Key Points
Question
Are children of consanguineous parents at increased risk of common mood disorders or psychoses?
Findings
In this population-wide cohort study of 363 960 participants, being a child of consanguineous parents was associated with having an increased likelihood of psychotropic medication use in adulthood. Children of first-cousin consanguineous parents are more than 3 times as likely to receive medications for common mood disorders and more than twice as likely to receive medications for psychoses compared with children of nonrelated parents.
Meaning
A child of first-cousin consanguineous parents is at increased risk of common mood disorders and psychoses.
Abstract
Importance
Approximately 1 in 10 children worldwide are born to consanguineous parents. The literature on consanguinity and mental health of progeny is scarce despite the fact that many of the factors associated with consanguineous unions are also associated with mental health.
Objective
To investigate if children of consanguineous parents are at increased risk of common mood disorders or psychoses.
Design, Setting, and Participants
This investigation was a retrospective population-wide cohort study of all individuals born in Northern Ireland between January 1, 1971, and December 31, 1986, derived from the Child Health System data set and linked to nationwide administrative data sources on prescription medication and death records. Data from the Child Health System data set identified all 447 452 births delivered to mothers residing in Northern Ireland between 1971 and 1986. The final data set comprised 363 960 individuals, alive and residing in Northern Ireland in 2014, with full data on all variables. The dates of analysis were June 1 to October 31, 2017.
Main Outcomes and Measures
Degree of parental consanguinity was assessed from questions asked of the parents during routine health visitor house calls within 2 weeks of the child’s birth. Potential mental ill health was estimated by receipt of psychotropic medication in 2010 to 2014. Ever or never use was used for the main analysis, with sensitivity analyses using a cutoff of at least 3 months’ prescriptions. Receipt of antidepressant or anxiolytic medications was used as a proxy for common mood disorders, whereas receipt of antipsychotic medications was used as a proxy indicator of psychoses.
Results
Of the 363 960 individuals (52.5% [191 102] male), 609 (0.2%) were born to consanguineous parents. After full adjustment for factors known to be associated with poor mental health, multilevel logistic regression models found that children of first-cousin consanguineous parents were more than 3 times as likely to be in receipt of antidepressant or anxiolytic medications (odds ratio, 3.01; 95% CI, 1.24-7.31) and more than twice as likely to be in receipt of antipsychotic medication (odds ratio, 2.13; 95% CI, 1.29-3.51) compared with children of nonrelated parents.
Conclusions and Relevance
A child of consanguineous parents is at increased risk of common mood disorders and psychoses.
This population-wide cohort study investigates if children of consanguineous parents are at increased risk of common mood disorders or psychoses.
Introduction
Across the world, approximately 1 in 10 children are the progeny of consanguineous parents despite concerns about the genetic safety of such a partnership.1 Consanguinity is defined as the union between 2 individuals related as second cousins or closer. The most commonly reported form of consanguineous partnership worldwide is between first cousins, who on average have coinherited one-eighth of their genes from one or more common ancestors. Therefore, first-cousin offspring will be homozygous at one-sixteenth of all loci (ie, they will receive identical gene copies from each parent at these sites in their genome).2,3 It is this shared genetic profile that is thought to lead to a higher prevalence of autosomal recessive disorders in children of consanguineous unions. The risk of abnormality or death in early childhood is approximately 5% in children of consanguineous couples compared with 2% to 2.5% for children of nonconsanguineous couples.4 Unsurprisingly, rates of miscarriage and stillbirth are higher among children of consanguineous parents.5,6 However, the results of some studies4,7 also suggest that consanguinity deleteriously affects late pregnancy and postpregnancy outcomes, including preeclampsia, prematurity, and low birth weight.4,7 A recent report from the United Kingdom stated that, in 1 London borough, 1 in 5 of all neonatal deaths were owing to their parents being related.8 Consanguinity has also been associated with increased risk of later-life effects such as cardiovascular disease, cancer, and Alzheimer disease.9
However, the validity of these associations and the magnitude of the risk have often been contested.10 Researchers in Australia found the risk of congenital defects in infants born to first-cousin marriages to be comparable to the risk to infants born to women older than 40 years.11 A narrative review12 on the effect of consanguinity on neonatal outcomes concluded that the findings were inconsistent, citing poor study design and inadequate adjustment for confounding factors as the reasons for the observed variability. In addition, the National Society of Genetic Counselors13 in North America concluded that risks quoted from studies based on non-Western populations may not be applicable to all consanguineous unions owing to underlying societal differences and ethnicity-related risk factors, suggesting that well-controlled studies evaluating the effect of consanguinity have not yet been conducted.
The literature on consanguinity and the mental health of progeny is scarce despite the fact that many of the factors associated with consanguineous unions are also associated with mental health outcomes.14,15,16 It is widely known that early-life factors such as parental deprivation and low birth weight are associated with poor mental health outcomes in adulthood.17,18 Furthermore, these factors are associated with consanguinity.19,20 Consanguineous pregnancies are also associated with younger maternal age, which is a risk factor for poor mental health in children.5,21 Children of consanguineous parents also face a certain degree of stigma, especially in communities where consanguinity is not the norm, and this stigma could negatively affect their mental well-being.13 Extant studies15,16,22,23,24 exploring the association between consanguinity and mental health have been limited by study cohort size, a lack of adequate controls, and inconsistent measurement of mental health. One recent study22 in Iran found no association between consanguinity and mental ill health in students aged 18 to 39 years as measured by the General Health Questionnaire 28; however, that study was based on a small sample of medical sciences students in 1 university and excluded anyone with a prediagnosed psychiatric disorder. There is a recognized need for further studies of the effect of consanguinity on late-onset disorders such as psychoses and common mood disorders that rigorously control for potential confounding variables like socioeconomic status, birth weight, maternal age, and rural dwelling.1
It is difficult to carry out a population-wide study on the effects of consanguinity in children owing to the lack of routine records on consanguineous marriage. First-cousin marriages are legal throughout the world with the exception of the United States, North Korea, and the People’s Republic of China.2 However, actual rates of consanguinity within populations are impossible to determine. It is estimated that consanguineous unions are increasing across Western Europe owing to migration from areas where consanguinity is commonplace.25,26
Data from church records are available on Roman Catholic consanguineous unions because special dispensation is required from the Catholic Church for such individuals to marry. Roman Catholics constitute the largest majority religion in Northern Ireland (NI), and recent data suggest that 1 in 625 (0.2%) of all Roman Catholic marriages in Ireland are consanguineous,26 with an estimated 0.1% to 0.2% of Roman Catholic marriages in Canada also being consanguineous.27 A random survey of 630 presentations to emergency departments in NI in 1955 found 0.3% of the population to be in consanguineous unions.28
This article presents the findings of a retrospective population-wide cohort study of data from the Child Health System (CHS) data set, which recorded information on all births in NI between January 1, 1971, and December 31, 1986, along with parental information, including degree of consanguinity. This cohort allows for the first population-wide data linkage study to date linking data from the CHS data set to primary care records, prescription medication data, and death records to investigate the association between consanguinity and the long-term mental health outcomes of progeny.
