OBJECTIVES:

The youngest children in a classroom are at increased risk of being medicated for attention-deficit/hyperactivity disorder (ADHD). We examined the association between children’s birth month and ADHD medication rates in Finland.

METHODS:

Using a population-based study, we analyzed ADHD medication use among children born in 2005 to 2007. Cases (n = 7054) were identified from the first purchase of medication for ADHD. Cox proportional hazard models and hazard ratios (HRs) were examined by birth month and sex. Finnish children start first grade in the year of their seventh birthday. The cutoff date is December 31.

RESULTS:

Risk of ADHD medication use increased throughout the year by birth month (ie, January through April to May through August to September through December). Among boys born in September to December, the association remained stable across cohorts (HR: 1.3; 95% confidence interval [CI]: 1.1–1.5). Among girls born in September to December, the HR in the 2005 cohort was 1.4 (95% CI: 1.1–1.8), whereas in the 2007 cohort it was 1.7 (95% CI: 1.3–2.2). In a restricted follow-up, which ended at the end of the year of the children’s eighth birthday, the HRs for boys and girls born in September to December 2007 were 1.5 (95% CI: 1.3–1.7) and 2.0 (95% CI: 1.5–2.8), respectively.

CONCLUSIONS:

Relative immaturity increases the likelihood of ADHD medication use in Finland. The association was more pronounced during the first school years. Increased awareness of this association is needed among clinicians and teachers.

What’s Known in This Subject:

The relative age effect (RAE) in attention-deficit/hyperactivity disorder (ADHD) indicates that ADHD medication use is more common in the relatively youngest children within a school grade. Within-country sex differences in ADHD medication use vary and are among the largest in Finland.

What This Study Adds:

This population-based study reveals that the RAE in ADHD medication use among children was more pronounced during the first school years (ages 6–8). The RAE in ADHD medication was stable among boys across 3 cohorts but increasing among girls.

Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental condition characterized by age-inappropriate levels and persistent patterns of inattention or hyperactivity-impulsivity or both, which interferes with social functioning and development.1,2  ADHD is recognized as a lifelong disorder.3,4  Prevalence of ADHD is estimated to be ∼5% in children.2,5  However, the relative age effect (RAE) refers to the fact that the youngest children in classrooms are more likely to receive a diagnosis and be prescribed medication for ADHD than their relatively older classmates.68 Researchers of previous studies have indicated that relative age plays a more important role with ADHD symptoms than the season of birth,8,9  particularly during childhood.10  One line of research suggests that RAE may represent a situation in which the more immature behavior of younger children is treated as ADHD.1113  There is empirical evidence of the RAE from countries with high prescription rates, such as Iceland,14  the United States,15  Canada,16  and Sweden,17  but also from countries with moderate rates, such as Norway18  and Taiwan.19 

The strength of the RAE may vary within countries across time. A Danish study found evidence for an RAE between 2000 and 2004,20  which then diminished during study years 2006 to 2012.20,21  These findings are possibly explained by the fact that ∼40% of the relatively younger children are held back one year in the Danish school system.20,22  The fact that the RAE is associated with school entry regulations further implies that the effect size may depend on the duration of the follow-up.23  This is an under-researched perspective.17  There is also notable variation between countries in prescribing rates for ADHD, and studies also differ by the years studied.6,24  For example, prevalence rates of ADHD medication use among the child population in Finland have been relatively low, particularly among girls,24,25  but the rates of ADHD diagnoses and medication use are increasing relatively rapidly.26  Sex differences are also important to consider because the rates of ADHD vary by sex in middle childhood. Emotion dysregulation and externalizing behaviors drive referral for ADHD.27,28  However, in recent studies on sex differences, it has been indicated that emotional and behavioral difficulties are stronger predictors of ADHD diagnosis in girls when compared with boys.29,30  In addition, girls with externalizing symptoms are more likely to have a referral for ADHD at a younger age compared with boys with similar behaviors.30  Interestingly, the findings with regard to sex differences in the RAE have been inconclusive.14,16,18,20  This may relate to the fact that sex differences in ADHD medication use vary between countries. In a recent study in which worldwide ADHD medication use in children was examined, the male-to-female ratio in medication use was the highest in Finland.24 

Our purpose with this study was to examine the extent to which relative age is associated with ADHD medication use among children in Finland born in 2005 to 2007. Importantly, we were able to use the study by Sayal et al,13  which examined the association between relative age and an ADHD diagnosis among Finnish children (born between 1991 and 2004) as a baseline. Altogether, with our study, we add to the global understanding of the RAE in ADHD by providing a more-complete picture of the phenomenon from a single jurisdiction where the rates of children with ADHD have increased somewhat rapidly and delayed school entry has decreased.13,26  First, we hypothesized that with delayed school entry being somewhat rare in Finland, there would be at least a modest RAE in prescribing ADHD medication for children. Second, given that difference in maturity is more extreme at young ages, we expected that the RAE in ADHD medication use would be stronger at ages 6 to 7, when children start primary education. Third, because of the prominent sex differences in the rates of ADHD, we expected that the RAE would be different in boys and girls.

The study outcome was ADHD medication prescribing. All permanent residents of Finland are covered under the National Health Insurance system. Reimbursements for medical expenses are available for products that the Pharmaceuticals Pricing Board has confirmed as reimbursable. A particular medicine can only be reimbursed with respect to specific medical indications noted in the summary of product characteristics. For example, children are eligible for reimbursements for the costs of ADHD prescription medication when the prescription indication is ADHD. Reimbursements from all Finnish pharmacies are recorded in the National Prescription Register maintained by the Social Insurance Institution. The register contains information on the child’s birth date and sex and the prescription dispensing date. Although the register does not include information on ADHD diagnoses, all children presumably had an adequate diagnosis of ADHD before a medication prescription was dispensed.

