Association between antibiotics use and the risk of attention deficit hyperactivity disorder in infants: a Real-World Evidence Study

doi:10.21203/rs.3.rs-3098753/v1

Attention deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder. Recently, infants using antibiotics showed an increased incidence of neurodevelopmental disorders. The purpose of this study was to investigate the association between antibiotics use and the risk of ADHD in infants.

In this population-based retrospective cohort study, the Taiwan National Health Insurance Research Database was used to collect data of infants. Prevalence of antibiotics use was analyzed in the infants (age, < 2 years) included in this study. The risk of developing ADHD was estimated using the Cox proportional hazards model. After controlling for other related influencing factors, infants who used antibiotics had a 1.12 times higher risk of ADHD than those who did not. The risk of ADHD increased through using penicillin and cephalosporin regardless of the duration of antibiotics use.

Therefore, antibiotics use in infants—especially penicillin and cephalosporin— was associated with a higher risk of ADHD.

antibiotics

attention deficit hyperactivity disorder (ADHD)

infant

neurodevelopmental disorder

National Health Insurance Research Database(NHIRD

gut&ndash

brain axis

Attention deficit hyperactivity disorder (ADHD) is a common neurodevelopmental disorder. Although ADHD etiologies remain unclear, genetic factors—in addition to several environmental factors—were reported to play a critical role. An umbrella review in 2020 indicated that ADHD risk factors include mothers with overweight (BMI > 25) or obesity (BMI > 30) in the pre-pregnancy period and the presence of hypertension, preeclampsia, diabetes, prenatal usage of antidepressants and acetaminophen, prenatal exposure to tobacco smoke, premature labor, low birth weight, types of birth (C-section imposes a higher risk than natural birth), breastfeeding (a higher risk is noted in the absence of breastfeeding during the first 3 months), lower education levels of the parents, and single or divorced parents (Kim et al., 2020). A systematic retrospective study on twins and siblings indicated that low birth weight, gestational age, and low household income can also increase the risk of ADHD (Carlsson et al., 2021). Another systematic retrospective study of parental age and offspring ADHD incidence discovered that children born to parents younger than 20 years experience the highest risk of ADHD. The risk declines with the increase in parental age till 31–35 years; the risk then increases subsequently (Min et al., 2021). Moreover, a meta-analysis in 2021 examining children aged 4–18 years with traumatic brain injuries reported that severe traumatic brain injury may increase the risk of ADHD (Asarnow et al., 2021).

Recent research has demonstrated that some gut bacteria can affect the nervous system of the host. Mind-altering probiotics in the gut microbiome can influence the physiology of the gut and central nervous system, and regulate multiple physiological functions of the human body via the gut–brain axis, thereby affecting the patients’ mental status (Wang and Kasper, 2014). The gut bacteria may increase the incidence of ADHD by overproducing reactive oxygen and nitrogen species, thereby causing neuroinflammation and oxidative stress (Checa-Ros et al., 2021). Considering this, factors affecting the growth of the gut microbiome appear to cause neurodevelopmental disorders. Antibiotics are administered to treat bacterial infections. They treat the symptoms of bacterial infection by inhibiting bacterial growth or killing bacteria, while dramatically altering the gut microbiome, thereby affecting brain functions via the gut–brain axis (Wu et al., 2021). The gut microbiome and brain are not fully developed and stable in infants; thus, using antibiotics to treat diseases during this period could induce neurodevelopmental disorders including ADHD. Therefore, the purpose of the present study was to investigate antibiotic use in infants in Taiwan and identify the correlation between antibiotics use and ADHD incidence.

Data sources

This study included data from the National Health Insurance Research Database (NHIRD) and Taiwan’s Maternal and Child Health Database (TMCHD). The NHIRD includes details of beneficiaries (approximately 99.8%) enrolled in Taiwan’s National Health Insurance (NHI) program. The TMCHD is a specific dataset designed to perform family study by further integration of information about newborns, which includes 99.78% of all the births nationwide (Wang et al., 2023). The data analyzed were obtained from the 2002–2017 NHIRD, and the maternal data were obtained by linking it with the TMCHD.

Study participants

The study population consisted of infants born between 2004 and 2012. The exclusion criteria were stillbirths; those who were diagnosed with ADHD, experienced a brain injury (ICD-9-CM: 850–854 or ICD-10-CM: S06), or died within 2 years of birth; and infants with missing data. After acquiring subjects, antibiotics use before the age of 2 years was analyzed. The subjects were divided into the “using antibiotics” and the “not using antibiotics” groups. We defined antibiotic use in infants as those who had outpatient or hospitalization declaration records showing a prescription of oral or injectable routine antibiotics (Anatomical Therapeutic Chemical (ATC) code: J01).

