptoms than heterosexual parents, but not in child externalising symptoms. No significant differences were found in depression. Other related outcomes: Gay men reported the highest parent-teacher relationships quality (3.47 (0.06)) and heterosexual men the lowest (3.05 (0.68)). Lesbians reported the least acceptance by other parents (4.03 (0.10)) than gay (4.39 (0.09)) and heterosexual parents (4.30 (0.66)). Lesbian mothers were the most likely to report negative adoption-related experiences with teachers (20.4%) than gay (5.4%) and heterosexual parents (9.1%). There were no other significant differences for parent-school variables. |
| Bos et al30 | Netherlands | 102 children (51 ([50%] L vs 51 ([50%] heterosexual) | Children: 15.9 (1.30); parents: NA | Cohort | Yes | Children’s psychological adjustment; parent–child relationship; other outcomes: adolescents substance use‡ | The adolescents with lesbian mothers scored significantly lower on the conduct problems and higher on self-esteem than their counterparts in heterosexual families. No significant differences were found in parent–child relationship quality. Other outcomes: No significant differences were found between adolescents in lesbian and heterosexual households in the consumption of cigarettes (0.76 (1.30) vs 0.76 (1.59)), alcohol ((0.60 (0.70) vs 0.82 (1.75))), and marijuana/hashish (0.26 (0.66) vs 0.35 (1.54)). |
| Bos et al31 | Netherlands | 200 families (100 ([50%] L vs 100 ([50%] heterosexual) | Children: 5.95 (4-8/1.30); parents: NA | Cross-sectional | Yes | Children’s psychological adjustment; couple relationship satisfaction* | No significant differences were found in total problem behaviour. Lesbian mothers seem to be significantly more satisfied with their partner as a co-parent than heterosexual mothers are. |
| Gartrell and Bos33 | USA | 171 children (78 ([45.6%] L vs 93 ([54.4%] Achenbach’s normative sample) | Children: 17.05; parents: NA | Cohort | Yes | Children’s psychological adjustment; children’s educational outcomes* | The children with lesbian mothers were rated significantly lower in externalising problem behaviour than the normative sample, not in internalising problem behaviour. The adolescents with lesbian mothers demonstrated higher levels of school/academic competence than gender-matched normative samples (5.2 (0.9) vs 2.8 (0.9)). |
| Averett et al25 | USA | 1384 children (155 ([11.20%] L vs 1,229 ([88.80%] heterosexual) | Children: (1.5–18); parents: 47.80 (8.64) | Cross-sectional | Yes | Children’s psychological adjustment; family functioning* | Same-sex parents reported significantly lower in externalising problems of their adopted children than heterosexual parents reported, but not in internalising problems. In 1.5- to 5-year-old group, no statistically significant differences were found in family functioning. In 6- to 18-year-old group, Same-sex parents reported significantly lower levels of family functioning. |
| McConnachie et al36 | UK | 111 families (68 ([61.3%] LG vs 43 ([38.7%] heterosexual) | Children:11.85 (10-14/1.20); parents: 48.26 (6.09) | Cohort | Yes | Children’s psychological adjustment; parents’ mental health; parent–child relationship*†‡ | No significant differences were found between same-sex parent families and heterosexual parent families in terms of children’s psychological adjustment, parent mental health and parent–child relationship. |
| Bos et al32 | The Netherlands | 190 families (95 ([50%] LG vs 95 ([50%] heterosexual) | Children:11; parents: 40 | Cross-sectional | Yes | Children’s psychological adjustment; parenting stress; social support* | No significant differences were found between same-sex and heterosexual families in children’s psychological adjustment, parenting stress and the use of formal and informal support(formal: 0.73 (0.87) vs 0.70 (1.04); informal: 1.95 (0.77) vs 2.03 (0.66)). |
| Baiocco et al28 | Italy | 390 children (195 ([50%] LG vs 195 ([50%] heterosexual) | Children: 4.33 (3-11/1.83); parents: 39.17 (7.04) | Cross-sectional | Yes | Children’s psychological adjustment; couple relationship satisfaction; family functioning; other outcomes: parental self-agency* | Gay and lesbian parent families reported fewer child psychological problems, higher levels of couple relationship satisfaction and family functioning, than did different-sex parent families. Other outcomes: Gay and lesbian parent families reported higher levels of parental self-agency (27.25 (4.25), 26.46 (4.07)) than did different-sex parent families (25.37 (4.62)), with no differences found between the gay and lesbian parent families. |
