Risk of psychiatric disorders among the surviving twins after a co-twin loss
- West China Biomedical Big Data Center, West China Hospital, Sichuan University, China
- Center of Public Health Sciences, Faculty of Medicine, University of Iceland, Iceland
- Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Sweden
- School of Medical Sciences, Örebro University, Sweden
- Institute of Environmental Medicine, Karolinska Institutet, Sweden
- Astrid Lindgren Children’s Hospital, Karolinska University Hospital, Sweden
- Department of Psychology, University of Southern California, United States
- Department of Epidemiology, Harvard TH Chan School of Public Health, United States
Figures
Figure 1
Study design.
Figure 2
The association between loss of a co-twin and subsequent risk of psychiatric disorder by age at the index date, analyses of matched twin cohort.
* Restricted cubic splines were applied on age at index date, with five knots placed at 5, 27.5, 50, 72.5, and 95 quantiles of the distribution of outcome events. Then, age-varying HRs were …
-
Figure 2—source data 1
Summary data for Figure 2.
- https://cdn.elifesciences.org/articles/56860/elife-56860-fig2-data1-v1.xls
Figure 3
Hazard ratios (HRs) with 95% confidence intervals (CIs) for subtypes of psychiatric disorders among twins exposed to loss of a co-twin compared to matched unexposed twins (analyses of matched twin cohort), for the entire follow-up period and for the first year after the index date.
* Cox regression models were stratified by matching identifiers (birth year and sex), and adjusted for education level, family income, marital status, history of severe somatic diseases, and family …
Tables
Table 1
Characteristics of the study cohorts.
| Matched twin cohort: twins exposed vs. unexposed to loss of a co-twin | Twin-sibling cohort: twins exposed to loss of a co-twin vs. their full siblings | |||
|---|---|---|---|---|
| Exposed twins | Matched unexposed twins | Exposed Twins’ | Exposed full siblings | |
| Number of individuals | 4528 | 22640 | 2761 | 4939 |
| Age at index date, median (IQR), year | 59(47-67) | 59(47-67) | 57(45-65) | 57(46-65) |
| Follow-up time, median (IQR), year | 7.4 (3.1–13.7) | 7.7 (3.2–14.2) | 8.0 (3.3–14.5) | 8.9 (3.8–15.4) |
| % of male | 50.8% | 50.8% | 51.7% | 49.7% |
| Education level, n (%) | ||||
| <9 years | 1171 (25.9) | 5041 (22.3) | 700 (25.4) | 1187 (24.0) |
| 9–12 years | 2439 (53.9) | 12151 (53.7) | 1520 (55.1) | 2775 (56.2) |
| >12 years | 844 (18.6) | 5185 (22.9) | 505 (18.3) | 939 (19.0) |
| Unknown | 74 (1.63) | 263 (1.16) | 36 (1.30) | 38 (0.77) |
| Yearly family income level, n (%) | ||||
| Lowest 20% | 504 (11.1) | 2275 (10.1) | 313 (11.3) | 561 (11.4) |
| Middle | 2487 (54.9) | 12092 (53.4) | 1499 (54.3) | 2653 (53.7) |
| Top 20% | 1050 (23.2) | 5850 (25.8) | 626 (22.7) | 1142 (23.1) |
| Unknown | 487 (10.8) | 2423 (10.7) | 323 (11.7) | 583 (11.8) |
| Marital status, n (%) | ||||
| Single | 1270 (28.1) | 6073 (26.8) | 808 (29.3) | 1213 (24.6) |
| Married/cohabiting | 2624 (58.0) | 13739 (60.7) | 1564 (56.7) | 3056 (61.9) |