Study design

In this retrospective cohort study, all subjects were tracked from their second birthday until the incidence of ADHD, death, or the end of the data collection period. The result was then analyzed based on the usage of antibiotics. The screening procedure is shown in Fig. 1.

Definition of Variables

In this study, confirmed diagnosis of ADHD (ICD-9-CM: 314, ICD-10-CM: F90) was defined as primary and secondary diagnoses with at least three outpatient visits or at least one hospitalization within a year. The antibiotics were classified into the following categories by the ATC code: (1) Penicillins (J01C), (2) Tetracyclines (J01A), (3) Macrolides (J01F), (4) Aminoglycosides (J01G), (5) Quinolones (J01M), (6) Cephalosporins (J01D), and (7) Others (J01B, J01E, J01R, J01X). Moreover, variable definitions were classified into infant characteristics and mother characteristics. The infant characteristics included sex, gestational age at birth (weeks), and birth weight (gram). Gestational age of < 37 weeks at birth was confirmed by diagnoses with CD-9-CM: 644 or ICD-10-CM: O60. The maternal characteristics included age at birth (years), insurance amount (NTD), insurance location, method of delivery, comorbidities, and medication used during pregnancy. The insurance location was based on the degree of urbanization, where Level 1 denoted the highest degree of urbanization and Level 7 denoted the lowest. Comorbidities like hypertension (including preeclampsia, ICD-9-CM: 401–405, 642, ICD-10-CM: I10-I15, O10, O11, O13, O14, O16) and diabetes (including gestational diabetes, ICD-9-CM: 249, 250, 648, ICD-10-CM: E8-E13, O24) were defined as primary and secondary diagnoses of diseases that incurred at least three outpatient visits or at least one hospitalization within a year before birth. The medications used during pregnancy were according to the prescription records during pregnancy, classified by the accumulated days of use into the following categories: no use, < 28 days of use, and ≥ 28 days of use by the mothers. The specific medications included antibiotics (J01), antidepressants (N06A), and acetaminophen (N02BE01).

Statistical analysis

SAS software version 9.4 (SAS Institute, Cary, NC, USA) was used to conduct the statistical analysis. Descriptive statistics were used to analyze the infant and mother characteristics, including the number of individuals, percentage, mean, and standard deviation. The Chi-square test was then conducted to analyze the distribution differences between variables. After further controlling for other related variables, the Cox proportional hazards model was used to analyze the difference in the risk of ADHD between infants in the using and not using antibiotics groups.

The significance level (α level) for all the statistical analyses in this study was set at 0.05; thus, p < 0.05 indicated statistical significance.

Description of the study population

Among the 1,601,689 infants included in this study, 1,141,146 (71.25%) and 460,543 (28.75%) were included in the using and not using antibiotics groups, respectively. Table 1 shows the basic characteristics of the infants and the distribution of antibiotic use. Among the included infants, 833,615 were male (52.05%), 1,462,043 had a gestational age of ≥ 37 weeks at birth (91.28%), and 1,473,075 were heavier than 2,500 grams at birth (91.97%).

Regarding the mother characteristics, the age of 551,549 (34.44%) and 578,908 (36.14%) mothers at birth was 25–29 and 30–34 years, respectively. The delivery method in 1,029,953 (64.30%) mothers was natural birth. The insurance amount for 772,467 people (48.23%) was within the range of 18,301–34,800 NTD; furthermore, the insurance location for 469,984 (29.34%) and 527,377 (32.93%) people were ranked at Levels 1 and 2, respectively. Hypertension and diabetes were not observed in 1,583,919 (98.89%) and 1,561,186 (97.47%) pregnant women, respectively. Among medications taken during pregnancy, acetaminophen was taken by 814,033 people (50.82%), antibiotics by 539,310 people (33.64%), and antidepressants by 8,194 people (0.52%).