| Calzo et al18 | USA | 21 103 children (296 ([1.40%] LGB vs 20,807 ([98.60%] heterosexual) | Children: (4–17); parents: NA | Cross-sectional | Yes | Children’s psychological adjustment; parents’ mental health* | Children of bisexual parents had higher SDQ scores than children of heterosexual parents. Children of lesbian and gay parents did not differ from children of heterosexual parents in emotional and mental health difficulties. Bisexual parents demonstrated substantially higher levels of distress than either lesbian or gay or heterosexual parents. No significant differences were found between gay or lesbian parents and heterosexual parents in distress levels. |
| Golombok et al35 | UK | 130 families (81 ([62.3%] LG vs 49 ([37.7%] heterosexual) | Children: 6 (4-8); parents: 42.22 (6.07) | Cross-sectional | Yes | Children’s psychological adjustment; children’s gender role behaviour; parents’ mental health; parenting stress; parent–child relationship*† | Children in gay and lesbian families were less externalising problems than children in heterosexual parent families, but not in internalising symptoms. No significant difference were found between same-sex and heterosexual families in terms of gender role behaviour. Gay fathers showed lower levels of both depression and parenting stress, but not in anxiety. Gay fathers showed lower levels of parenting stress. Gay fathers showed higher levels of parent–child relationship. |
| Bos and Sandfort29 | The Netherlands | 131 children (63 ([48.09%] L vs 68 ([51.91%] heterosexual) | Children: (8-12); parents: NA | Cross-sectional | Yes | Children’s psychological adjustment; children’s gender identity/sexual orientation‡ | No differences were found in psychosocial adjustment between children with same-sex parents and children with opposite-sex parents. Children in lesbian families felt less parental pressure to conform to gender stereotypes(1.46 (0.53) vs 1.69 (0.65)), were less likely to experience their own gender as superior(2.15 (0.49) vs 2.51 (0.64))and were more likely to be uncertain about future heterosexual romantic involvement(2.32 (0.71) vs 2.08 (0.76)). |
| Sullins39 | USA | 207 007 children (512 ([0.25%] same-sex vs 206,495 ([99.75%] heterosexual) | Children: (0–18); parents: NA | Cross-sectional | No | Children’s psychological adjustment* | Children of same-sex parents have more than twice as prevalent in emotional problems as children of opposite-sex parents (RR: 2.4, 95% CI: 1.7 to 3.0). |
| Regnerus20 | USA | 1155 children (236 ([20.43%] LG vs 919 ([79.57%] heterosexual) | Children: (18–39); parents: NA | Cross-sectional | No | Children’s psychological adjustment; children’s physical health; children’s gender identity /sexual orientation; children’s educational outcomes‡ | The children with same-sex parents worse on the CES-D (depression) index (2.19 vs 1.83) and apt to have thought recently about committing suicide (18% vs 5%), than children in heterosexual married families, but not in anxiety (2.67 vs 2.46). Children of lesbian parents reported worse physical health (3.38), but no significant differences between children of gay parents (3.58) and children of heterosexual married families (3.75). Children of lesbian mother families and gay father families were less likely to self-identify as entirely heterosexual than children of heterosexual married families (61% vs 71% vs 90%). Children of parents who have lesbian or gay romantic relationship worse on educational attainment than do respondents from different-sex married families (2.39 vs 2.64 vs 3.19). |
| Crouch et al38 | Australia | 5525 children (500 ([9.05%] LGBTQ or other vs 5,025 (90.95%] population normative data) | Children: (0–17); parents: NA | Cross-sectional | No | Children’s psychological adjustment; children’s physical health* | Children in same-sex families had higher scores in measures of general behaviour compared with population normative data (β: 2.93, 95% CI: 0.35 to 5.52), but not in SDQ scores, role-emotional/behavioural, mental health and self-esteem scores. Children in sexual minority families had higher scores on measures of general health compared with population normative data (β=5.60, 95% CI=2.69 to 8.52, p = < 0.001), but not in child health for physical functioning, role-physical, and bodily pain (β=0.78, 95% CI = −1.79 to 3.35; β = −0.89, 95% CI = −3.60 to 1.81; β = −1.70, 95% CI = −4.60 to 1.20). |
| Reczek et al37 | USA | 167 047 children (633 ([0.4%] same-sex vs 166,414 ([99.6%] heterosexual) | Children: (4-17); parents: NA | Cross-sectional | No | Children’s psychological adjustment; children’s physical health* | Same-sex cohabiting parents were significantly more likely than different-sex married parents to report higher levels of child emotional difficulties (5.3% vs 0.8%), less likely to respond ‘certainly true’ to the question about children being well-behaved (65.8% vs 79.1 %) and more likely to report good/fair/poor health (20.7% vs 13.8 %). No significant differences were found between children in same-sex married families and children in heterosexual married families on physical health outcomes (18.5% vs 13.8%). |