ADHD medication use was examined by age-, sex-, and birth month–matched population, with follow-up data from children born in 2005, 2006, and 2007 with permanent residence in Finland (N = 182 802) at the end of December 31, 2017. The children were followed from their sixth birthday until the end of calendar years of their 12th (born 2005), 11th (born 2006), or 10th (born 2007) birthday.

The first purchase of medication for ADHD was the main outcome. Altogether, 7054 cases with medication were identified. Medication for ADHD included all the substances reimbursable in Finland within the category of centrally acting sympathomimetic drugs according to the Anatomic Therapeutic Chemical Classification System (N06BA): methylphenidate, atomoxetine, dexamphetamine, and lisdexamphetamine. The prevalence of ADHD medication use was defined as the number of children who had at least one reimbursed purchase of ADHD medication compared with the number of children in the total population.

Children were classified into 3 groups depending on their month of birth: the oldest were born in January to April, the middle group in May to August, and the relatively youngest in September to December. Finnish children must attend kindergarten (refers to a preprimary education in Finland) for one year before compulsory basic education begins. The cutoff date for school eligibility is December 31. Children start kindergarten in the calendar year of their sixth birthday and first grade of comprehensive school (basic education) in the year of their seventh birthday. The school year starts in mid-August. Although delayed school entry is possible, this is a relatively uncommon procedure, and the number of children whose school entry is delayed is decreasing. According to Statistics Finland (V. Hämäläinen, personal communication, 2019), 1100 children were held back one year in 2010 (875, 713, and 698 children in 2012, 2013, and 2014, respectively). Furthermore, delayed entry into primary education because of inattention and hyperactivity deficits is rare. Instead, children are expected to receive either special or intensified support at school if needed.

The time of the first ADHD medication purchase was analyzed by a series of sex-stratified Cox proportional hazards regression models by using right censoring.31  The follow-up ended on December 31, 2017. Children born in the first months of the calendar year were set as a reference group. Inspection of log-minus-log survival curves did not indicate a violation of the proportional hazard assumption. For sensitivity analyses, we further assessed whether hazard ratios (HRs) change over time. Regarding use of medication, it is possible that children born in January to April will later move toward their classmates born in September to December, which would result in different HRs depending on the duration of the study.17,23  In a restricted model, the follow-up ended in the year of the children’s eighth birthday. In additional analyses, we also examined incidence rate ratios (IRRs). This relates to the fact that children born in September to December are relatively close in age compared with children born in January to April in the following year. Pairwise comparisons were examined for children born in September to December 2005 and 2006 relative to children born in January to April 2006 and 2007. Analyses were performed by using SPSS version 25 (IBM SPSS Statistics, IBM Corporation) and the computing environment R (ggplot2 package).

The study population included 182 802 Finnish children born in 2005 to 2007 (93 374 boys, 89 428 girls). By the end of follow-up, 7054 children had received ADHD medication at least once (5775 boys, 1279 girls). Cumulative proportions for ADHD medication use in boys and girls born in 2005 to 2007 are shown in Fig 1. Cumulative proportions increased somewhat steadily until age 9 and then started to slow down across both sexes. At the end of follow-up (December 31, 2017), 5.9% to 6.4% of boys and 1.4% to 1.6% of girls had at least 1 reimbursed purchase of ADHD medication.

FIGURE 1

Cumulative proportions of ADHD medication use across cohorts by sex.

FIGURE 1

Cumulative proportions of ADHD medication use across cohorts by sex.

Close modal

In Cox regression analyses, it was shown that the HRs for boys (Table 1) were somewhat stable across cohorts. For boys born in May to August (HRs ∼1.1–1.2) and September to December (HRs ∼1.3), ADHD medication use was more common compared with those born in January to April. Among girls born in either May to August or September to December 2005, the HRs were 1.3 and 1.4, respectively, when compared with those born between January and April. Among girls born in either May to August or September to December 2007, the HRs were 1.6 and 1.7, respectively. After restricting the follow-up to the year of the children’s eighth birthday, the RAE became more pronounced (Tables 1 and 2). Among boys born in May to August and September to December, the HRs were 1.2 to 1.3 and 1.5 to 1.7, respectively, compared with those born in January to April. Among girls, HRs for later-born children (born in May to August and September to December) were 1.7 to 1.9 and 2.0 to 2.1, respectively, compared with their early-born peers (January to April).