Table 1

Basic characteristics of the infants and the distribution of antibiotic use
Variables	Total		Infant antibiotic use				P
	Total		No		Yes		P
	N	%	N	%	N	%
Total	1,601,689	100.00	460,543	28.75	1,141,146	71.25
Infant characteristics
Sex							< 0.001
Male	833,615	52.05	222,811	48.38	610,804	53.53
Female	768,074	47.95	237,732	51.62	530,342	46.47
Gestational age at birth (weeks)							< 0.001
≥ 37	1,462,043	91.28	421,751	91.58	1,040,292	91.16
< 37	139,646	8.72	38,792	8.42	100,854	8.84
Birth weight (grams)							< 0.001
≥ 2,500	1,473,075	91.97	422,744	91.79	1,050,331	92.04
< 2,500	128,614	8.03	37,799	8.21	90,815	7.96
Mother characteristics
Age at birth (years)							< 0.001
< 25	238,236	14.87	63,941	13.88	174,295	15.27
25–29	551,549	34.44	152,343	33.08	399,206	34.98
30–34	578,908	36.14	169,403	36.78	409,505	35.89
35–39	204,073	12.74	64,540	14.01	139,533	12.23
≥ 40	28,923	1.81	10,316	2.24	18,607	1.63
Mean ± SD	29.99 ± 4.78		30.27 ± 4.85		29.87 ± 4.74
Insurance amount (NTD)							< 0.001
Low (≤ 18,300)	453,088	28.29	131,130	28.47	321,958	28.21
Middle (18,301–34,800)	772,467	48.23	212,781	46.20	559,686	49.05
High (≥ 34,801)	376,134	23.48	116,632	25.32	259,502	22.74
Insurance location ^a							< 0.001
Level 1	469,984	29.34	143,976	31.26	326,008	28.57
Level 2	527,377	32.93	153,003	33.22	374,374	32.81
Level 3	300,238	18.75	85,633	18.59	214,605	18.81
Level 4	189,404	11.83	51,375	11.16	138,029	12.10
Level 5	20,818	1.30	5,168	1.12	15,650	1.37
Level 6	44,256	2.76	10,407	2.26	33,849	2.97
Level 7	49,612	3.10	10,981	2.38	38,631	3.39
Method of Delivery							< 0.001
Natural birth	1,029,953	64.30	302,251	65.63	727,702	63.77
C-section	571,736	35.70	158,292	34.37	413,444	36.23
Hypertension							< 0.001
No	1,583,919	98.89	455,766	98.96	1,128,153	98.86
Yes	17,770	1.11	4,777	1.04	12,993	1.14
Diabetes							< 0.001
No	1,561,186	97.47	449,229	97.54	1,111,957	97.44
Yes	40,503	2.53	11,314	2.46	29,189	2.56
Medication use during pregnancy
Antibiotics							< 0.001
No	1,062,379	66.33	322,835	70.10	739,544	64.81
< 28 days	527,520	32.94	135,433	29.41	392,087	34.36
≥ 28 days	11,790	0.74	2,275	0.49	9,515	0.83
Antidepressants							< 0.001
No	1,593,495	99.49	458,452	99.55	1,135,043	99.47
< 28 days	6,509	0.41	1,640	0.36	4,869	0.43
≥ 28 days	1,685	0.11	451	0.10	1,234	0.11
Acetaminophen							< 0.001
No	787,656	49.18	244,863	53.17	542,793	47.57
< 28 days	767,530	47.92	206,970	44.94	560,560	49.12
≥ 28 days	46,503	2.90	8,710	1.89	37,793	3.31

^a Level 1 denotes the highest and Level 7 denotes the lowest degree of urbanization.

The correlation between infant antibiotic use and the occurrence of ADHD

As shown in Table 2, model 1 demonstrated that after controlling for other related variables, infants who used antibiotics had a 1.12 times higher risk of developing ADHD than those who did not (95% CI = 1.10–1.14). In model 2, regarding antibiotic types, infants using penicillin and cephalosporin had higher risks of ADHD than those not using these antibiotics. Among the penicillin users, the cumulative use period groups of 1–7, 8–14, 15–21, and > 21 days had risks of 1.08 (95% CI = 1.07–1.10), 1.10 (95% CI = 1.07–1.14), 1.08 (95% CI = 1.03–1.14), and 1.22 (95% CI = 1.16–1.28), respectively. The corresponding risks among the cephalosporin users were 1.09 (95% CI = 1.07–1.10), 1.05 (95% CI = 1.01–1.10), 1.12 (95% CI = 1.03–1.21), and 1.10 (95% CI = 1.01–1.21), respectively. The ADHD risk increases with the increase in the cumulative days of use. In contrast, infants using tetracycline, macrolide, aminoglycoside, quinolone, or other antibiotics showed no significant difference in ADHD risks from the non-users. The male infants (Hazard Ratio (HR) = 3.12, 95% CI = 3.03–3.12), infants with gestational age of < 37 weeks (HR = 1.35, 95% CI = 1.32–1.38), and infants that had a birth weight of < 2,500 grams (HR = 1.33, 95% CI = 1.30–1.36) all showed higher risks of developing ADHD.