Methods
Study Population and Design
We used the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for reporting. Data from the CHS data set were used to form a historical cohort of all 447 452 births delivered to mothers residing in NI during a 15-year period (between 1971 and 1986).29,30 Details were collated on the child (including gestational age, birth weight, and delivery method) from obstetric records at the time of delivery and on the mother (including mother’s age, parity, and area of residence) and the father (including father’s age and degree of consanguinity to the mother) by health visitors in the home, typically within 1 to 2 weeks of the birth.30 Health visitors are public health practitioners that provide support to all families in NI as part of the free-at-the-point-of-service National Health Service.31 After the introduction of the unique individual Health and Care Number (HCN) in 1998 (which replaced the previously used Community Health Index [CHI] identifier), the CHS data set was updated, allowing for direct one-to-one linkage to other contemporary health care–related data sets. However, not all individuals were successfully assigned a new HCN owing to name changes, marriages, and duplication errors. All CHS data with an HCN were linked to current population-wide data on prescription medication from the Enhanced Prescribing Database (EPD) and death records to investigate the mental health profile of our cohort. The final study data set comprised 363 960 individuals born between 1971 and 1986, alive and residing in NI in 2014, with full data on all variables (447 452 minus 74 738 with missing HCN, 3328 deaths, and 5426 with missing data) (eFigure in the Supplement). The dates of analysis were June 1 to October 31, 2017.
The EPD contains information on all prescriptions dispensed in community pharmacies in NI from 2010 onward.32 Northern Ireland’s health system includes free prescription medication, and every individual is registered with a general practitioner (GP) at birth. For this study, prescribed medication was collated for the calendar years 2010 to 2014 inclusive.
Individual-level informed consent was not required because only nonidentifiable data were made available to the research team. Ethical approval was obtained from the Office for Research Ethics Committees Northern Ireland.
Child Characteristics
Child sex was identified from the CHS data set. Age was calculated as of the study midpoint (June 15, 2012) and grouped as 26 to 29, 30 to 33, 34 to 37, or 38 to 41 years. Birth weight and gestational age were used to calculate a small for gestational age (SGA) variable as per the global reference for fetal weight and birth weight percentiles.33 An infant was considered SGA if he or she weighed below the 10th percentile of the sex-specific, population-based birth weight reference curve for gestational age. Being SGA has been linked to increased risk of long-term health and social consequences such as neurocognitive impairment, hyperactivity, and lower educational attainment.34,35,36 Delivery method was categorized as natural, natural assisted, or cesarean delivery. Births were identified as singleton (n = 357 351) or multiple (n = 6609) to allow for sensitivity analyses limited to singleton births only. Ethnicity information was not available; however, less than 0.8% of the NI population at the time of the CHS were nonwhite.37
Parental Characteristics
Maternal and paternal ages were obtained from the CHS data set. Each sex contained a large age range, so only ages within 3 SDs of the mean were accepted, with all others deemed at high risk of error and placed in the “unknown” age group category. Parental age was defined as younger than 18 years, 18 to 35 years, and older than 35 years because parents younger than 18 years and older than 35 years have been identified as having a high risk of psychiatric morbidity in offspring.38,39,40 Maternal parity was also identified and categorized as firstborn, parity 1, parity 2, or parity 3 or more. The mother’s address at the time of the child’s birth was used to assign area-level deprivation.41 Areas are ranked from most affluent to most deprived based on the number of households in receipt of income-related state benefits and tax credits. Degree of consanguinity between the parents was based on response to questions from the health visitor and was identified as nonrelated parents, first-cousin pairing, second-cousin pairing, or not known.
Prescribed Medication
Receipt of psychotropic medication was used as a proxy indicator of psychopathology. Individuals were classified as being in receipt of antipsychotic medication if they received at least 1 prescription for antipsychotics (British National Formulary [BNF] category 4.3.6) and were classified as being in receipt of medications for common mood disorders if they received at least 1 prescription for antidepressant medication (BNF category 4.3.4) or anxiolytic medication (BNF category 4.3.1) over the 5-year study period (2010-2014). The BNF is the standard reference digest for medications in the United Kingdom.42 Ever or never use was used for the main analysis, and sensitivity analyses were carried out using a cutoff of at least 3 months’ prescriptions, yielding similar results (eTable 1 in the Supplement).
Data Linkage
The prescribing data were linked to the CHS data set using the unique individual HCN. Linkages were undertaken by the data custodians, and the resultant research data set containing only fully anonymized data was made available to the research team within a secure analysis environment.
Analytic Approach
Analysis was divided into 3 stages. First, descriptive analysis of the cohort aimed to investigate the demographic profile of children born to consanguineous partnerships. Second, multilevel multivariable regression models were constructed to assess the likelihood of medication use for common mood disorders given the degree of consanguinity between the parents, adjusting for factors known to be associated with mental ill health and multilevel adjustment for the natural clustering of individuals within GP practices. Receipt of antidepressant or anxiolytic medications was used as a proxy indicator of common mood disorders. This method has been validated in previous studies.43,44 Owing to small numbers in each of the consanguinity categories, measures of area deprivation and area rurality were added to the multilevel models separately to ensure convergence. Third, as per the method above, multilevel multivariable regression models were constructed to estimate the likelihood of psychotropic medication use given the degree of parental consanguinity. Receipt of psychotropic medication was used as a proxy measure of psychoses.32 Sensitivity analyses were carried out repeating each of the multilevel regression analyses limited to singleton births only (n = 357 351), yielding similar results.