TABLE 1

Risk of ADHD Medication Use Among Boys Born 2005–2007 by Birth Month

200520062007
PersonsCases With ADHD Medication, n (%)HR (95% CI)PersonsCases With ADHD Medication, n (%)HR (95% CI)PersonsCases With ADHD Medication, n (%)HR (95% CI)
Total 30 580 1967 (6.4) — 31 414 1941 (6.2) — 31 380 1867 (5.9) — 
Follow-up on December 31, 2017          
 January to April (ref) 9911 529 (5.3) 10 500 572 (5.4) 10 350 548 (5.3) 
 May to August 10 638 705 (6.6) 1.24*** (1.10–1.40) 10 822 665 (6.1) 1.13* (1.01–1.27) 10 927 652 (6.0) 1.13* (1.01–1.27) 
 September to December 10 031 733 (7.3) 1.39*** (1.24–1.54) 10 092 704 (7.0) 1.29*** (1.16–1.45) 10 103 666 (6.6) 1.26***(1.12–1.41) 
Follow-up at the end of the year of eighth birthday          
 January to April (ref) 9911 225 (2.3) 10 500 306 (2.9) 10 350 322 (3.1) 
 May to August 10 638 350 (3.3) 1.27** (1.07–1.50) 10 822 362 (3.3) 1.15 (0.99–1.34) 10 927 422 (3.9) 1.25** (1.08–1.44) 
 September to December 10 031 407 (4.4) 1.67*** (1.42–1.97) 10 092 423 (4.2) 1.45*** (1.25–1.68) 10 103 472 (4.7) 1.51*** (1.31–1.74) 
200520062007
PersonsCases With ADHD Medication, n (%)HR (95% CI)PersonsCases With ADHD Medication, n (%)HR (95% CI)PersonsCases With ADHD Medication, n (%)HR (95% CI)
Total 30 580 1967 (6.4) — 31 414 1941 (6.2) — 31 380 1867 (5.9) — 
Follow-up on December 31, 2017          
 January to April (ref) 9911 529 (5.3) 10 500 572 (5.4) 10 350 548 (5.3) 
 May to August 10 638 705 (6.6) 1.24*** (1.10–1.40) 10 822 665 (6.1) 1.13* (1.01–1.27) 10 927 652 (6.0) 1.13* (1.01–1.27) 
 September to December 10 031 733 (7.3) 1.39*** (1.24–1.54) 10 092 704 (7.0) 1.29*** (1.16–1.45) 10 103 666 (6.6) 1.26***(1.12–1.41) 
Follow-up at the end of the year of eighth birthday          
 January to April (ref) 9911 225 (2.3) 10 500 306 (2.9) 10 350 322 (3.1) 
 May to August 10 638 350 (3.3) 1.27** (1.07–1.50) 10 822 362 (3.3) 1.15 (0.99–1.34) 10 927 422 (3.9) 1.25** (1.08–1.44) 
 September to December 10 031 407 (4.4) 1.67*** (1.42–1.97) 10 092 423 (4.2) 1.45*** (1.25–1.68) 10 103 472 (4.7) 1.51*** (1.31–1.74) 

Cox proportional hazards regression model. Cases include all the children who had ≥1 reimbursed purchase of ADHD medication by the end of follow-up (% = cumulative incidence). CI, confidence interval; ref, reference category; —, not applicable.

***

P ≤ .001

**

P ≤ .01

*

P ≤ .05.

TABLE 2

Risk of ADHD Medication Use Among Girls Born 2005–2007 by Birth Month

200520062007
PersonsNo. Cases With ADHD Medication (%)HR (95% CI)PersonsNo. Cases With ADHD Medication (%)HR (95% CI)PersonsNo. Cases With ADHD Medication (%)HR (95% CI)
Total 29 528 459 (1.6) — 30 060 401 (1.3) — 29 840 419 (1.4) — 
Follow-up at December 31, 2017          
 January to April (ref) 9753 122 (1.3) 10 004 99 (1.0) 9979 98 (1.0) 
 May to August 10 177 165 (1.6) 1.30* (1.03–1.64) 10 306 149 (1.4) 1.47** (1.14–1.89) 10 205 160 (1.6) 1.60*** (1.25–2.06) 
 September to December 9598 172 (1.8) 1.44** (1.14–1.81) 9750 153 (1.6) 1.59*** (1.24–2.05) 9656 161 (1.7) 1.71*** (1.33–2.19) 
Follow-up at the end of year of eighth birthday          
 January to April (ref) 9753 41 (0.4) 10 004 43 (0.4) 9979 55 (0.6) 
 May to August 10 177 72 (0.7) 1.69** (1.15–2.47) 10 306 82 (0.8) 1.85*** (1.28–2.68) 10 205 96 (0.9) 1.71** (1.23–2.38) 
 September to December 9598 81 (0.8) 2.01*** (1.38–2.93) 9750 87 (0.9) 2.08*** (1.44–3.00) 9656 107 (1.1) 2.02*** (1.46–2.79) 
200520062007
PersonsNo. Cases With ADHD Medication (%)HR (95% CI)PersonsNo. Cases With ADHD Medication (%)HR (95% CI)PersonsNo. Cases With ADHD Medication (%)HR (95% CI)
Total 29 528 459 (1.6) — 30 060 401 (1.3) — 29 840 419 (1.4) — 
Follow-up at December 31, 2017          
 January to April (ref) 9753 122 (1.3) 10 004 99 (1.0) 9979 98 (1.0) 
 May to August 10 177 165 (1.6) 1.30* (1.03–1.64) 10 306 149 (1.4) 1.47** (1.14–1.89) 10 205 160 (1.6) 1.60*** (1.25–2.06) 
 September to December 9598 172 (1.8) 1.44** (1.14–1.81) 9750 153 (1.6) 1.59*** (1.24–2.05) 9656 161 (1.7) 1.71*** (1.33–2.19) 
Follow-up at the end of year of eighth birthday          
 January to April (ref) 9753 41 (0.4) 10 004 43 (0.4) 9979 55 (0.6) 
 May to August 10 177 72 (0.7) 1.69** (1.15–2.47) 10 306 82 (0.8) 1.85*** (1.28–2.68) 10 205 96 (0.9) 1.71** (1.23–2.38) 
 September to December 9598 81 (0.8) 2.01*** (1.38–2.93) 9750 87 (0.9) 2.08*** (1.44–3.00) 9656 107 (1.1) 2.02*** (1.46–2.79) 

Cox proportional hazards regression model. Cases include all the children who had ≥1 reimbursed purchase of ADHD medication by the end of follow-up (% = cumulative incidence). CI, confidence interval; ref, reference category; —, not applicable.

***

P ≤ .001

**

P ≤ .01

*

P ≤ .05.