All mother characteristics displayed significant differences, except for use of acetaminophen cumulatively for < 28 days during pregnancy, which showed no significant difference. The maternal age of > 40 years at childbirth imposed the highest risk. The ADHD risk in infants delivered by C-section was 1.10 times higher than in infants of natural birth (95% CI = 1.09–1.12). Regarding insurance amount and location, those with < 18,300 NTD and those in location Level 1 showed the highest risk of occurrence. The group with comorbidities of hypertension (HR = 1.46, 95% CI = 1.38–1.54) and diabetes (HR = 1.23, 95% CI = 1.18–1.29) showed higher risks than those without those diseases. Medication use during pregnancy showed the highest risk in those with a cumulative use of ≥ 28 days for antibiotics (HR = 1.29, 95% CI = 1.20–1.40), antidepressants (HR = 1.98, 95% CI = 1.69–2.32), and acetaminophen (HR = 1.15, 95% CI = 1.10–1.20).

Table 2

ADHD risk in infants using antibiotics
Variables	Unadjusted model					Adjusted model 1					Adjusted model 2
Variables	HR	95% CI			P	HR	95% CI			P	HR	95% CI			P
Infant antibiotic use
Any antibiotics
No (control group)	1					1
Yes	1.18	1.16	–	1.20	< 0.001	1.12	1.10	–	1.14	< 0.001	-		-		-
Penicillins
No (control group)	1										1
1–7 days	1.13	1.12	–	1.15	< 0.001	-		-		-	1.08	1.07	–	1.10	< 0.001
8–14 days	1.19	1.16	–	1.23	< 0.001	-		-		-	1.10	1.07	–	1.14	< 0.001
15–21 days	1.18	1.12	–	1.24	< 0.001	-		-		-	1.08	1.03	–	1.14	0.004
> 21 days	1.34	1.28	–	1.41	< 0.001	-		-		-	1.22	1.16	–	1.28	< 0.001
Tetracyclines
No (control group)	1										1
1–7 days	0.98	0.91	–	1.05	0.532	-		-		-	0.92	0.86	–	0.99	0.019
8–14 days	1.04	0.74	–	1.46	0.835	-		-		-	0.93	0.66	–	1.31	0.694
Macrolides
No (control group)	1										1
1–7 days	1.09	1.07	–	1.11	< 0.001	-		-		-	1.01	0.99	–	1.03	0.314
8–14 days	1.14	1.04	–	1.25	0.005	-		-		-	1.00	0.91	–	1.10	0.953
15–21 days	0.94	0.77	–	1.15	0.564	-		-		-	0.83	0.68	–	1.02	0.076
> 21 days	1.01	0.82	–	1.25	0.926	-		-		-	0.87	0.70	–	1.08	0.206
Aminoglycosides
No (control group)	1										1
1–7 days	1.18	1.09	–	1.27	< 0.001	-		-		-	1.13	1.05	–	1.22	0.002
8–14 days	1.76	0.79	–	3.89	0.164	-		-		-	1.71	0.77	–	3.82	0.188
Quinolones
No (control group)	1										1
1–7 days	1.02	0.97	–	1.08	0.377	-		-		-	0.97	0.92	–	1.02	0.202
8–14 days	0.90	0.69	–	1.17	0.421	-		-		-	0.80	0.62	–	1.05	0.107
15–21 days	1.21	0.77	–	1.92	0.407	-		-		-	1.04	0.65	–	1.65	0.871
> 21 days	1.09	0.52	–	2.28	0.823	-		-		-	0.96	0.46	–	2.02	0.921
Cephalosporins
No (control group)	1										1
1–7 days	1.14	1.13	–	1.16	< 0.001	-		-		-	1.09	1.07	–	1.10	< 0.001
8–14 days	1.16	1.11	–	1.21	< 0.001	-		-		-	1.05	1.01	–	1.10	0.030
15–21 days	1.25	1.15	–	1.36	< 0.001	-		-		-	1.12	1.03	–	1.21	0.008
> 21 days	1.26	1.15	–	1.38	< 0.001	-		-		-	1.10	1.01	–	1.21	0.037
Other antibiotics
No (control group)	1										1
1–7 days	1.05	1.03	–	1.07	< 0.001	-		-		-	1.00	0.97	–	1.02	0.690
8–14 days	1.17	1.07	–	1.28	< 0.001	-		-		-	1.08	0.99	–	1.18	0.086
