Limits...
Adoption Does Not Increase the Risk of Mortality among Taiwanese Girls in a Longitudinal Analysis.

Mattison SM, Brown MJ, Floyd B, Feldman MW - PLoS ONE (2015)

Bottom Line: Adopted children often experience health and well-being disadvantages compared to biological children remaining in their natal households.Using Cox proportional hazards models in which adoption is included as a time-dependent covariate, we show that adoption of girls does not increase the risk of mortality, as previously suggested; in fact, it is either protective or neutral with respect to mortality.These results suggest that socio-structural variables may produce positive outcomes for adopted children, even compared to biological children who remain in the care of their parents.

View Article: PubMed Central - PubMed

Affiliation: Department of Biology and Women's, Gender, and Sexuality Studies Program, Boston University, 5 Cummington Mall, Boston, Massachusetts, 02215, United States of America; Department of Anthropology, University of Auckland, Private Bag 92019, Auckland Centre, 1142, New Zealand.

ABSTRACT
Adopted children often experience health and well-being disadvantages compared to biological children remaining in their natal households. The degree of genetic relatedness is thought to mediate the level of parental investment in children, leading to poorer outcomes of biologically unrelated children. We explore whether mortality is related to adoption in a historical Taiwanese population where adoption rarely occurred among kin. Using Cox proportional hazards models in which adoption is included as a time-dependent covariate, we show that adoption of girls does not increase the risk of mortality, as previously suggested; in fact, it is either protective or neutral with respect to mortality. These results suggest that socio-structural variables may produce positive outcomes for adopted children, even compared to biological children who remain in the care of their parents.

Show MeSH
Survival plots, by gender and adoption status.Adoption status is significantly associated with predicted survivorship, but only for adopted girls. Confidence intervals have been eliminated from the figure as they obscure the main results, but do not overlap for adopted girls, only.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4414473&req=5

pone.0122867.g003: Survival plots, by gender and adoption status.Adoption status is significantly associated with predicted survivorship, but only for adopted girls. Confidence intervals have been eliminated from the figure as they obscure the main results, but do not overlap for adopted girls, only.

Mentions: Cox proportional hazards models estimating the effects of age, gender, and other covariates on the hazard of mortality in our dataset confirmed relationships apparent in the above descriptions. These analyses revealed that adoption was significantly and inversely associated with mortality, but only for girls (Table 2 and S4 Table-all ages). This effect was dramatic: Fig 3 shows that roughly 78% of adopted girls could be expected to survive 40 years, compared to approximately 72% of non-adopted boys and girls. Furthermore, adoption became more protective over the duration of the study (i.e., with calendar time), even after its prevalence lessened. Our results provide partial support for the suggestion that it was ADIL, not AD, whose survivorship was highest, since the effect of adoption on mortality was strongest in areas where ADIL (minor marriage) was highest: although the interaction term was not significant (Table 2, cf. S4 Table), regions in which minor marriage prevalence was high experienced lower mortality than those where the prevalence of minor marriage was lower [17,53]. However, the effect was present for both males and females when models were run separately for each gender (S6 Table); the specific means by which male adoptees may have benefitted from a high prevalence of minor marriage is unclear. The north-south gradient of increasing mortality in Taiwan [41] does not fully overlap with prevalence of ADIL and cannot account for the separate effect of adoption on mortality. Other results of these analyses are in the anticipated directions [41]: age was the strongest contributor to the hazard of mortality; high socio-economic status (as indicated by the head of household’s occupation) relative to employment in agriculture generally was associated with lower mortality; and higher parity increased the hazard of mortality.


Adoption Does Not Increase the Risk of Mortality among Taiwanese Girls in a Longitudinal Analysis.

Mattison SM, Brown MJ, Floyd B, Feldman MW - PLoS ONE (2015)

Survival plots, by gender and adoption status.Adoption status is significantly associated with predicted survivorship, but only for adopted girls. Confidence intervals have been eliminated from the figure as they obscure the main results, but do not overlap for adopted girls, only.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4414473&req=5

pone.0122867.g003: Survival plots, by gender and adoption status.Adoption status is significantly associated with predicted survivorship, but only for adopted girls. Confidence intervals have been eliminated from the figure as they obscure the main results, but do not overlap for adopted girls, only.
Mentions: Cox proportional hazards models estimating the effects of age, gender, and other covariates on the hazard of mortality in our dataset confirmed relationships apparent in the above descriptions. These analyses revealed that adoption was significantly and inversely associated with mortality, but only for girls (Table 2 and S4 Table-all ages). This effect was dramatic: Fig 3 shows that roughly 78% of adopted girls could be expected to survive 40 years, compared to approximately 72% of non-adopted boys and girls. Furthermore, adoption became more protective over the duration of the study (i.e., with calendar time), even after its prevalence lessened. Our results provide partial support for the suggestion that it was ADIL, not AD, whose survivorship was highest, since the effect of adoption on mortality was strongest in areas where ADIL (minor marriage) was highest: although the interaction term was not significant (Table 2, cf. S4 Table), regions in which minor marriage prevalence was high experienced lower mortality than those where the prevalence of minor marriage was lower [17,53]. However, the effect was present for both males and females when models were run separately for each gender (S6 Table); the specific means by which male adoptees may have benefitted from a high prevalence of minor marriage is unclear. The north-south gradient of increasing mortality in Taiwan [41] does not fully overlap with prevalence of ADIL and cannot account for the separate effect of adoption on mortality. Other results of these analyses are in the anticipated directions [41]: age was the strongest contributor to the hazard of mortality; high socio-economic status (as indicated by the head of household’s occupation) relative to employment in agriculture generally was associated with lower mortality; and higher parity increased the hazard of mortality.

Bottom Line: Adopted children often experience health and well-being disadvantages compared to biological children remaining in their natal households.Using Cox proportional hazards models in which adoption is included as a time-dependent covariate, we show that adoption of girls does not increase the risk of mortality, as previously suggested; in fact, it is either protective or neutral with respect to mortality.These results suggest that socio-structural variables may produce positive outcomes for adopted children, even compared to biological children who remain in the care of their parents.

View Article: PubMed Central - PubMed

Affiliation: Department of Biology and Women's, Gender, and Sexuality Studies Program, Boston University, 5 Cummington Mall, Boston, Massachusetts, 02215, United States of America; Department of Anthropology, University of Auckland, Private Bag 92019, Auckland Centre, 1142, New Zealand.

ABSTRACT
Adopted children often experience health and well-being disadvantages compared to biological children remaining in their natal households. The degree of genetic relatedness is thought to mediate the level of parental investment in children, leading to poorer outcomes of biologically unrelated children. We explore whether mortality is related to adoption in a historical Taiwanese population where adoption rarely occurred among kin. Using Cox proportional hazards models in which adoption is included as a time-dependent covariate, we show that adoption of girls does not increase the risk of mortality, as previously suggested; in fact, it is either protective or neutral with respect to mortality. These results suggest that socio-structural variables may produce positive outcomes for adopted children, even compared to biological children who remain in the care of their parents.

Show MeSH