Missing HCN
A total of 74 738 individuals (16.7%) were not included in the cohort because they were unable to be assigned an HCN when the unique identifier was updated from CHI to HCN. A CHI-to-HCN lookup was created matching individuals on name, address, and date of birth and allowing a present HCN to be assigned to the historical CHS data set. Individuals with incomplete data in these fields may not have been successfully assigned an HCN. The HCN indicator was used to link the CHS with the EPD data set. This proportion of the population was further explored to assess whether it varied significantly from the study cohort. Female sex (odds ratio [OR], 1.35; 95% CI, 1.33-1.37) was associated with missing HCN, likely owing to marital name changes or migration since assignment of the original CHI. Older age (OR, 3.09; 95% CI, 3.01-3.17 for 38-41 years compared with 26-29 years), SGA (OR, 1.88; 95% CI, 1.83-1.93), and first-cousin consanguineous parents (OR, 1.74; 95% CI, 1.35-2.25) were also associated with missing HCN, likely owing to the higher mortality risk in this group (eTable 2 in the Supplement).
Results
Of the 363 960 individuals born between 1971 and 1986 in our cohort (52.5% [191 102] male), 609 (0.2%) were born to consanguineous parents, including 349 to second-cousin consanguineous parents and 260 to first-cousin consanguineous parents. These results are listed in Table 1.
Table 1. Proportion of the Population With Consanguineous Parents by Level of Consanguinity and Demographic Characteristics.
| Variable | All, No. (%) (N = 363 960) |
% | P Valuea | |||
|---|---|---|---|---|---|---|
| Not Related (n = 344 183) |
First Cousins (n = 260) |
Second Cousins (n = 349) |
Not Known (n = 19 168) |
|||
| Sex | ||||||
| Male | 191 102 (52.5) | 52.5 | 53.9 | 51.3 | 51.7 | .82 |
| Female | 172 858 (47.5) | 47.5 | 46.1 | 48.7 | 48.3 | |
| Age, y | ||||||
| 26-29 | 97 399 (26.8) | 27.1 | 43.1 | 39.8 | 21.1 | <.01 |
| 30-33 | 95 663 (26.3) | 26.4 | 18.1 | 21.2 | 24.3 | |
| 34-37 | 87 065 (23.9) | 23.0 | 22.3 | 17.5 | 40.5 | |
| 38-41 | 83 833 (23.0) | 23.5 | 16.5 | 21.5 | 14.1 | |
| SGA | ||||||
| No | 342 412 (94.1) | 94.1 | 92.3 | 93.7 | 93.8 | .45 |
| Yes | 21 548 (5.9) | 5.9 | 7.7 | 6.3 | 6.2 | |
| Delivery methodb | ||||||
| Vaginal | 290 841 (79.9) | 80.0 | 82.3 | 78.2 | 79.0 | .46 |
| Vaginal assisted (ie, forceps, vacuum) and cesarean delivery | 73 119 (20.1) | 20.0 | 17.7 | 21.8 | 21.0 | |
| Parity | ||||||
| Firstborn | 96 685 (26.6) | 26.5 | 20.0 | 27.8 | 27.9 | <.01 |
| 1 | 105 750 (29.1) | 29.3 | 23.5 | 22.6 | 25.8 | |
| 2 | 64 794 (17.8) | 17.9 | 12.7 | 16.1 | 16.0 | |
| ≥3 | 70 968 (19.5) | 19.5 | 33.5 | 22.4 | 19.7 | |
| Unknown | 25 763 (7.1) | 6.9 | 10.4 | 11.2 | 10.6 | |
| Mother’s age, mean, y | 27.7 | 27.7 | 27.0 | 27.0 | 27.3 | .08 |
| Father’s age, mean, y | 30.2 | 30.1 | 37.4 | 35.2 | 31.6 | <.01 |
| Deprivation at birth | ||||||
| Not deprived | 200 238 (55.0) | 55.4 | 41.9 | 58.7 | 48.0 | <.01 |
| Deprived | 156 797 (43.1) | 42.7 | 53.1 | 38.4 | 49.8 | |
| Not known | 6925 (1.9) | 1.9 | 5.0 | 2.9 | 2.2 | |
| Urbanicity at birth | ||||||
| Urban | 144 647 (39.7) | 39.8 | 31.2 | 22.4 | 39.2 | <.01 |
| Rural | 212 369 (58.3) | 58.3 | 63.9 | 74.8 | 58.6 | |
| Not known | 6944 (1.9) | 1.9 | 5.0 | 2.9 | 2.2 | |
| Common mood medication use | ||||||
| No | 269 201 (74.0) | 74.0 | 64.2 | 68.8 | 73.2 | <.01 |
| Yes | 94 759 (26.0) | 26.0 | 35.8 | 31.2 | 26.8 | |
| Antipsychotic medication use | ||||||
| No | 354 156 (97.3) | 97.3 | 91.5 | 95.7 | 96.9 | <.01 |
| Yes | 9804 (2.7) | 2.7 | 8.5 | 4.3 | 3.1 | |
Abbreviation: SGA, small for gestational age.
P values represent χ2 test for difference between not related and related populations only (excluding the “Not Known” column).
Delivery method summarized as “Natural” or “Other” owing to small cell counts.
There was no significant difference in the sex distribution of offspring of consanguineous parents. However, a larger proportion of consanguineous offspring were younger, with 43.1% (112 of 260) of the first-cousin group aged 26 to 29 years compared with just 27.1% (93 105 of 344 183) of the nonrelated group. There was no significant difference in SGA or delivery method between consanguineous offspring vs nonconsanguineous offspring, but consanguineous offspring tended to come from larger families, with almost half (46.2% [120 of 260]) of children of first cousins being third born or greater (ie, parity ≥2). Father’s age was also older in first-cousin consanguineous unions (mean age, 37.4 years) compared with nonrelated parents (mean age, 30.1 years). A greater proportion of consanguineous offspring were from deprived and rural areas.