15–21 days	1.03	0.86	–	1.23	0.775	-		-		-	0.95	0.80	–	1.14	0.612
> 21 days	1.23	1.03	–	1.47	0.021	-		-		-	1.09	0.92	–	1.31	0.324
Infant characteristics
Sex
Male (control group)	1					1					1
Female	0.32	0.32	–	0.33	< 0.001	0.32	0.32	–	0.33	< 0.001	0.32	0.32	–	0.33	< 0.001
Gestational age at birth (weeks)
≥ 37 (control group)	1					1					1
< 37	1.35	1.32	–	1.38	< 0.001	1.11	1.08	–	1.14	< 0.001	1.11	1.08	–	1.14	< 0.001
Birth weight (grams)
≥ 2,500 (control group)	1					1					1
< 2,500	1.33	1.30	–	1.36	< 0.001	1.30	1.26	–	1.33	< 0.001	1.30	1.26	–	1.33	< 0.001
Mother characteristics
Age at birth (years)
< 25	0.94	0.89	–	0.99	0.012	1.00	0.95	–	1.05	0.902	0.99	0.94	–	1.05	0.811
25–29	0.82	0.78	–	0.86	< 0.001	0.86	0.82	–	0.91	< 0.001	0.86	0.82	–	0.90	< 0.001
30–34	0.79	0.75	–	0.83	< 0.001	0.83	0.79	–	0.88	< 0.001	0.83	0.79	–	0.87	< 0.001
35–39	0.86	0.81	–	0.90	< 0.001	0.88	0.84	–	0.93	< 0.001	0.88	0.83	–	0.93	< 0.001
≥ 40 (control group)	1					1					1
Insurance amount (NTD)
Low (≤ 18,300) (control group)	1					1					1
Middle (18,301–34,800)	0.90	0.88	–	0.91	< 0.001	0.89	0.87	–	0.90	< 0.001	0.88	0.87	–	0.90	< 0.001
High (≥ 34,801)	0.80	0.78	–	0.81	< 0.001	0.77	0.76	–	0.79	< 0.001	0.77	0.76	–	0.79	< 0.001
Insurance location^a
Level 1 (control group)	1					1					1
Level 2	0.90	0.89	–	0.92	< 0.001	0.88	0.86	–	0.89	< 0.001	0.88	0.86	–	0.89	< 0.001
Level 3	0.81	0.80	–	0.83	< 0.001	0.77	0.75	–	0.78	< 0.001	0.77	0.75	–	0.78	< 0.001
Level 4	0.76	0.74	–	0.78	< 0.001	0.71	0.69	–	0.73	< 0.001	0.71	0.69	–	0.73	< 0.001
Level 5	0.73	0.69	–	0.78	< 0.001	0.67	0.63	–	0.72	< 0.001	0.67	0.63	–	0.72	< 0.001
Level 6	0.68	0.65	–	0.71	< 0.001	0.61	0.58	–	0.64	< 0.001	0.61	0.58	–	0.64	< 0.001
Level 7	0.68	0.65	–	0.71	< 0.001	0.61	0.59	–	0.64	< 0.001	0.61	0.58	–	0.64	< 0.001
Method of Delivery
Natural birth (control group)	1					1					1
C-section	1.10	1.09	–	1.12	< 0.001	1.05	1.03	–	1.06	< 0.001	1.05	1.03	–	1.06	< 0.001
Hypertension
No (control group)	1					1					1
Yes	1.46	1.38	–	1.54	< 0.001	1.32	1.25	–	1.40	< 0.001	1.32	1.25	–	1.40	< 0.001
Diabetes
No (control group)	1					1					1
Yes	1.23	1.18	–	1.29	< 0.001	1.21	1.16	–	1.26	< 0.001	1.20	1.16	–	1.26	< 0.001
Medication use during pregnancy
Antibiotics
No (control group)	1					1					1
< 28 days	1.08	1.07	–	1.10	< 0.001	1.06	1.05	–	1.08	< 0.001	1.06	1.04	–	1.08	< 0.001
≥ 28 days	1.29	1.20	–	1.40	< 0.001	1.20	1.11	–	1.30	< 0.001	1.20	1.10	–	1.30	< 0.001
Antidepressants
No (control group)	1					1					1
< 28 days	1.60	1.46	–	1.74	< 0.001	1.53	1.40	–	1.67	< 0.001	1.53	1.40	–	1.67	< 0.001
≥ 28 days	1.98	1.69	–	2.32	< 0.001	1.83	1.56	–	2.15	< 0.001	1.83	1.56	–	2.15	< 0.001
Acetaminophen
No (control group)	1					1					1
< 28 days	1.01	1.00	–	1.03	0.090	0.97	0.95	–	0.98	< 0.001	0.96	0.95	–	0.98	< 0.001
≥ 28 days	1.15	1.10	–	1.20	< 0.001	1.05	1.00	–	1.09	0.052	1.04	0.99	–	1.08	0.101

^a Level 1 denotes the highest and Level 7 denotes the lowest degree of urbanization.