ADHD medication incidence per 1000 persons across 3 cohorts are presented in Tables 3 and 4. Further analyses showed that incidence ratios were significant in the years of primary education entry. After comparing boys (Table 3) born at the end of years 2005 and 2006 (ie, September to December) and who are relatively close in age with children born in the first months of the following calendar years, the IRRs at age 6 (ie, the year of kindergarten entry) were 1.8 and 1.7, whereas the IRRs at age 7 were 1.4. In girls (Table 4), the corresponding analyses yielded significant IRRs at age 7 (IRRs 2.5 and 2.8), when children enter school. IRR estimates were >1 for each age interval. This indicates that there was no catch-up effect among those born in January to April.

TABLE 3

Incidence per 1000 Persons for ADHD Medication Among Boys Born 2005–2007 by Birth Month

Age, yNo. Cases and Incidence (95% CI) January to AprilNo. Cases and Incidence (95% CI) May to AugustNo. Cases and Incidence (95% CI) September to DecemberIRR (95% CI)
2005200620072005200620072005200620072005 September to December versus 2006 January to April2006 September to December versus 2007 January to April
45; 4.5 (3.4–6.1) 69; 6.6 (5.2–8.3) 68; 6.6 (5.2–8.3) 101; 9.5 (7.8–11.5) 74; 6.8 (5.4–8.6) 111; 10.2 (8.4–12.2) 116; 10.6 (8.7–12.8) 112; 11.1 (9.2–13.3) 131; 13.0 (10.9–15.3) 1.76 (1.31–2.38)a 1.69 (1.25–2.29)a 
82; 8.2 (6.7–10.3) 104; 9.9 (8.1–12.0) 117; 11.3 (9.4–13.5) 114; 10.7 (8.9–12.9) 134; 12.4 (10.5–14.7) 156; 14.3 (12.2–16.7) 140; 14.0 (11.8–16.5) 156; 15.5 (13.2–18.1) 172; 17.0 (14.7–19.7) 1.41 (1.09–1.82)a 1.37 (1.08–1.74)a 
98; 9.9 (8.1–12.0) 133; 12.7 (10.7–15.0) 137; 13.2 (11.2–15.6) 135; 12.7 (10.7–15.0) 154; 14.2 (12.2–16.7) 155; 14.2 (12.1–16.6) 151; 15.1 (12.8–17.6) 155; 15.4 (13.1–18.0) 170; 16.8 (14.5–19.5) 1.19 (0.94–1.50) 1.16 (0.92–1.46) 
88; 8.9 (7.2–10.9) 112; 10.7 (8.9–12.8) 124; 12.0 (10.0–14.3) 108; 10.2 (8.4–12.2) 130; 12.0 (10.1–14.3) 159; 14.6 (12.5–17.0) 131; 13.1 (11.0–15.5) 140; 13.9 (11.8–16.4) 172; 17.0 (14.7–19.7) 1.22 (0.95–1.58) 1.16 (0.91–1.48) 
10 88; 8.9 (7.2–10.9) 88; 8.4 (6.8–10.3) 102; 9.9 (8.1–12.0) 103; 9.7 (8.0–11.7) 117; 10.8 (9.0–12.9) 71; 6.5 (5.2–8.2) 101; 10.1 (8.3–12.2) 131; 13.0 (10.9–15.4) 22; 2.2 (1.4–3.3) 1.20 (0.90–1.60) 1.31 (1.02–1.71)a 
11 69; 7.0 (5.5–8.8) 66; 6.3 (4.9–8.0) — 101; 9.5 (7.8–11.5) 56; 5.2 (4.0–6.7) — 77; 7.7 (6.1–9.6) 10; 1.0 (0.5–1.8) — 1.22 (0.88–1.70) — 
12 59; 6.0 (4.6–7.7) — — 43; 4.0 (3.0–5.4) — — 17; 1.7 (1.1–2.7) — — — — 
Age, yNo. Cases and Incidence (95% CI) January to AprilNo. Cases and Incidence (95% CI) May to AugustNo. Cases and Incidence (95% CI) September to DecemberIRR (95% CI)
2005200620072005200620072005200620072005 September to December versus 2006 January to April2006 September to December versus 2007 January to April
45; 4.5 (3.4–6.1) 69; 6.6 (5.2–8.3) 68; 6.6 (5.2–8.3) 101; 9.5 (7.8–11.5) 74; 6.8 (5.4–8.6) 111; 10.2 (8.4–12.2) 116; 10.6 (8.7–12.8) 112; 11.1 (9.2–13.3) 131; 13.0 (10.9–15.3) 1.76 (1.31–2.38)a 1.69 (1.25–2.29)a 
82; 8.2 (6.7–10.3) 104; 9.9 (8.1–12.0) 117; 11.3 (9.4–13.5) 114; 10.7 (8.9–12.9) 134; 12.4 (10.5–14.7) 156; 14.3 (12.2–16.7) 140; 14.0 (11.8–16.5) 156; 15.5 (13.2–18.1) 172; 17.0 (14.7–19.7) 1.41 (1.09–1.82)a 1.37 (1.08–1.74)a 
98; 9.9 (8.1–12.0) 133; 12.7 (10.7–15.0) 137; 13.2 (11.2–15.6) 135; 12.7 (10.7–15.0) 154; 14.2 (12.2–16.7) 155; 14.2 (12.1–16.6) 151; 15.1 (12.8–17.6) 155; 15.4 (13.1–18.0) 170; 16.8 (14.5–19.5) 1.19 (0.94–1.50) 1.16 (0.92–1.46) 
88; 8.9 (7.2–10.9) 112; 10.7 (8.9–12.8) 124; 12.0 (10.0–14.3) 108; 10.2 (8.4–12.2) 130; 12.0 (10.1–14.3) 159; 14.6 (12.5–17.0) 131; 13.1 (11.0–15.5) 140; 13.9 (11.8–16.4) 172; 17.0 (14.7–19.7) 1.22 (0.95–1.58) 1.16 (0.91–1.48) 
10 88; 8.9 (7.2–10.9) 88; 8.4 (6.8–10.3) 102; 9.9 (8.1–12.0) 103; 9.7 (8.0–11.7) 117; 10.8 (9.0–12.9) 71; 6.5 (5.2–8.2) 101; 10.1 (8.3–12.2) 131; 13.0 (10.9–15.4) 22; 2.2 (1.4–3.3) 1.20 (0.90–1.60) 1.31 (1.02–1.71)a 
11 69; 7.0 (5.5–8.8) 66; 6.3 (4.9–8.0) — 101; 9.5 (7.8–11.5) 56; 5.2 (4.0–6.7) — 77; 7.7 (6.1–9.6) 10; 1.0 (0.5–1.8) — 1.22 (0.88–1.70) — 
12 59; 6.0 (4.6–7.7) — — 43; 4.0 (3.0–5.4) — — 17; 1.7 (1.1–2.7) — — — — 