| Divorced/widowed | 634 (14.0) | 2828 (12.5) | 389 (14.1) | 670 (13.6) |
| History of severe somatic diseases*, n (%) | ||||
| Yes | 893 (19.7) | 2664 (11.8) | 514 (18.6) | 924 (18.7) |
| No | 3635 (80.3) | 19976 (88.2) | 2247 (81.4) | 4015 (81.3) |
| Family history of psychiatric disorders/suicide, n (%) | ||||
| Yes | 2257 (49.9) | 8440 (37.3) | 1378 (49.9) | 2505 (50.7) |
| No | 2271 (50.2) | 14200 (62.7) | 1383 (50.1) | 2434 (49.3) |
| Cause of the co-twin’s death, n (%) | ||||
| Unnatural death | ||||
| Yes | 1020 (22.5) | - | 666 (24.1) | 1213 (24.6) |
| No | 3508 (77.5) | - | 2095 (75.9) | 3726 (75.4) |
| Zygosity of twins, n (%) | ||||
| Monozygotic twins | 746 (16.5) | 2367 (10.5) | 423 (15.4) | - |
| Dizygotic twins | 3016 (66.6) | 15722 (69.4) | 1851 (67.0) | - |
| Unknown | 766 (16.9) | 4551 (20.1) | 487 (17.6) | - |
-
* Involved somatic diseases included myocardial infarction, congestive heart failure, cerebrovascular disease, chronic pulmonary disease, connective tissue disease, diabetes, renal diseases, liver diseases, ulcer diseases, and HIV infection/AIDS.
Table 2
Hazard ratios (HRs) with 95% confidence intervals (CIs) for any psychiatric disorder among the surviving twins after co-twin loss (overall and by follow-up time), compared to matched unexposed twins or their full siblings, derived from different Cox models.
| Model information | Matched twin cohort: twins exposed vs. unexposed to loss of a co-twin | Twin-sibling cohort: twins exposed to loss of a co-twin vs. their full siblings | ||
|---|---|---|---|---|
| Number of cases (Crude incidence rate, per 1000 person years), exposed/unexposed twins | Hr (95% CI)* | Number of cases (Crude incidence rate, per 1000 person years), exposed twins/exposed siblings | Hr (95% CI)* | |
| Model 1 Controlled/adjusted for sex, birth year | 526 (12.29)/1521 (6.90) | 1.80 (1.63–2.00) | 328 (11.92)/415 (7.90) | 1.54 (1.31–1.81) |
| Model 2 above + socioeconomic status (education level, marital status, family income) | 1.79 (1.62–1.99) | 1.55 (1.32–1.83) | ||
| Model 3 above + history of severe somatic disease | 1.71 (1.55–1.90) | 1.55 (1.31–1.82) | ||
| Model 4 above + family history of psychiatric disorder | 1.65 (1.48–1.83) | - | ||
| By follow-up period† | ||||
| Within 1 month | 15 (43.47)/10 (5.79) | 7.16 (3.07–16.70) | 10 (47.55)/3 (7.96) | 7.21 (1.12–46.5) |
| 2–11 months | 45 (11.38)/116 (5.85) | 1.69 (1.18–2.42) | 29 (12.05)/30 (6.92) | 1.46 (0.83–2.57) |
| 2–9 years | 307 (12.24)/864 (6.75) | 1.61 (1.40–1.85) | 180 (11.39)/217 (7.30) | 1.58 (1.26–1.97) |
| 10 years and onwards | 159 (11.88)/531 (7.51) | 1.59 (1.31–1.92) | 109 (12.00)/165 (9.15) | 1.36 (1.03–1.81) |
-
* Cox regression models were stratified by matching identifiers (birth year and sex, in matched twin cohort) or family identifiers (in twin-sibling cohort), and adjusted for covariates mentioned in the ‘model information’ column. Time since the index date was used as underlying time scale.
† HRs were derived from fully adjusted Cox regression models, that is model four for matched twin cohort and model three for twin-sibling cohort.
