. 2020 Mar 18;17(3):e1003078. doi: 10.1371/journal.pmed.1003078

Table 3. Cause-specific mortality rate ratio (MMR) for the childhood obesity cohort compared to the comparison group.

Cause of death	Number of deaths		Mortality rate per 10,000 person-years	Crude MRR (95% CI); p-value	Adjusted MRR (95% CI); p-value
Cause of death	Childhood obesity cohort	Comparison group	Mortality rate per 10,000 person-years	Crude MRR (95% CI); p-value	Adjusted MRR (95% CI); p-value
Injuries and external causes	7	14	1.10	2.44 (0.98–6.04); 0.054	2.38 (0.96–5.94); 0.063
Endogenous causes	15	17	1.68	4.30 (2.15–8.61); <0.001	4.04 (2.00–8.17); <0.001
Suicide and self-harm	16	31	2.46	2.51 (1.38–4.60); 0.003	2.15 (1.17–3.95); 0.014
Excluding individuals with genetic syndromes^a and malignant tumors^b in childhood
Injuries and external causes	7	14	1.33	2.49 (1.01–6.17); 0.049	2.43 (0.97–6.06); 0.057
Endogenous causes	11	12	1.46	4.56 (2.01–10.33); <0.001	4.50 (2.00–10.32); <0.001
Suicide and self-harm	15	31	2.92	2.40 (1.30–4.45); 0.005	2.03 (1.09–3.78); 0.026

Crude and adjusted models yielded MRRs for childhood obesity cohort versus comparison group; adjusted model controlled for sex, Nordic origin, and parental SES. Endogenous causes of death included deaths from pathogens, acquired disorders, congenital disorders, etc. Suicide and self-harm included intentional deaths from suicide as well as deaths with unintentional or unclear intention from poisoning, e.g., illicit drugs. Injuries and other external causes included deaths such as traffic accidents and homicide.

^aFragile X, Klinefelter, Laurence–Moon–Bardet–Biedl, Down, Noonan, Prader–Willi, Silver–Russell, and Turner.

^bIncluding benign brain tumors.