There was a clear stepwise increase in the proportion of consanguineous offspring in receipt of psychotropic medication with degree of consanguinity. More than one-third (35.8% [93 of 260]) of children of first-cousin consanguineous unions were in receipt of antidepressant or anxiolytic medications compared with just over one-quarter (26.0% [89 412 of 344 183]) of nonrelated offspring. Furthermore, 8.5% (22 of 260) of first-cousin consanguineous parent offspring were in receipt of antipsychotic medication compared with 4.3% (15 of 349) of second-cousin consanguineous parent offspring and 2.7% (9167 of 344 183) of nonrelated offspring.
In the multilevel regression models, female sex (OR, 1.79; 95% CI, 1.72-1.88), middle age (OR, 1.11; 95% CI, 1.04-1.19 for those aged 38-41 years compared with those aged 26-29 years), and residence in a deprived area at birth (OR, 1.10; 95% CI, 1.04-1.15 for deprived compared with nondeprived areas) were associated with increased likelihood of being in receipt of medications for common mood disorders, while residence in a rural area at birth was associated with decreased likelihood of medication use (OR, 0.91; 95% CI, 0.85-0.97) (Table 2). These values reflect well-established associations between sociodemographic factors and mental ill health and affirm the robustness of prescribed antidepressant or anxiolytic medications as a measure of common mood disorders. There was a clear stepwise increase in the ORs for antidepressant or anxiolytic medication use given the degree of consanguinity of parents. After full adjustment for factors known to be associated with poor mental health, children of first-cousin consanguineous parents were more than 3 times as likely to be in receipt of medications for common mood disorders compared with children of nonrelated parents (OR, 3.01; 95% CI, 1.24-7.31). The association between being a child of second-cousin consanguineous parents and receiving medications for common mood disorders was elevated but not statistically significant at the conventional 5% level (OR, 1.31; 95% CI, 0.63-2.71). Restricting analysis to singleton births did not affect these associations (OR, 3.01; 95% CI, 1.23-7.41 for first cousin and OR, 1.31; 95% CI, 0.63-2.71 for second cousin) (full sensitivity results are available in eTable 3 in the Supplement).
Table 2. Multilevel Regression Models to Investigate the Likelihood of Antidepressant or Anxiolytic Medication Use Given Parental Consanguinity, Adjusting for the Clustering of Individuals Within GP Practices.
| Variable | OR (95% CI) | ||
|---|---|---|---|
| Unadjusted | Model 1 | Model 2 | |
| Consanguineous parents | |||
| Not related | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| First cousins | 3.01 (1.24-7.31) | 2.99 (1.23-7.27) | 3.01 (1.24-7.31) |
| Second cousins | 1.32 (0.64-2.72) | 1.30 (0.63-2.70) | 1.31 (0.63-2.71) |
| Not known | 1.03 (0.94-1.14) | 1.00 (0.90-1.10) | 1.00 (0.90-1.10) |
| Sex | |||
| Male | NA | 1 [Reference] | 1 [Reference] |
| Female | NA | 1.79 (1.72-1.88) | 1.79 (1.71-1.87) |
| Age, y | |||
| 26-29 | NA | 1 [Reference] | 1 [Reference] |
| 30-33 | NA | 1.05 (0.99-1.12) | 1.05 (0.99-1.12) |
| 34-37 | NA | 1.10 (1.03-1.17) | 1.09 (1.02-1.17) |
| 38-41 | NA | 1.11 (1.04-1.19) | 1.11 (1.04-1.19) |
| SGA | |||
| No | NA | 1 [Reference] | 1 [Reference] |
| Yes | NA | 1.06 (0.97-1.16) | 1.06 (0.97-1.17) |
| Delivery method | |||
| Vaginal | NA | 1 [Reference] | 1 [Reference] |
| Vaginal assisted (ie, forceps, vacuum) | NA | 1.03 (0.95-1.10) | 1.02 (0.95-1.10) |
| Cesarean delivery | NA | 0.97 (0.89-1.06) | 0.97 (0.89-1.05) |
| Parity | |||
| Firstborn | NA | 1 [Reference] | 1 [Reference] |
| 1 | NA | 0.97 (0.91-1.04) | 0.97 (0.92-1.04) |
| 2 | NA | 0.93 (0.87-1.00) | 0.93 (0.87-1.00) |
| ≥3 | NA | 1.00 (0.93-1.07) | 1.01 (0.94-1.09) |
| Unknown | NA | 0.93 (0.84-1.03) | 0.93 (0.84-1.04) |
| Mother’s age, y | |||
| <18 | NA | 1.06 (0.85-1.33) | 1.04 (0.86-1.34) |
| 18-35 | NA | 1 [Reference] | 1 [Reference] |
| >35 | NA | 1.01 (0.92-1.10) | 1.00 (0.92-1.10) |
| Not known | NA | 1.05 (0.67-1.64) | 1.04 (0.67-1.63) |
| Father’s age, y | |||
| <18 | NA | 1.90 (0.86-4.21) | 1.91 (0.87-4.23) |
| 18-35 | NA | 1 [Reference] | 1 [Reference] |
| >35 | NA | 0.93 (0.86-1.00) | 0.93 (0.86-1.00) |
| Not known | NA | 1.11 (1.02-1.20) | 1.12 (1.03-1.22) |
| Deprivation at birth | |||
| Not deprived | NA | 1 [Reference] | NA |
| Deprived | NA | 1.10 (1.04-1.15) | NA |
| Not known | NA | 0.88 (0.73-1.04) | NA |
| Urbanicity at birth | |||
| Urban | NA | NA | 1 [Reference] |
| Rural | NA | NA | 0.91 (0.85-0.97) |
| Not known | NA | NA | 0.79 (0.65-0.94) |
| Variance | 0.352759 | 0.351842 | 0.352919 |
| P value | <.001 | <.001 | <.001 |
| Variance partition coefficient | 0.097 | 0.097 | 0.097 |
Abbreviations: GP, general practitioner; NA, not applicable; OR, odds ratio; SGA, small for gestational age.