Use of antibiotics in infants

The results demonstrated that 71.25% of infants younger than 2 years had used at least 1 antibiotic, which is similar to the 70% reported in a US study, lower than the 81.7% in Finland, but much higher than the 35.2% in the Netherlands (Aversa et al., 2021; Lavebratt et al., 2019; Slob et al., 2021). Regarding the types of antibiotics used, penicillins, cephalosporins, and macrolides were the main categories, consistent with relevant studies in other countries where these three types are also the most commonly used antibiotics. The usage proportions were 52.5%, 40.9%, and 15.7%, respectively. In comparison, the ratios in the US are 42.9%, 29.1%, and 19.2%; in South Korea they are 56.6%, 25.3%, and 17.0%; in Italy (Lazio region) they are 51.5%, 24.7%, and 23.2%; in Germany they are 38.4%, 42.2%, and 18.2%; in Spain they are 62.1%, 20.6%, and 17.0%; and in Norway they are 66.7%, 1.8%, and 21.9% (Youngster et al., 2017). The proportions of penicillins and macrolides usage reported in this study are approximately the same as in other countries; however, the proportion of cephalosporin is higher (except for in Germany). The usage proportions are similar to the US, perhaps due to Taiwan's medical standards mostly being based on US guidelines. The cumulative duration of antibiotic use was mainly short-term use of 1–7 days. A cumulative use of less than 14 days’ accounts for between 93.6% and 100% of cases, in line with the recommendations of the Taiwan Centers for Disease Control (Taiwan Centers for Disease Control, 2015).

Regarding mother characteristics, those from location with higher levels of urbanization showed higher prescription rates, with Level 1 and 2 regions accounting for approximately 50% of the issued prescriptions. The prescription rate was highest for the group with medium insurance amounts. These results are similar to a study conducted in Canada. However, 36.2% of infants born by C-section took antibiotics within 2 years of birth, a proportion nearly twice as high as the 20.0% within 1 year in the Canadian study, and the 23.8% within 2 years in the US study. However, it is much lower than the 75% within 2 years in the Danish study (Aversa et al., 2021; Axelsson et al., 2019; Hamad et al., 2019). There are significant differences in the results of studies from different countries. The reasons for such discrepancy cannot be determined from the present literature. This issue deserves further investigation in the future.

The correlation between antibiotic use in infants and the occurrence of ADHD

The elevated risk of ADHD among those who have used antibiotics is found to be 1.32 times greater (95% CI = 1.15–1.53) than that of the study from the US, and 1.16 times greater (younger than 6 months), 1.12 times greater (6–11 months), and 1.26 times greater (1–2 years) than that of the study from Finland that grouped antibiotic exposure by age. A meta-analysis of 6 studies also showed a similar odds ratio of 1.18 (95% CI = 1.1–1.27). Moreover, all these results are statistically significant, indicating that infants who have used antibiotics will indeed have an increased risk of developing ADHD (Aversa et al., 2021; Lavebratt et al., 2019; Yu et al., 2022).

Regarding antibiotic type and cumulative days of use, the US study (Aversa et al., 2021) only showed a significant difference in the HR for female infants using penicillin, which was 1.50 (95% CI = 1.14–1.96). In contrast, the Canadian study (Hamad et al., 2019) showed significant differences for all antibiotic types tested, with cephalosporins having the highest risk at 1.23 (95% CI = 1.07–1.42), and only cumulative use exceeding 21 days was found to have a significant difference at 1.38 (95% CI = 1.17–1.64). In this study, penicillin and cephalosporins showed significant differences, and the risk increased for all durations of cumulative use. Based on the results of these studies, it can be suggested that if infants need antibiotics to treat infections, penicillins and cephalosporins should be avoided whenever possible.

Regarding maternal characteristics, the lowest risk was observed in the mothers of age 30–34 years, with a risk of 0.79 (95% CI = 0.75–0.83), followed by mothers in 25–29 year group with a risk of 0.82 (95% CI = 0.78–0.86). Although these groups had the highest antibiotic usage, they had the lowest risk of developing ADHD. This may be attributed to the fact that ADHD is primarily influenced by genetic factors, and women ranged 25–35 years of age are in their prime childbearing years, where the genetic risk is reduced. Therefore, the genetic factors outweigh the impact of antibiotic usage, resulting in such outcomes.