IRRs are between boys born in the last third of the year (September to December) and boys born in the first third (January to April) of the following year. CI, confidence interval; —, not applicable.

a

IRR is significant at P < .05.

TABLE 4

Incidence per 1000 Persons for ADHD Medication Among Girls Born 2005–2007 by Birth Month

Age, yNo. Cases and Incidence (95% CI) January to AprilNo. Cases and Incidence (95% CI) May to AugustNo. Cases and Incidence (95% CI) September to DecemberIRR (95% CI)
2005200620072005200620072005200620072005 September to December versus 2006 January to April2006 September to December versus 2007 January to April
6; 0.6 (0.2–1.3) 11; 1.1 (0.6–1.9) 12; 1.2 (0.7–2.0) 18; 1.8 (1.1–2.7) 19; 1.8 (1.1–2.8) 23; 2.3 (1.5–3.3) 14; 1.5 (0.8–2.4) 14; 1.4 (0.8–2.4) 26; 2.7 (1.8–3.9) 1.33 (0.60–3.01) 1.19 (0.55–2.64) 
18; 1.8 (1.1–2.9) 11; 1.1 (0.6–1.9) 14; 1.4 (0.8–2.3) 23; 2.3 (1.5–3.3) 27; 2.6 (1.8–3.8) 31; 3.0 (2.1–4.3) 26; 2.7 (1.8–3.9) 38; 3.9 (2.8–5.3) 45; 4.7 (3.4–6.2) 2.46 (1.23–5.19)a 2.78 (1.53–5.29)a 
17; 1.7 (1.0–2.7) 21; 2.1 (1.3–3.2) 29; 2.9 (2.0–4.1) 31; 3.0 (2.1–4.3) 36; 3.5 (2.5–4.8) 42; 4.1 (3.0–5.5) 41; 4.3 (3.1–5.7) 35; 3.6 (2.5–4.9) 36; 3.7 (2.7–5.1) 2.04 (1.21–3.50)a 1.24 (0.75–2.04) 
26; 2.7 (1.8–3.9) 22; 2.2 (1.3–3.2) 25; 2.5 (1.7–3.6) 34; 3.3 (2.4–4.6) 33; 3.2 (2.2–4.4) 45; 4.4 (3.3– 5.8) 31; 3.2 (2.2–4.5) 36; 3.7 (2.6–5.1) 47; 4.9 (3.6–6.4) 1.47 (0.85–2.57) 1.47 (0.89–2.48) 
10 20; 2.1 (1.3–3.1) 19; 1.9 (1.2–2.9) 18; 1.8 (1.1–2.8) 17; 1.7 (1.0–2.6) 25; 2.4 (1.6–3.5) 19; 1.9 (1.2–2.9) 26; 2.7 (1.8–3.9) 24; 2.5 (1.6–3.6) 7; 0.7 (0.3–1.4) 1.43 (0.79–2.62) 1.37 (0.74–2.55) 
11 17; 1.8 (1.0–2.7) 15; 1.5 (0.9–2.4) — 28; 2.8 (1.9–3.9) 9; 0.9 (0.4–1.6) — 30; 3.1 (2.1–4.4) 6; 0.6 (0.2–1.3) — 2.09 (1.13–3.97)a — 
12 18; 1.9 (1.1–2.9) — — 14; 1.4 (0.8–2.3) — — 4; 0.4 (0.1–1.0) — — — — 
Age, yNo. Cases and Incidence (95% CI) January to AprilNo. Cases and Incidence (95% CI) May to AugustNo. Cases and Incidence (95% CI) September to DecemberIRR (95% CI)
2005200620072005200620072005200620072005 September to December versus 2006 January to April2006 September to December versus 2007 January to April
6; 0.6 (0.2–1.3) 11; 1.1 (0.6–1.9) 12; 1.2 (0.7–2.0) 18; 1.8 (1.1–2.7) 19; 1.8 (1.1–2.8) 23; 2.3 (1.5–3.3) 14; 1.5 (0.8–2.4) 14; 1.4 (0.8–2.4) 26; 2.7 (1.8–3.9) 1.33 (0.60–3.01) 1.19 (0.55–2.64) 
18; 1.8 (1.1–2.9) 11; 1.1 (0.6–1.9) 14; 1.4 (0.8–2.3) 23; 2.3 (1.5–3.3) 27; 2.6 (1.8–3.8) 31; 3.0 (2.1–4.3) 26; 2.7 (1.8–3.9) 38; 3.9 (2.8–5.3) 45; 4.7 (3.4–6.2) 2.46 (1.23–5.19)a 2.78 (1.53–5.29)a 
17; 1.7 (1.0–2.7) 21; 2.1 (1.3–3.2) 29; 2.9 (2.0–4.1) 31; 3.0 (2.1–4.3) 36; 3.5 (2.5–4.8) 42; 4.1 (3.0–5.5) 41; 4.3 (3.1–5.7) 35; 3.6 (2.5–4.9) 36; 3.7 (2.7–5.1) 2.04 (1.21–3.50)a 1.24 (0.75–2.04) 
26; 2.7 (1.8–3.9) 22; 2.2 (1.3–3.2) 25; 2.5 (1.7–3.6) 34; 3.3 (2.4–4.6) 33; 3.2 (2.2–4.4) 45; 4.4 (3.3– 5.8) 31; 3.2 (2.2–4.5) 36; 3.7 (2.6–5.1) 47; 4.9 (3.6–6.4) 1.47 (0.85–2.57) 1.47 (0.89–2.48) 
10 20; 2.1 (1.3–3.1) 19; 1.9 (1.2–2.9) 18; 1.8 (1.1–2.8) 17; 1.7 (1.0–2.6) 25; 2.4 (1.6–3.5) 19; 1.9 (1.2–2.9) 26; 2.7 (1.8–3.9) 24; 2.5 (1.6–3.6) 7; 0.7 (0.3–1.4) 1.43 (0.79–2.62) 1.37 (0.74–2.55) 
11 17; 1.8 (1.0–2.7) 15; 1.5 (0.9–2.4) — 28; 2.8 (1.9–3.9) 9; 0.9 (0.4–1.6) — 30; 3.1 (2.1–4.4) 6; 0.6 (0.2–1.3) — 2.09 (1.13–3.97)a — 
12 18; 1.9 (1.1–2.9) — — 14; 1.4 (0.8–2.3) — — 4; 0.4 (0.1–1.0) — — — — 