Table 3 lists the results of the multilevel models investigating the association between antipsychotic medication use and consanguinity of parents. Being older (OR, 1.15; 95% CI, 1.08-1.23 for those aged 38-41 years compared with those aged 26-29 years), greater than fourth born (OR, 1.15; 95% CI, 1.07-1.23 for parity ≥3 compared with firstborn), and from a deprived area (OR, 1.34; 95% CI, 1.28-1.41 for deprived compared with nondeprived) were associated with increased likelihood of receiving antipsychotic medication, while being female (OR, 0.57; 95% CI, 0.55-0.60) and from a rural area (OR, 0.92; 95% CI, 0.85-0.99 for rural compared with urban) were associated with decreased likelihood of receiving antipsychotic medication. After full adjustment for factors known to be associated with poor mental health, children of first-cousin consanguineous parents were more than twice as likely to be in receipt of antipsychotic medication compared with children of nonrelated parents (OR, 2.13; 95% CI, 1.29-3.51). Restricting analysis to singleton births did not affect these associations (OR, 2.19; 95% CI, 1.32-3.61 for first cousin and OR, 1.37; 95% CI, 0.78-2.40 for second cousin) (full sensitivity results are available in eTable4 in the Supplement).
Table 3. Multilevel Regression Models to Investigate the Likelihood of Antipsychotic Medication Use Given Parental Consanguinity, Adjusting for the Clustering of Individuals Within GP Practices.
| Variable | OR (95% CI) | ||
|---|---|---|---|
| Unadjusted | Model 1 | Model 2 | |
| Consanguineous parents | |||
| Not related | 1 [Reference] | 1 [Reference] | 1 [Reference] |
| First cousins | 2.30 (1.40-3.77) | 2.09 (1.26-3.44) | 2.13 (1.29-3.51) |
| Second cousins | 1.39 (0.80-2.42) | 1.39 (0.79-2.43) | 1.37 (0.79-2.40) |
| Not known | 0.97 (0.89-1.06) | 0.92 (0.84-1.01) | 0.92 (0.84-1.01) |
| Sex | |||
| Male | NA | 1 [Reference] | 1 [Reference] |
| Female | NA | 0.57 (0.55-0.60) | 0.57 (0.55-0.60) |
| Age, y | |||
| 26-29 | NA | 1 [Reference] | 1 [Reference] |
| 30-33 | NA | 1.04 (0.98-1.11) | 1.04 (0.98-1.11) |
| 34-37 | NA | 1.10 (1.03-1.17) | 1.10 (1.03-1.17) |
| 38-41 | NA | 1.15 (1.08-1.23) | 1.15 (1.08-1.22) |
| SGA | |||
| No | NA | 1 [Reference] | 1 [Reference] |
| Yes | NA | 1.16 (1.07-1.26) | 1.18 (1.09-1.28) |
| Delivery method | |||
| Vaginal | NA | 1 [Reference] | 1 [Reference] |
| Vaginal assisted (ie, forceps, vacuum) | NA | 1.04 (0.97-1.11) | 1.03 (0.96-1.10) |
| Cesarean delivery | NA | 1.09 (1.01-1.18) | 1.08 (1.00-1.17) |
| Parity | |||
| Firstborn | NA | 1 [Reference] | 1 [Reference] |
| 1 | NA | 1.04 (0.98-1.11) | 1.05 (0.98-1.11) |
| 2 | NA | 1.08 (1.00-1.15) | 1.08 (1.01-1.16) |
| ≥3 | NA | 1.15 (1.07-1.23) | 1.18 (1.10-1.26) |
| Unknown | NA | 1.03 (0.93-1.13) | 1.03 (0.93-1.14) |
| Mother’s age, y | |||
| <18 | NA | 1.12 (0.93-1.35) | 1.15 (0.96-1.39) |
| 18-35 | NA | 1 [Reference] | 1 [Reference] |
| >35 | NA | 0.98 (0.90-1.06) | 0.97 (0.90-1.05) |
| Not known | NA | 0.90 (0.60-1.35) | 0.88 (0.59-1.33) |
| Father’s age, y | |||
| <18 | NA | 1.43 (0.84-2.42) | 1.45 (0.85-2.47) |
| 18-35 | NA | 1 [Reference] | 1 [Reference] |
| >35 | NA | 1.02 (0.95-1.10) | 1.01 (0.94-1.09) |
| Not known | NA | 1.32 (1.23-1.42) | 1.35 (1.26-1.45) |
| Deprivation at birth | |||
| Not deprived | NA | 1 [Reference] | NA |
| Deprived | NA | 1.34 (1.28-1.41) | NA |
| Not known | NA | 1.13 (0.94-1.35) | NA |
| Urbanicity at birth | |||
| Urban | NA | NA | 1 [Reference] |
| Rural | NA | NA | 0.92 (0.85-0.99) |
| Not known | NA | NA | 0.89 (0.74-1.07) |
| Variance | 0.354988 | 0.318330 | 0.348805 |
| P value | <.001 | <.001 | <.001 |
| Variance partition coefficient | 0.097 | 0.088 | 0.096 |
Abbreviations: GP, general practitioner; NA, not applicable; OR, odds ratio; SGA, small for gestational age.
Risk of psychotropic medication use was also elevated in children of second-cousin consanguineous parents but was not statistically significant at the conventional 5% level (OR, 1.37; 95% CI, 0.79-2.40). Likelihood ratio tests for interactions found no interaction between rurality and consanguinity (χ2 = 6.37, P = .38) or between deprivation and consanguinity (χ2 = 7.99, P = .63).
Discussion
This study shows that a child of first-cousin consanguineous parents is at increased risk of common mood disorders and psychoses. In the study population, 0.2% of children were born to consanguineous parents, which is consistent with previous estimates of population consanguinity in Ireland and among Roman Catholic populations.26,27,28 Female sex, middle age, and deprivation were associated with receipt of antidepressant or anxiolytic medications, validating this measure because these factors are known in the literature to be associated with risk of depression and anxiety disorders.45,46 Children of first-cousin consanguineous parents were more than 3 times as likely to be in receipt of medications for common mood disorders compared with children of nonrelated parents. In addition, children of first-cousin consanguineous parents were more than twice as likely to be in receipt of antipsychotic medication compared with children of nonrelated parents. Male sex, older age, birth weight (SGA), parity, and deprivation also were significantly associated with antipsychotic risk, validating this measure further because these factors are known to be associated with risk of psychoses.18,47,48
There are several theories as to why consanguinity may result in mental ill health in progeny. First, high heritability points to a major role for inherited genetic variants in the etiology of psychiatric disorders.49 In recent years, genome-wide association studies50,51 of schizophrenia, bipolar disorder, and major depression have provided strong support for a substantial polygenic contribution of a large number of small genetic effects. An alternative view is that most of the variance for certain complex diseases is owing to moderately highly penetrant rare variants.52 As a form of assortative mating, consanguinity increases polygenic loading and thus is likely associated with a higher risk of mental disorder in progeny.53 However, this is only true if each of the parents carries common susceptibility loci.