Since this study only analyzed data from the databases, there are some limitations. First, real-life medication usage may differ from the patterns presented by the data. However, as prescription drugs, antibiotics are less likely to raise concerns over the out-of-pocket costs, and prematurely discontinuing treatment may lead to antibiotic resistance issues; therefore, the medication compliance should be consistent with that shown in the database. In addition, the NHIRD analysis could not reveal whether pregnant women and infants had taken probiotics as nutritional supplements, which would have altered the ecology of the gut microbiome and consequently indirectly affected the results. Furthermore, the database does not collect information on healthy behaviors and lifestyles, such as whether a woman smoked during pregnancy, whether the mother chooses to breastfeed, her education level, and her marital status. These variables may all affect the outcomes of the study.

This study confirms a significant correlation between infants using antibiotics and the occurrence of ADHD, especially in those who use penicillin and cephalosporin, regardless of the duration of use. Therefore, when infants show symptoms of infection, it is recommended to avoid using penicillins and cephalosporins as much as possible in order to reduce the risk of developing ADHD later in life in infants.

Author contributions: Conceptualization: Chih-Kang Lin, Ya-Chun Tseng; Methodology: Chih-Kang Lin, Ya-Chun Tseng, Kuang-Hua Huang; Formal analysis and investigation: Chih-Kang Lin, Hsing-Yu Hsu, Tung-Han Tsai; Writing-original draft: Chih-Kang Lin, Ya-Chun Tseng, Hsing-Yu Hsu, Tung-Han Tsai; Writing-review and editing: Kuang-Hua Huang. All authors have read and agreed to the published version of the manuscript. Chih-Kang Lin and Ya-Chun Tseng have contributed equally to this work.

Funding: This study was funded by the Cheng Ching General Hospital, Taiwan (Protocol No. 11106).

Conflicts of Interest: The authors have no competing interests to declare that are relevant to the content of this article.

Ethics approval: This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the Ethics Committee of Cheng Ching General Hospital, Taiwan (Date 22 July 2022/No. HP220021).

Informed Consent: The Health and Welfare Data Science Center (HWDC) have conducted deidentification of insured individuals to protect their privacy. The personal identities have been removed from the database used in the study, rendering it anonymous. Due to this anonymization process, the requirement for informed consent is waived.

Data Availability: The Databases used to support the findings of this study was provided by the HWDC, Ministry of Health and Welfare (MOHW) under license and so cannot be made freely available. Requests for access to these data should be made to HWDC (https://dep.mohw.gov.tw/dos/cp-5119-59201-113.html, accessed on 20-May-2023).

Acknowledgments: We thank Crimson Interactive Pvt. Ltd. (Ulatus) – www.ulatus.tw for their assistance in manuscript translation and editing. This study is based on the data released by the Health and Welfare Data Science Center, Ministry of Health and Welfare. The interpretation and conclusions contained herein do not represent those of the Ministry of Health and Welfare.