IRR are between girls born in the last third of the year (September to December) and girls born in the first third (January to April; reference category) of the following year. CI, confidence interval; —, not applicable.

a

IRR is significant at P < .05.

With this study, we provide up-to-date evidence for the association between children’s birth month and ADHD from a single jurisdiction. In line with our hypothesis, our findings reveal an association among children between younger relative age in the school year and receiving medication for ADHD. The findings further reveal that differences in maturity, which are stronger at younger ages, moderate the association between children’s birth month and ADHD medication use because the RAE was stronger across sexes in the years of school entry. We also observed that the RAE in medication for ADHD seemed more pronounced in the female population than the male population. Importantly, the comparison of 3 consecutive cohorts indicated that the association between ADHD medication use and relative age was stable among the male population but showed signs of increase among the female population.

Our findings are consistent with earlier studies on the influence of relative age on the increased likelihood of receiving ADHD medication.6,7  The observed RAE among children (born between 2005 and 2007) in our study also corroborates the findings from the Finnish study that revealed greater cumulative incidence of clinically diagnosed ADHD among relatively younger children within the Finnish male and female populations born between 1991 and 2004.13  Most importantly, along with the aforementioned study, our findings increase the understanding of the RAE in ADHD by providing a comprehensive picture of the association within a single country. Although researchers of previous studies have suggested relatively low prescribing rates for ADHD in Finland,24  our results indicated that the prevalence of ADHD medication use has increased somewhat rapidly among Finnish children. In our study, we observed that 7.3% of boys born in September to December in 2005 had received at least one reimbursed ADHD medication purchase by the end of follow-up, whereas in girls, the comparable figure was 1.8%.

An important discovery that emerged was that the RAE in ADHD among Finnish children is particularly prominent in the early school years. This is in line with the results of a Swedish study,17  whereas a Norwegian study concluded that the RAE increased from grade 3 onwards.18  The finding that the relatively youngest children in classrooms are more likely to be prescribed medication for ADHD than their relatively older peers in the years of school entry underscores the notion that school entry regulations may play an important role.7  For example, children in Finland and Denmark start first grade in the calendar year of their seventh birthday, and the cutoff date for school eligibility in both countries is December 31. However, in findings from the Danish study, it is shown that although the use of medication for Danish children with ADHD had increased during the years 2000 to 2012, the RAE on ADHD medication use decreased and further vanished during the aforementioned study years.20  This may stem from the fact that ∼40% of children born in October to December are held back one year in the Danish school system.2022  In Finland, delayed school entry is rare and ever decreasing. Meanwhile, the prevalence of ADHD medication use has increased steadily. Therefore, it is possible that in countries with an RAE, the more immature behavior of younger children within the school year is treated as ADHD.7,11  In line with this notion, teacher ratings of whether children display clinically significant ADHD symptomology are more strongly associated with relative immaturity when compared with parent ratings.15,17  In addition to school entry regulations, we speculate that the increased availability and improved use of school contingency management could also protect from the RAE among children with ADHD susceptibility.27,28 

Our findings indicate a large male-to-female discrepancy in ADHD medication use among the Finnish child population. This was also addressed in a recent retrospective observational study in which researchers examined trends in ADHD medication use among children in 13 countries.24  Furthermore, our study findings extend previous research by showing that the association between relative age and ADHD medication use is more prominent among the female population.7  This finding is in line with studies from Canada16  and Norway.18  In fact, the observed HRs in the male and female populations are similar to those of a Canadian study.16  In comparison, researchers of studies from Iceland,14  Denmark,21  and Sweden17  did not discover sex-based differences. Regarding sex differences, our findings indicate that the prescribing rates for ADHD medication among boys born toward the end of the year are close to the upper threshold of pooled estimates of ADHD prevalence.5  Consequently, there is less room for an increase in ADHD medication use in boys compared with girls. This might explain why the RAE appeared to remain stable in boys born in 2005 to 2007.