A second theory suggests that having consanguineous parents is associated with “social stigma,” especially in Western societies where consanguineous partnerships are considered taboo.13 Being a member of a minority population and having even perceived discrimination are known to be associated with poor mental health outcomes.54,55 However, it is not known how many of the children in our cohort were aware of the genetic relationship of their parents.
Third, the observed association may be owing to some unmeasured confounding associated with the likelihood of consanguinity and to decreased mental health. However, the study design allowed for a robust examination of the mental health risk associated with consanguineous parents: the data were population wide, capturing an entire cohort born over 15 years, and contained detailed neonatal information on the individual and detailed sociodemographic information on the parents. The prevalence of consanguineous parents recorded in this study is in keeping with other estimates,26,27,28 and the associations between mental health and a range of sociodemographic factors reflect those found in other studies worldwide.17,46,47,48 The analysis included regression modeling, adjusting for a range of confounders known to be associated with mental health, and multilevel modeling allowed for excellent adjustment of the potential unknown confounding associated with the natural clustering of individuals within GP practices. The results illustrate a clear increasing, stepwise association between level of consanguinity and mental ill health, suggesting a quasi–dose-response association, supporting a causal association between consanguineous parents and mental health of progeny.
Strengths and Limitations
Our study has significant strengths and limitations. Regarding its strengths, this study is the first population-wide study to date of consanguinity and mental health of progeny, and it uses an objective measure of mental ill health in the form of prescribed medication data.
Its caveats concern the information limitations of the data, including prescription data without accompanying diagnosis codes or indication for use. However, prescription medication receipt as an indicator of mental ill health has been used effectively in previous studies44,56,57,58 worldwide. Consanguinity was identified by parents’ response to a question asked by a health visitor in their home, but some individuals may not have identified themselves as consanguineous parents owing to fears of stigma, discrimination, and even legal prosecution.13 However, there is no legal impairment to consanguinity in NI, so fear of legal prosecution is unlikely to be a factor herein. There is no information on the mental health of the parents of our cohort. Parental mental health is known to be related to the mental health of the children; however, almost all consanguineous parents would have had to have poor mental health themselves to produce the associations observed in this study, and there is no evidence to suggest poorer mental health among consanguineous couples. Last, to experience the outcome of interest, participants must have been alive in 2010 to 2014. However, psychopathology is known to be associated with mortality risk, meaning there is mortality bias in our results. This factor likely excludes those with the most severe mental disorders, biasing the results toward the null, but does not affect the robustness of the observed associations.
Conclusions
Despite the recent debate around the physical genetic risk of consanguineous parents, more research is required on the psychological effects of consanguineous parents on progeny. The results of this study suggest a significant association of consanguinity with mental health independent of birth weight, mother’s parity, parental age, deprivation, and rurality. However, to effectively analyze the effect of consanguinity on physical and mental ill health, there is a need to implement accurate record keeping of marriage between cousins. This study demonstrates the ability of population-wide data linkage to explore hard-to-reach populations, and we call on other countries with similar large-scale administrative data sources to use their data to explore the effects of consanguinity on offspring. We suggest that these findings will be of value to health promotion and public health professionals and to those commissioning antenatal, pediatric, and clinical genetic services. Sensitive advice about the risks should be provided to communities that favor consanguineous unions to assist in reproductive decision making.
eFigure. Flowchart Illustrating the Generation of the Study Data Set
eTable 1. Multilevel Regression Models to Determine the Likelihood of ≥3 Months’ Antidepressant and/or Anxiolytic or ≥3 Months’ Antipsychotic Medication Prescriptions Given Parental Consanguinity, Adjusting for the Clustering of Individuals Within GP Practices
eTable 2. Regression Analysis to Determine Likelihood of Missing HCN Given Neonatal Factors and Consanguinity of Parents
eTable 3. Multilevel Regression Models to Determine the Likelihood of Antidepressant and/or Anxiolytic Medication Given Parental Consanguinity for Singleton Births Only, Adjusting for the Clustering of Individuals Within GP Practices
eTable 4. Multilevel Regression Models to Determine the Likelihood of Antipsychotic Medication Given Parental Consanguinity for Singleton Births Only, Adjusting for the Clustering of Individuals Within GP Practices.
References
- 1.Bittles AH, Black ML. Consanguinity, human evolution, and complex diseases. Proc Natl Acad Sci U S A. 2010;107(suppl 1):1779-1786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Bittles AH, Black ML. The impact of consanguinity on neonatal and infant health. Early Hum Dev. 2010;86(11):737-741. [DOI] [PubMed] [Google Scholar]
- 3.Shawky RM, Elsayed SM, Zaki ME, El-din SM, Kamal FM. Consanguinity and its relevance to clinical genetics. Egypt J Med Hum Genet. 2013;14(2):157-164. [Google Scholar]
- 4.Kapurubandara S, Melov S, Shalou E, Alahakoon I. Consanguinity and associated perinatal outcomes, including stillbirth. Aust N Z J Obstet Gynaecol. 2016;56(6):599-604. [DOI] [PubMed] [Google Scholar]
- 5.Maghsoudlou S, Cnattingius S, Aarabi M, et al. . Consanguineous marriage, prepregnancy maternal characteristics and stillbirth risk: a population-based case-control study. Acta Obstet Gynecol Scand. 2015;94(10):1095-1101. [DOI] [PubMed] [Google Scholar]
- 6.Kuntla S, Goli S, Sekher TV, Doshi R. Consanguineous marriages and their effects on pregnancy outcomes in India. Int J Sociol Soc Policy. 2013;33(7/8):437-452. [Google Scholar]
- 7.Anvar Z, Namavar-Jahromi B, Saadat M. Association between consanguineous marriages and risk of pre-eclampsia. Arch Gynecol Obstet. 2011;283(suppl 1):5-7. [DOI] [PubMed] [Google Scholar]
- 8.Bulman M. London borough finds one in five child deaths caused by parents being related. The Independent Online May 17, 2017. http://www.independent.co.uk/news/uk/home-news/london-borough-child-deaths-redbridge-parents-related-cousins-pakistani-families-council-report-a7741146.html. Accessed June 17, 2017.