| Goldberg and Garcia34 | USA | 136 families (80 ([58.8%] LG vs 56 ([41.2%] heterosexual) | Children: (8-16); parents: NA | Cross-sectional | No | Children’s psychological adjustment* | CBCL total score were uncorrelated with family type (r = −0.14, p>0.1). |
| Bos et al40 | USA | 190 families (95 ([50%] L vs 95 ([50%] heterosexual) | Children: 10.67 (6-17/5.04); parents: 47.61 (6.45) | Cross-sectional | No | Children’s physical health; parenting stress; parent–child relationship; couple relationship satisfaction* | No differences were observed between same-sex parents and heterosexual parents in children’s health(4.63 (0.64) vs 4.58 (0.72)), parent–child relationship and couple relationship satisfaction. Same-sex parents reported significantly more stress than did different-sex parents. |
| Goldberg and Garcia42 | USA | 181 parents (104([57.5%] LG vs 77([42.5%] heterosexual) | Children: 6.06; parents: NA | Cohort | No | Children’s gender role behaviour* | No significant differences among lesbian, gay, or heterosexual parent families for girls’ gender role behaviour(37.86 (11.43), 39.38 (11.70), 34.80 (9.24), respectively); boys with lesbian mothers are less masculine than boys in other family types(56.47 (11.76), 62.42 (11.92), 64.49 (9.32) respectively). The degree of gender differentiation did not differ between heterosexual vs gay parent families. |
| Cenegy et al19 | USA | 27 810 families (1,005([3.6%] same-sex vs 26,805([96.4%] heterosexual) | Children: 3.68 (0–17); parents: 36.93 | Cross-sectional | No | Children’s physical health* | Poorer health among children in same-sex couple as well as heterosexual cohabiting couple and never married single-parent families (2.22, 2.31, 2.38, respectively) than heterosexual married families (1.97). |
| Carnaghi et al41 | Italy | 80 children (NA G vs NA heterosexual) | Children: 21.23 (1.97); parents: NA | Cross-sectional | No | Children’s gender role behaviour; children’s gender identity / sexual orientation‡ | The son of the GMM and GMF was perceived as being similarly masculine as the son of the heterosexual couple. The son of the GFF was perceived as less masculine than the sons of all the other parent descriptions.(heterosexual, GMM, GMF, GFF: 3.04 (0.14), 3.05 (0.13), 3.29 (0.13), 3.68 (0.15) respectively). The son raised by heterosexual couple would develop as heterosexual than the son raised by the GMM, the GMF, the GFF(5.29 (0.14) vs 4.81 (0.15), 4.78 (0.14), 4.64 (0.15)). |
| Boertien and Bernardi44 | USA | 1 952 490 children (7,792([0.4%] same-sex vs 1,944,698([99.6%] heterosexual) | Children: 12 (8-16); parents: 43.85 | Cross-sectional | No | Children’s educational outcomes*† | The initial differences between same-sex and different-sex parent families observed in 2008 disappeared within a few years (probability of being behind in school in 2008: 5.1% vs 3.0%; in 2010, 2.4% vs 2.9%). |
| Watkins46 | USA | 1 012 927 children (4,430([0.44%] same-sex vs 1,008,497([99.56%] heterosexual) | Children: NA; parents: NA | Cross-sectional | No | Children’s educational outcomes* | Among the children of married couples, Children who reside with gay, lesbian, heterosexual couples have a grade retention rate of 4.94%, 4.76% and 4.43% respectively. Among the children of unmarried couples, Children who reside with heterosexual, gay and lesbian parents have a grade retention rate of 6.09%, 6.13%, and 4.90%, respectively. |
| Kabátek and Perales45 | The Netherlands | 1 454 577 children (3,006([0.21] same-sex vs 1,451,571([99.79%] heterosexual) | Children: NA; parents: NA | Cross-sectional | No | Children’s educational outcomes*† | Children in same-sex parent families attain overall Cito test scores that are five percentile points higher than those of children in different-sex parent families, and they are more likely to enter an academic (ie, advanced) high school track (AME=21.61%, p<0.01), graduate from high school (AME=1.47%, p=0.01), and enrol in college (AME=11.20%, p=0.02). No differences were found in grade-retention rate (AME = −3.45%, p=0.32). |
| Allen43 | Canada | 1 401 466 parents (1,392([0.1%] LG vs 1,400,074(99.9%] heterosexual) | Children: (0–22); parents: NA | Cross-sectional | No | Children’s educational outcomes c | Children of gay and lesbian parents have a lower graduation rate compared with children of married opposite-sex families (60% vs 52% vs 72%). |