Additionally, because of a significant sex discrepancy in ADHD medication use, the average prevalence ratio among girls is well below the population average. We speculate that over the long-term, as the prevalence ratio among girls moves closer to the population average, the RAE may weaken.32  Another possible explanation is that, at present, different manifestations of ADHD may go unnoticed in the female population in Finland, and those with ADHD medication are perhaps more likely to suffer from co-occurring disorders, such as oppositional defiant disorder. Recent studies on sex differences indicate that emotional and behavioral difficulties are stronger predictors of ADHD diagnosis in girls when compared with boys.29,30  There is also evidence that emotional and behavioral difficulties are also in greater contrast to what is perceived as normative functioning in girls compared with boys.30  Interestingly, researchers of one study discovered that delayed school entry improved, in particular, social functioning among girls in early school years.21  Nevertheless, there is a need for studies in which researchers examine the extent to which relative age is associated with long-term outcomes and social functioning.

The main strength of our study is the population-based design using prescription register data that cover all reimbursed outpatient ADHD medication purchases in children. However, it is important to acknowledge that with our register-based data, we could not rule out whether the increased risk for ADHD medication use among relatively younger children relates to the misidentification of ADHD or perhaps to the fact that relative immaturity aggravates ADHD traits. Nevertheless, the empirical findings support the idea that school entry plays an important role. Further studies with better used information from various national registers and adjustment for confounding factors are needed to examine the extent to which the RAE is associated with medication use and diagnoses and whether ADHD presentations moderate the RAE. Importantly, as the Finnish Current Care Guideline of ADHD enhances the notion that it is pivotal to assess children’s symptomology relative to age, clinicians have become more aware of the RAE in ADHD. Therefore, it is important to replicate our analyses in children born after 2007 in the near future to examine the extent to which increased awareness is associated with standards in prescribing for children with ADHD.

With our study, we provide evidence that children born in May to August and September to December (ie, relatively younger children in a school class) are more likely to receive ADHD medication in Finland compared with children born in January to April. The risk of a modest RAE was stable among boys but appeared to be increasing among girls. These findings may stem from the fact that ADHD medication use among boys has become common in Finland, whereas ADHD medication use among girls is still relatively low. The RAE in ADHD medication use was more pronounced in the years when children enter school. Therefore, there is a need to consider school entry regulations and, particularly, increase the availability of school contingency management programs to reduce the RAE. Importantly, there is a need for prospective studies in which researchers examine whether relative age is associated with youth and adult outcomes in children with ADHD susceptibility.

Mr Vuori conceptualized and designed the study, coordinated data collection, conducted the initial analyses, drafted the initial manuscript, and reviewed and revised the manuscript; Ms Koski-Pirilä conceptualized and designed the study, collected data, and reviewed and revised the manuscript; Ms Martikainen, Ms Saastamoinen, Profs Sourander and Aronen, and Drs Puustjärvi and Chudal conceptualized and designed the study and critically reviewed the manuscript for important intellectual content; and all authors approved the final manuscript as submitted and agree to be accountable for all aspects of the work.

Deidentified individual participant data will not be made available.

FUNDING: No external funding.