Table 3
Hazard ratios (HRs) with 95% confidence intervals (CIs) for any psychiatric disorder among the surviving twins after co-twin loss, compared to matched unexposed twins (matched twin cohort) or their full siblings (twin-sibling cohort), by zygosity of the twin pairs
| Twins who lost a monozygotic twin vs. matched unexposed monozygotic twins or their full siblings | Twins who lost a dizygotic twin vs. matched unexposed dizygotic twins or their full siblings | |||
|---|---|---|---|---|
| Number of cases (Crude incidence rate, per 1000 person years), exposed/unexposed | Hr (95% CI)* | Number of cases (Crude incidence rate, per 1000 person years), exposed/unexposed | Hr (95% CI)* | |
| Matched twin cohort | 92 (14.81)/153 (7.26) | 1.86 (1.40–2.47) | 303 (10.89)/1024 (7.08) | 1.33 (1.15–1.54) |
| By follow-up period | ||||
| Within 1 month | 6 (105.9)/2 (11.12) | 9.47 (1.88–47.8) | 8 (34.81)/5 (4.17) | 4.20 (1.15–15.3) |
| 2–11 months | 12 (18.7)/14 (6.84) | 2.54 (1.01–6.37) | 22 (8.37)/72 (5.23) | 1.23 (0.71–2.13) |
| 2–9 years | 46 (12.40)/93 (7.42) | 1.40 (0.95–2.07) | 190 (11.41)/603 (6.98) | 1.36 (1.13–1.63) |
| 10 years and onwards | 28 (15.56)/44 (6.99) | 2.50 (1.45–4.34) | 83 (10.00)/344 (7.97) | 1.22 (0.93–1.60) |
| By cause of the co-twin’s death | ||||
| Unnatural death | 28 (21.32)/26 (5.58) | 4.29 (2.18–8.45) | 81 (9.79)/254 (5.97) | 1.44 (1.09–1.91) |
| Natural death | 64 (13.07)/127 (7.74) | 1.46 (1.05–2.03) | 222 (11.36)/770 (7.54) | 1.30 (1.10–1.54) |
| By gender difference | ||||
| Twins with same gender | 92 (14.81)/153 (7.26) | 1.86 (1.40–2.47) | 118 (10.51)/240 (6.93) | 1.43 (1.13–1.82) |
| Twins with different gender | - | - | 185 (11.15)/784 (7.13) | 1.33 (1.10–1.62) |
| Twin-sibling cohort | 53 (14.37)/57 (7.24) | 2.45 (1.56–3.85) | 193 (10.73)/279 (8.42) | 1.29 (1.05–1.59) |
-
* Cox regression models were stratified by matching identifiers (birth year and sex, in matched twin cohort) or family identifiers (in twin-sibling cohort), and adjusted for education level, family income, marital status, history of severe somatic diseases, and family history of psychiatric disorder. Time since the index date was used as underlying time scale.
Additional files
-
Source code 1
SAS script for the primary analyses.
- https://cdn.elifesciences.org/articles/56860/elife-56860-code1-v1.zip
-
Supplementary file 1
Supplementary Tables.
Supplementary Table 1. International Classification of Diseases (ICD) codes, eighth (ICD-8; 1969–1986), ninth (ICD-9; 1987–1996), and tenth (ICD-10; 1997–2013) Swedish revisions for diagnoses used in this study Supplementary Table 2. Hazard ratios (HRs) with 95% confidence intervals (CIs) for any psychiatric disorder among the surviving twins after co-twin loss, compared to matched unexposed twins or their full siblings, by different characteristics Supplementary Table 3. Hazard ratios (HRs) with 95% confidence intervals (CIs) for any psychiatric disorder, identified by only primary diagnosis in the National Patient Register or by only diagnosis from National Inpatient Register, among the surviving twins after co-twin loss, compared to matched unexposed twins (matched twin cohort) or their full siblings (twin-sibling cohort).
- https://cdn.elifesciences.org/articles/56860/elife-56860-supp1-v1.docx
-
Transparent reporting form
- https://cdn.elifesciences.org/articles/56860/elife-56860-transrepform-v1.pdf