| Farr and Vázquez47 | USA | 209 parents (110([52.6%] LG vs 99([47.4%] heterosexual) | Children: 8.36 (1.66); parents: 47.56 (5.87) | Cohort | Yes | Parents’ mental health; parent–child relationship a,c | No significant differences were found in parents’ mental health and parent–child relationship between same-sex and heterosexual parent families. |
| Farr and Patterson50 | USA | 104 families (54([51.9%] LG vs 50([48.1%] heterosexual) | Children: 3.01 (1.08-6/1.31); parents: 41.69 (30-60/5.51) | Cross-sectional | Yes | Parent–child relationship a | Lesbian couples showed the most supportive and least undermining behaviour, whereas gay couples showed the least supportive behaviour, and heterosexual couples the most undermining behaviour. |
| Gelderen et al48 | UK, The Netherlands, France | 140 families (99([70.7%] LG vs 41([29.3%] heterosexual) | Children: (0.28–0.38); parents: 34.8 (22-59/5.07) | Cross-sectional | Yes | Parents’ mental health; parenting stress; couple relationship satisfaction a | No significant differences were found in anxiety and depression, parenting stress, relationship satisfaction between same-sex couples and different-sex couples after controlling for caregiver role. |
| Goldberg et al51 | USA | 42 parents (30([71.4%] LG vs 12([28.6%] heterosexual) | Children: 4.63 (0–16); parents: 37.55 (6.17) | qualitative | No | Couple relationship satisfaction a | There were few differences in intimate relationships by parent sexual orientation. Same-sex couples cited some additional stressors related to their sexual minority status. |
| Farr49 | USA | 184 parents (103([56%] LG vs 81([44%] heterosexual) | Children: 8.38 (1.62); parents: 47.50 (5.56) | Cohort | Yes | Parenting stress; family functioning a | No significant differences were found between family types in parent adjustment and family functioning. |
| Solomon et al53 | USA | 985 parents (573([41.8%] LG vs 412([58.2%] heterosexual) | Children:14.29 (11.76); parents: 43.38 (7.27) | Cross-sectional | No | Social support a | The three types of couples (lesbians in civil unions, lesbians not in civil unions, and heterosexual married women) did not differ on perceived social support from friends(15.55 (4.45) vs 15.80 (4.22) vs 14.86 (4.32)). Lesbians in civil unions and not in civil unions perceived significantly less social support from family than did heterosexual married women.(12.14 (6.60) vs 11.29 (6.58) vs 15.33 (5.15)). Gay men in both types of couples perceived significantly more social support from friends than did heterosexual married men(15.26 (4.76) vs 11.42 (5.35)). The three groups (gay men in civil unions, gay men not in civil unions, and heterosexual married men) did not differ significantly in perceived social support from family (11.90 (6.48) vs 11.07 (6.46) vs 12.80 (5.89)). |
| Goldberg et al52 | USA | 84 parents (60([71.4%] LG vs 24([28.6%] heterosexual) | Children: 4.63 (0–16); parents: 37.55 (6.17) | qualitative | No | Social support a | Lesbian and gay participants faced additional concerns regarding the security of their placement due to the possibility for discrimination. |
| Goldberg et al55 | USA | 45 parents (30([66.67%] LG vs 15([33.33%] heterosexual) | Children: 5.78 (1.51); parents: NA | qualitative | No | Other outcomes: parental school involvement a | Gay male couples (n=9) and heterosexual couples (n=8) more often described differential involvement, whereby one partner was more involved at school than the other; lesbian couples were slightly less likely to describe a pattern of differential involvement (n=5). |
| Goldberg and Smith54 | USA | 105 parents (65([61.90%] LG vs 40([38.10%] heterosexual) | Children: 3.47 (0.99); parents: NA | Cross-sectional | No | Other outcomes: preschool selection considerations and experiences of school mistreatment a | Same-sex parents were more likely to consider racial diversity than heterosexual parents when selecting children’s preschool (47.5% vs 28%). In reporting on their experiences with schools, heterosexual parents were more likely to perceive mistreatment due to their adoptive status than sexual-minority parents. |
Data are mean, range, mean (SD), or mean (range), unless otherwise stated. Children refer to parents’ offspring of all ages throughout the article. In sample size column, LGBTQ refer to lesbian, gay, bisexual, transgender, and queer parents, respectively. In outcome measures column, *parent-reported data, †teacher-reported data and ‡child-reported data. In main finding column, the data of each study included in meta-analysis were listed in online supplemental appendix tables S4A–S4F.
AME, estimates from logistic regression models are expressed as average marginal effects (AMEs); GFF, gay male parents both described as feminine; GMF, gay male parents, one described as masculine and one as feminine; GMM, gay male parents both described as masculine; NA, Not Applicable.