     
  • ADHD

    attention-deficit/hyperactivity disorder

  •  
  • HR

    hazard ratio

  •  
  • IRR

    incidence rate ratio

  •  
  • RAE

    relative age effect

1
Faraone
SV
,
Asherson
P
,
Banaschewski
T
, et al
.
Attention-deficit/hyperactivity disorder
.
Nat Rev Dis Primers
.
2015
;
1
:
15020
2
Sayal
K
,
Prasad
V
,
Daley
D
,
Ford
T
,
Coghill
D
.
ADHD in children and young people: prevalence, care pathways, and service provision
.
Lancet Psychiatry
.
2018
;
5
(
2
):
175
186
3
Wolraich
ML
,
Chan
E
,
Froehlich
T
, et al
.
ADHD diagnosis and treatment guidelines: a historical perspective
.
Pediatrics
.
2019
;
144
(
4
):
e20191682
4
Franke
B
,
Michelini
G
,
Asherson
P
, et al
.
Live fast, die young? A review on the developmental trajectories of ADHD across the lifespan
.
Eur Neuropsychopharmacol
.
2018
;
28
(
10
):
1059
1088
5
Thomas
R
,
Sanders
S
,
Doust
J
,
Beller
E
,
Glasziou
P
.
Prevalence of attention-deficit/hyperactivity disorder: a systematic review and meta-analysis
.
Pediatrics
.
2015
;
135
(
4
).
6
Holland
J
,
Sayal
K
.
Relative age and ADHD symptoms, diagnosis and medication: a systematic review
.
Eur Child Adolesc Psychiatry
.
2019
;
28
(
11
):
1417
1429
7
Whitely
M
,
Raven
M
,
Timimi
S
, et al
.
Attention deficit hyperactivity disorder late birthdate effect common in both high and low prescribing international jurisdictions: a systematic review
.
J Child Psychol Psychiatry
.
2019
;
60
(
4
):
380
391
8
Root
A
,
Brown
JP
,
Forbes
HJ
, et al
.
Association of relative age in the school year with diagnosis of intellectual disability, attention-deficit/hyperactivity disorder, and depression
.
JAMA Pediatr
.
2019
;
173
(
11
):
1068
1075
9
Goodman
R
,
Gledhill
J
,
Ford
T
.
Child psychiatric disorder and relative age within school year: cross sectional survey of large population sample
.
BMJ
.
2003
;
327
(
7413
):
472
10
Brikell
I
,
Kuja-Halkola
R
,
Larsson
JO
, et al
.
Relative immaturity in childhood and attention-deficit/hyperactivity disorder symptoms from childhood to early adulthood: exploring genetic and environmental overlap across development
.
J Am Acad Child Adolesc Psychiatry
.
2016
;
55
(
10
):
886
895
11
Evans
WN
,
Morrill
MS
,
Parente
ST
.
Measuring inappropriate medical diagnosis and treatment in survey data: the case of ADHD among school-age children
.
J Health Econ
.
2010
;
29
(
5
):
657
673
12
Merten
EC
,
Cwik
JC
,
Margraf
J
,
Schneider
S
.
Overdiagnosis of mental disorders in children and adolescents (in developed countries)
.
Child Adolesc Psychiatry Ment Health
.
2017
;
11
:
5
13
Sayal
K
,
Chudal
R
,
Hinkka-Yli-Salomäki
S
,
Joelsson
P
,
Sourander
A
.
Relative age within the school year and diagnosis of attention-deficit hyperactivity disorder: a nationwide population-based study
.
Lancet Psychiatry
.
2017
;
4
(
11
):
868
875
14
Zoëga
H
,
Valdimarsdóttir
UA
,
Hernández-Díaz
S
.
Age, academic performance, and stimulant prescribing for ADHD: a nationwide cohort study
.
Pediatrics
.
2012
;
130
(
6
):
1012
1018
15
Elder
TE
.
The importance of relative standards in ADHD diagnoses: evidence based on exact birth dates
.
J Health Econ
.
2010
;
29
(
5
):
641
656
16
Morrow
RL
,
Garland
EJ
,
Wright
JM
,
Maclure
M
,
Taylor
S
,
Dormuth
CR
.
Influence of relative age on diagnosis and treatment of attention-deficit/hyperactivity disorder in children
.
CMAJ
.
2012
;
184
(
7
):
755
762
17
Halldner
L
,
Tillander
A
,
Lundholm
C
, et al
.
Relative immaturity and ADHD: findings from nationwide registers, parent- and self-reports
.
J Child Psychol Psychiatry
.
2014
;
55
(
8
):
897
904
18
Karlstad
Ø
,
Furu
K
,
Stoltenberg
C
,
Håberg
SE
,
Bakken
IJ
.
ADHD treatment and diagnosis in relation to children’s birth month: nationwide cohort study from Norway
.
Scand J Public Health
.
2017
;
45
(
4
):
343
349
19
Chen
MH
,
Lan
WH
,
Bai
YM
, et al
.
Influence of relative age on diagnosis and treatment of attention-deficit hyperactivity disorder in Taiwanese children
.
J Pediatr
.
2016
;
172
:
162
167.e1
20
Pottegård
A
,
Hallas
J
,
Hernández-Díaz
,
Zoëga
H
.
Children’s relative age in class and use of medication for ADHD: a Danish nationwide study
.
J Child Psychol Psychiatry
.
2014
;
55
(
11
):
1244
1250
21
Dee
TS
,
Sievertsen
HH
.
The gift of time? School starting age and mental health
.
Health Econ
.
2018
;
27
(
5
):
781
802
22
Dalsgaard
S
,
Humlum
MK
,
Nielsen
HS
,
Simonsen
M
.
Common Danish standards in prescribing medication for children and adolescents with ADHD
.
Eur Child Adolesc Psychiatry
.
2014
;
23
(
9
):
841
844
23
Hernán
MA
.
The hazards of hazard ratios
.
Epidemiology
.
2010
;
21
(
1
):
13
15
24
Raman
SR
,
Man
KKC
,
Bahmanyar
S
, et al
.
Trends in attention-deficit hyperactivity disorder medication use: a retrospective observational study using population-based databases
.
Lancet Psychiatry
.
2018
;
5
(
10
):
824
835
25
Furu
K
,
Karlstad
Ø
,
Zoega
H
, et al
.
Utilization of stimulants and atomoxetine for attention-deficit/hyperactivity disorder among 5.4 million children using population-based longitudinal data
.
Basic Clin Pharmacol Toxicol
.
2017
;
120
(
4
):
373
379
26
Vuori
M
,
Koski-Pirilä
A
,
Martikainen
JE
,
Saastamoinen
L
.
Gender- and age-stratified analyses of ADHD medication use in children and adolescents in Finland using population-based longitudinal data, 2008–2018
.
Scand J Public Health
.
2020
;
48
(
3
):
303
307
27
Faraone
SV
,
Rostain
AL
,
Blader
J
, et al
.
Practitioner review: emotional dysregulation in attention-deficit/hyperactivity disorder - implications for clinical recognition and intervention
.
J Child Psychol Psychiatry
.
2019
;
60
(
2
):
133
150
28
Daley
D
,
Van Der Oord
S
,
Ferrin
M
, et al
.
Practitioner review: current best practice in the use of parent training and other behavioural interventions in the treatment of children and adolescents with attention deficit hyperactivity disorder
.
J Child Psychol Psychiatry
.
2018
;
59
(
9
):
932
947
29
Ottosen
C
,
Larsen
JT
,
Faraone
SV
, et al
.
Sex differences in comorbidity patterns of attention-deficit/hyperactivity disorder
.
J Am Acad Child Adolesc Psychiatry
.
2019
;
58
(
4
):
412
422.e3
30
Mowlem
FD
,
Rosenqvist
MA
,
Martin
J
,
Lichtenstein
P
,
Asherson
P
,
Larsson
H
.
Sex differences in predicting ADHD clinical diagnosis and pharmacological treatment
.
Eur Child Adolesc Psychiatry
.
2019
;
28
(
4
):
481
489
31
Tierney
JF
,
Stewart
LA
,
Ghersi
D
,
Burdett
S
,
Sydes
MR
.
Practical methods for incorporating summary time-to-event data into meta-analysis
.
Trials
.
2007
;
8
:
16
32
Barnett
AG
,
van der Pols
JC
,
Dobson
AJ
.
Regression to the mean: what it is and how to deal with it
.
Int J Epidemiol
.
2005
;
34
(
1
):
215
220

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.