- 9.Fareed M, Afzal M. Genetics of consanguinity and inbreeding in health and disease. Ann Hum Biol. 2017;44(2):99-107. [DOI] [PubMed] [Google Scholar]
- 10.Bhopal RS, Petherick ES, Wright J, Small N. Potential social, economic and general health benefits of consanguineous marriage: results from the Born in Bradford cohort study. Eur J Public Health. 2014;24(5):862-869. [DOI] [PubMed] [Google Scholar]
- 11.Paul DB, Spencer HG. “It’s ok, we’re not cousins by blood”: the cousin marriage controversy in historical perspective. PLoS Biol. 2008;6(12):2627-2630. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Abbas HA, Yunis K. The effect of consanguinity on neonatal outcomes and health. Hum Hered. 2014;77(1-4):87-92. [DOI] [PubMed] [Google Scholar]
- 13.Bennett RL, Motulsky AG, Bittles A, et al. . Genetic counseling and screening of consanguineous couples and their offspring: recommendations of the National Society of Genetic Counselors. J Genet Couns. 2002;11(2):97-119. [DOI] [PubMed] [Google Scholar]
- 14.Hamamy H. Consanguineous marriages: preconception consultation in primary health care settings. J Community Genet. 2012;3(3):185-192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ahmed AH. Consanguinity and schizophrenia in Sudan. Br J Psychiatry. 1979;134:635-636. [DOI] [PubMed] [Google Scholar]
- 16.Chaleby K, Tuma TA. Cousin marriages and schizophrenia in Saudi Arabia. Br J Psychiatry. 1987;150:547-549. [DOI] [PubMed] [Google Scholar]
- 17.Reiss F. Socioeconomic inequalities and mental health problems in children and adolescents: a systematic review. Soc Sci Med. 2013;90:24-31. [DOI] [PubMed] [Google Scholar]
- 18.Van Lieshout RJ, Boyle MH, Saigal S, Morrison K, Schmidt LA. Mental health of extremely low birth weight survivors in their 30s. Pediatrics. 2015;135(3):452-459. [DOI] [PubMed] [Google Scholar]
- 19.Mumtaz G, Tamim H, Kanaan M, et al. ; National Collaborative Perinatal Neonatal Network . Effect of consanguinity on birth weight for gestational age in a developing country. Am J Epidemiol. 2007;165(7):742-752. [DOI] [PubMed] [Google Scholar]
- 20.Mumtaz G, Nassar AH, Mahfoud Z, et al. . Consanguinity: a risk factor for preterm birth at less than 33 weeks’ gestation. Am J Epidemiol. 2010;172(12):1424-1430. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Lopez-Castroman J. Young maternal age and old paternal age induce similar risk of mental disorders in offspring. Evid Based Ment Health. 2014;17(4):100. [DOI] [PubMed] [Google Scholar]
- 22.Hosseinpour M, Deris F, Solati-Dehkordi K, Heidari-Soreshjani S, Karimi N, Teimori H. The effect of consanguineous marriage on mental health among the students of the Shahrekord University of Medical Sciences. J Clin Diagn Res. 2016;10(11):GC01-GC04. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bener A, Dafeeah EE, Samson N. Does consanguinity increase the risk of schizophrenia? study based on primary health care centre visits. Ment Health Fam Med. 2012;9(4):241-248. [PMC free article] [PubMed] [Google Scholar]
- 24.Bener A, Abou-Saleh MT, Mohammad RM, Dafeeah EE, Ventriglio A, Bhugra D. Does consanguinity increase the risk of mental illnesses? a population-based study. Int J Cult Ment Health. 2016;9(2):172-181. [Google Scholar]
- 25.Ajaz M, Ali N, Randhawa G. UK Pakistani views on the adverse health risks associated with consanguineous marriages. J Community Genet. 2015;6(4):331-342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Barrett P. A review of consanguinity in Ireland: estimation of frequency and approaches to mitigate risks. Ir J Med Sci. 2016;185(1):17-28. [DOI] [PubMed] [Google Scholar]
- 27.De Braekeleer M, Ross M. Inbreeding in Saguenay-Lac-St-Jean (Quebec, Canada): a study of Catholic Church dispensations 1842-1971. Hum Hered. 1991;41(6):379-384. [DOI] [PubMed] [Google Scholar]
- 28.Kilpatrick SJ, Mathers JD, Stevenson AC. The importance of population fertility and consanguinity data being available in medico-social studies: some data on consanguineous marriages in Northern Ireland. Ulster Med J. 1955;24(2):113-122. [PMC free article] [PubMed] [Google Scholar]
- 29.O’Rorke MA, Black C, Murray LJ, Cardwell CR, Gavin AT, Cantwell MM. Do perinatal and early life exposures influence the risk of malignant melanoma? a Northern Ireland birth cohort analysis. Eur J Cancer. 2013;49(5):1109-1116. [DOI] [PubMed] [Google Scholar]
- 30.Cantwell MM, Forman MR, Middleton RJ, Murray LJ. Association of early life factors and brain tumour risk in a cohort study. Br J Cancer. 2008;99(5):796-799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Cowley S, Caan W, Dowling S, Weir H. What do health visitors do? a national survey of activities and service organisation. Public Health. 2007;121(11):869-879. [DOI] [PubMed] [Google Scholar]
- 32.Maguire A, Hughes C, Cardwell C, O’Reilly D. Psychotropic medications and the transition into care: a national data linkage study. J Am Geriatr Soc. 2013;61(2):215-221. [DOI] [PubMed] [Google Scholar]
- 33.Mikolajczyk RT, Zhang J, Betran AP, et al. . A global reference for fetal-weight and birthweight percentiles. Lancet. 2011;377(9780):1855-1861. [DOI] [PubMed] [Google Scholar]
- 34.Strauss RS. Adult functional outcome of those born small for gestational age: twenty-six-year follow-up of the 1970 British Birth Cohort. JAMA. 2000;283(5):625-632. [DOI] [PubMed] [Google Scholar]
- 35.Bickle Graz M, Tolsa JF, Fischer Fumeaux CJ. Being small for gestational age: does it matter for the neurodevelopment of premature infants? a cohort study. PLoS One. 2015;10(5):e0125769. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.World Health Organization Born Too Soon: The Global Action Report on Preterm Birth Geneva, Switzerland: World Health Organization. http://www.who.int/pmnch/media/news/2012/201204_borntoosoon-report.pdf. Published 2012. Accessed June 29, 2017.