Asarnow RF, Newman N, Weiss RE, Su E (2021) Association of attention-deficit/hyperactivity disorder diagnoses with pediatric traumatic brain injury: A meta-analysis. JAMA Pediatr 175(10):1009–1016. https://doi.org/10.1001/jamapediatrics.2021.2033
Aversa Z, Atkinson EJ, Schafer MJ, Theiler RN, Rocca WA, Blaser MJ, LeBrasseur NK (2021) Association of infant antibiotic exposure with childhood health outcomes. Mayo Clinic Proceedings, 96(1), 66–77. https://doi.org/10.1016/j.mayocp.2020.07.019
Axelsson PB, Clausen TD, Petersen AH, Hageman I, Pinborg A, Kessing LV, Bergholt T, Rasmussen SC, Keiding N, Løkkegaard ECL (2019) Investigating the effects of cesarean delivery and antibiotic use in early childhood on risk of later attention deficit hyperactivity disorder. J Child Psychol Psychiatry Allied Discip 60(2):151–159. https://doi.org/10.1111/jcpp.12961
Carlsson T, Molander F, Taylor MJ, Jonsson U, Bölte S (2021) Early environmental risk factors for neurodevelopmental disorders – A systematic review of twin and sibling studies. Dev Psychopathol 33(4):1448–1495. https://doi.org/10.1017/S0954579420000620
Checa-Ros A, Jeréz-Calero A, Molina-Carballo A, Campoy C, Muñoz-Hoyos A (2021) Current evidence on the role of the gut microbiome in ADHD pathophysiology and therapeutic implications. Nutrients 13(1):249. https://doi.org/10.3390/nu13010249
Hamad AF, Alessi-Severini S, Mahmud SM, Brownell M, Kuo IF (2019) Antibiotic exposure in the first year of life and the risk of attention-deficit/hyperactivity disorder: A population-based cohort study. Am J Epidemiol 188(11):1923–1931. https://doi.org/10.1093/aje/kwz178
Kim JH, Kim JY, Lee J, Jeong GH, Lee E, Lee S, Lee KH, Kronbichler A, Stubbs B, Solmi M, Koyanagi A, Hong SH, Dragioti E, Jacob L, Brunoni AR, Carvalho AF, Radua J, Thompson T, Smith L, Fusar-Poli P (2020) Environmental risk factors, protective factors, and peripheral biomarkers for ADHD: An umbrella review. Lancet Psychiatry 7(11):955–970. https://doi.org/10.1016/S2215-0366(20)30312-6
Lavebratt C, Yang LL, Giacobini M, Forsell Y, Schalling M, Partonen T, Gissler M (2019) Early exposure to antibiotic drugs and risk for psychiatric disorders: A population-based study. Translational Psychiatry 9(1):317. https://doi.org/10.1038/s41398-019-0653-9
Min X, Li C, Yan Y (2021) Parental age and the risk of ADHD in offspring: A systematic review and meta-analysis. Int J Environ Res Public Health 18(9):4939. https://doi.org/10.3390/ijerph18094939
Slob EMA, Brew BK, Vijverberg SJH, Dijs T, van Beijsterveldt CEM, Koppelman GH, Bartels M, Dolan CV, Larsson H, Lundström S, Lichtenstein P, Gong T, -van der Maitland AH, Kraneveld AD, Almqvist C, Boomsma DI (2021) Early-life antibiotic use and risk of attention-deficit hyperactivity disorder and autism spectrum disorder: Results of a discordant twin study. Int J Epidemiol 50(2):475–484. https://doi.org/10.1093/ije/dyaa168
Taiwan Centers for Disease Control (2015) National action plan antimicrobial stewardship program. Retrieved June 20, 2023, from https://www.cdc.gov.tw/En/InfectionReport/Info/9YUAXbFsmorP5T10V8qvMA?infoId=rsr-iHn43cvyhpnTNZMp3Q
Wang CM, Yang ST, Yang CC, Chiu HY, Lin HY, Tsai ML, Lin HC, Chang YC (2023) Maternal and neonatal risk factors of asthma in children: Nationwide population based study. J Microbiol Immunol Infect 56(1):182–191. https://doi.org/10.1016/j.jmii.2022.10.005
Wang Y, Kasper LH (2014) The role of microbiome in central nervous system disorders. Brain Behav Immun 38:1–12. https://doi.org/10.1016/j.bbi.2013.12.015
Wu WL, Adame MD, Liou CW, Barlow JT, Lai TT, Sharon G, Schretter CE, Needham BD, Wang MI, Tang W, Ousey J, Lin YY, Yao TH, Abdel-Haq R, Beadle K, Gradinaru V, Ismagilov RF, Mazmanian SK (2021) Microbiota regulate social behaviour via stress response neurons in the brain. Nature 595(7867):409–414. https://doi.org/10.1038/s41586-021-03669-y
Youngster I, Avorn J, Belleudi V, Cantarutti A, Díez-Domingo J, Kirchmayer U, Park BJ, Peiró S, Sanfélix-Gimeno G, Schröder H, Schüssel K, Shin JY, Shin SM, Simonsen GS, Blix HS, Tong A, Trifirò G, Ziv-Baran T, Kim SC (2017) Antibiotic use in children – A cross-national analysis of 6 countries. J Pediatr 182:239–244. https://doi.org/10.1016/j.jpeds.2016.11.027
Yu HY, Zhou YY, Pan LY, Zhang X, Jiang HY (2022) Early life antibiotic exposure and the subsequent risk of autism spectrum disorder and attention deficit hyperactivity disorder: A systematic review and meta-analysis. J Autism Dev Disord 52(5):2236–2246. https://doi.org/10.1007/s10803-021-05121-6

No competing interests reported.

Association between antibiotics use and the risk of attention deficit hyperactivity disorder in infants: a Real-World Evidence Study

Status:

Version 1

Abstract

Figures

Introduction

Material and methods

Data sources

Study participants

Study design

Definition of Variables

Statistical analysis

Results

Description of the study population

The correlation between infant antibiotic use and the occurrence of ADHD

Discussion

Use of antibiotics in infants

The correlation between antibiotic use in infants and the occurrence of ADHD

Conclusions

Declarations

References

Additional Declarations

Status:

Version 1