- 37.Northern Ireland Statistics and Research Agency 2001. Census key statistics. https://www.nisra.gov.uk/sites/nisra.gov.uk/files/publications/2001-census-results-key-statistics-report-tables.pdf. Published 2001. Accessed August 5, 2017.
- 38.Chang Z, Lichtenstein P, D’Onofrio BM, et al. . Maternal age at childbirth and risk for ADHD in offspring: a population-based cohort study. Int J Epidemiol. 2014;43(6):1815-1824. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.D’Onofrio BM, Rickert ME, Frans E, et al. . Paternal age at childbearing and offspring psychiatric and academic morbidity. JAMA Psychiatry. 2014;71(4):432-438. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Malaspina D, Gilman C, Kranz TM. Paternal age and mental health of offspring. Fertil Steril. 2015;103(6):1392-1396. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Northern Ireland Statistics and Research Agency Northern Ireland multiple deprivation measure 2010 (NIMDM2010). https://www.nisra.gov.uk/statistics/deprivation/northern-ireland-multiple-deprivation-measure-2010-nimdm2010. Published May 26, 2010. Accessed August 5, 2017.
- 42.BMJ Publishing Group Ltd and Royal Pharmaceutical Society British National Formulary (BNF) 74. London, United Kingdom: Joint Formulary Committee 74; 2017. [Google Scholar]
- 43.Maguire A, French D, O’Reilly D. Residential segregation, dividing walls and mental health: a population-based record linkage study. J Epidemiol Community Health. 2016;70(9):845-854. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Gardarsdottir H, Egberts AC, van Dijk L, Sturkenboom MC, Heerdink ER. An algorithm to identify antidepressant users with a diagnosis of depression from prescription data. Pharmacoepidemiol Drug Saf. 2009;18(1):7-15. [DOI] [PubMed] [Google Scholar]
- 45.Piccinelli M, Wilkinson G. Gender differences in depression: critical review. Br J Psychiatry. 2000;177:486-492. [DOI] [PubMed] [Google Scholar]
- 46.Mair C, Diez Roux AV, Galea S. Are neighbourhood characteristics associated with depressive symptoms? a review of evidence. J Epidemiol Community Health. 2008;62(11):940-946, 8 p following 946. [DOI] [PubMed] [Google Scholar]
- 47.Ring N, Tantam D, Montague L, Newby D, Black D, Morris J. Gender differences in the incidence of definite schizophrenia and atypical psychosis: focus on negative symptoms of schizophrenia. Acta Psychiatr Scand. 1991;84(6):489-496. [DOI] [PubMed] [Google Scholar]
- 48.Croudace TJ, Kayne R, Jones PB, Harrison GL. Non-linear relationship between an index of social deprivation, psychiatric admission prevalence and the incidence of psychosis. Psychol Med. 2000;30(1):177-185. [DOI] [PubMed] [Google Scholar]
- 49.Ripke S, Neale B, Corvin A; Schizophrenia Working Group of the Psychiatric Genomics Consortium . Biological insights from 108 schizophrenia-associated genetic loci. Nature. 2014;511(7510):421-427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Bigdeli TB, Ripke S, Bacanu SA, et al. ; Schizophrenia Working Group of the Psychiatric Genomics Consortium . Genome-wide association study reveals greater polygenic loading for schizophrenia in cases with a family history of illness. Am J Med Genet B Neuropsychiatr Genet. 2016;171B(2):276-289. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Power RA, Tansey KE, Buttenschøn HN, et al. ; CONVERGE Consortium, CARDIOGRAM Consortium, GERAD1 Consortium . Genome-wide association for major depression through age at onset stratification: Major Depressive Disorder Working Group of the Psychiatric Genomics Consortium. Biol Psychiatry. 2017;81(4):325-335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Gibson G. Rare and common variants: twenty arguments. Nat Rev Genet. 2012;13(2):135-145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Rudan I, Campbell H, Carothers AD, Hastie ND, Wright AF. Contribution of consanguinuity to polygenic and multifactorial diseases. Nat Genet. 2006;38(11):1224-1225. [DOI] [PubMed] [Google Scholar]
- 54.McGuire TG, Miranda J. New evidence regarding racial and ethnic disparities in mental health: policy implications. Health Aff (Millwood). 2008;27(2):393-403. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Kessler RC, Mickelson KD, Williams DR. The prevalence, distribution, and mental health correlates of perceived discrimination in the United States. J Health Soc Behav. 1999;40(3):208-230. [PubMed] [Google Scholar]
- 56.Martikainen P, Valkonen T. Mortality after death of spouse in relation to duration of bereavement in Finland. J Epidemiol Community Health. 1996;50(3):264-268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 57.Crump C, Sundquist K, Sundquist J, Winkleby MA. Neighborhood deprivation and psychiatric medication prescription: a Swedish national multilevel study. Ann Epidemiol. 2011;21(4):231-237. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 58.Sundquist K, Frank G, Sundquist J. Urbanisation and incidence of psychosis and depression: follow-up study of 4.4 million women and men in Sweden. Br J Psychiatry. 2004;184:293-298. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eFigure. Flowchart Illustrating the Generation of the Study Data Set
eTable 1. Multilevel Regression Models to Determine the Likelihood of ≥3 Months’ Antidepressant and/or Anxiolytic or ≥3 Months’ Antipsychotic Medication Prescriptions Given Parental Consanguinity, Adjusting for the Clustering of Individuals Within GP Practices
eTable 2. Regression Analysis to Determine Likelihood of Missing HCN Given Neonatal Factors and Consanguinity of Parents
eTable 3. Multilevel Regression Models to Determine the Likelihood of Antidepressant and/or Anxiolytic Medication Given Parental Consanguinity for Singleton Births Only, Adjusting for the Clustering of Individuals Within GP Practices
eTable 4. Multilevel Regression Models to Determine the Likelihood of Antipsychotic Medication Given Parental Consanguinity for Singleton Births Only, Adjusting for the Clustering of Individuals Within GP Practices.