Table 1.
Clinical and general population epidemiologic studies evaluating the pediatric obesity-headache association
| Author (y) | Study design (database) | Sample population | Headache sample size | Sex | Race | Mean age yrs (range) | HA Dx | BMI (SR/M) | Findings | Notes |
|---|---|---|---|---|---|---|---|---|---|---|
| [31] | Pro, CS, Multicenter Clin Trial | Source: Obesity and Gen Ped Clinics in Israel Size: 273 BMI≥5th and <85th% : 116 BMI≥85th - <95th%: 45 BMI≥95th%: 112 |
Total HA participants (Any HA Type): 39 Total EM: 15 Total CM: 7 Total TTH: 13 Total CTTH: 2 |
61 % F Stratified | Hebrew-speaking children (specifics NR.) | 13.3 (9–17) | ICHD-2 | M-BMI | The OR of headaches in children 9–17 yrs of age was nearly a 4-fold increase in girls with BMI≥95th% (OR 3.93; CI: 1.28–12.1) compared with normal weighted girls. There was no reported comparison of odds of headache in obese boys to normal weighted boys in this dataset; the odds of headache in boys with BMI≥95th% (OR 2.28; CI: 0.58–8.91) was reported in relationship to normal weighted girls. The odds of EM in those across BMI categories were not reported. EM was reported in 8.9 % (10/112) of those with BMI≥95th%, 4.4 % (2/45) of those with BMI≥85th% and <95th%; and 2.5 % (3/116) of those with BMI <85th, (P value not reported). |
Excluded all with known chronic diseases and eating disorders. Excluded those receiving any medications. |
| [32•] | Retro, Long, Multicenter, Clin Trial compared with historically reported GP data | Source: 7 Ped HA Clinics in the US and historical NHANES data Size: initial: 913 3 mo: 213 6 mo: 174 BMI≤5th%: 20 (1) BMI≥5th – <85th%: 580 BMI≥85th <95th%: 152 BMI≥95th%: 160 |
Total HA participants (Any HA Type): 913 (2) Total EM: MO: 645 MA: 146 Total p-M: 40 Total TTH: 38 Total CDH: 212 CM: 204 CTTH: 8 |
60 % F | White 85 % Black 13 % Asian 1 % Hisp <1 % |
11.9 (3–18) | ICHD-2 | M-BMI | Prevalence of having a BMI≥95th% (17.5%) in pediatric headache patients was similar to the historical prevalence rates of those reported in the NHANES gen pop study (17.1 %), P>0.05. BMI percentile was significantly correlated to HA frequency (P=0.003) and disability by PedMIDAS score (P=0.02), such that as BMI percentile increased so did HA frequency and disability score. At 3 mos, unadjusted changes in BMI in those with BMI≥85th percentile, positively correlated to reduction of HA frequency (r=0.32, P=0.01). |
Obesity prevalence in HA patients were compared with historically reported obesity prevalence rates in general population data from NHANES. Patients were allowed to use preventive medications. |
| [38] | Retro, CS, Clin Trial | Source: Ped Neuro Clinic in Turkey Size: 124 rel BMI <110 (3): 80 rel BMI≥110 and <120: 20 rel BMI≥120: 22 |
Total HA participants (All EM): 124 EM without aura: 88 EM with aura: 36 |
62 % F | NR | 12.9 (4–17) | ICHD-2 | M-rel BMI | 17.7 % of children with EM had rel BMI≥120. Of children with EM, those with a rel BMI≥120 reported more frequent HA/mo (5.3±2.6) compared with those with a rel BMI≥110 and ±120 (4.4±2.4 HA/mo) and those with a rel BMI ±110, (3.6±2.2 HA/mo), P=0.018. HA severity and duration was not significantly different across BMI categories. |
Excluded those with transformed migraine and tension-type headache. Excluded those with systemic disease and other chronic medical problems. No participants were taking medications. |
| [33••] | Pro, CS, GP Trial of adolescents in school | Source: Junior High and High Schools in Norway Total: 5588 Overweight and obese (4) : 891 Male: 421 Female: 470 |
Total HA participants (Any HA Type): 1591 M: 554 F: 1047 Total Mig: 392 M: 126 F: 266 Total TTH: 950 M: 320 F: 630 |
52 % F Stratified | White 98 % Other 2 % |
NR (13–18) | ICHD-2 | M-BMI | The OR of recurrent HA were 40 % greater in those whose BMI was above the overweight cutoff (OR 1.4, CI: 1.2–1.6, P<0.0001), and was seen independently in boys (OR 1.4, CI: 1.1–1.8, P=0.01) and girls (OR 1.4, CI: 1.2–1.8, P<0.0001). The OR of migraine were 60 % greater in those whose BMI was above overweight cutoff (OR 1.6, CI: 1.4–2.2), P<0.0001] compared with those whose BMI was below overweight cutoff. The OR of TTH were 40 % greater in those whose BMI was above overweight cutoff (OR 1.4, CI: 1.1–1.6, P<0.0001) compared with those whose BMI was below overweight cutoff. |
Participants not reported to be excluded based on medication use. HA frequency NR. |
| [35] | Retro, CS, Clin Trial compared with historical GP norms from NHANES | Source: Ped HA Clinic in Columbus, OH Size: 925 BMI≥5th – <85th%: 610 BMI≥85th – <95th%: 130 BMI≥95th%: 185 |
Total HA participants (any type): 925 Total EM and p-EM: 532 Total CM% p-CM w/o Rx overuse: 79 Total TTH and p-TTH: 141 Total CTTH and p-CTTH w/o Rx overuse: 73 Total CDH: 252 w/ Rx overuse: 100 w/o Rx overuse: 152 |
58 % F | NR | 12.5 (5–17) | ICHD-2 | M-BMI | A total of 22 % (95 % CI: 17.1–27.4) of CDH pts had BMI≥95th% compared with unidentified historical norms (16.3 %). A total of 26 %(95 % CI: 16.5–37.6) of those with chronic or p-CTTH w/o MOH had BMI≥95th% compared with unidentified historical norms (16 %). Population estimates of 16.3 % with BMI >95th% and 31.9 % with BMI >85th% compared with CDC growth chart norms (which were based on data from 1960s – 1990s) included children as young as 2 yrs. When re-calculations are done excluding children age 2–5 yrs making the gen pop more representative of study population then percentiles for BMI >95th% and >85th% become 17.4 % and 33.8 %, respectively. This would nullify the 2 above findings that were noted to be significant |
Historical Norms came from NHANES, which included children age 2–19 yrs while clinical population was age 5–17 yrs. |
| [37] | Pro, Long, GP Trial of middle school students | Source: Taitung, Taiwan Middle Schools Size: 3342 BMI <95th% (5): 2617 BMI≥95th%: 517 |
Total EM or p-EM participants at onset: 820 Total incidence CDH: 63 CM: 37 CTTH: 22 Unknown: 4 |
49 % F | NR | 13.2 (13–14) | Mod-ICHD-2 | SR-BMI | Association between EM and obesity NR. Risk of incident CDH was not increased in those with BMI≥95th% (RR 1.58, CI: 0.89–2.80) compared with those with BMI <95 %. However, incident CDH was higher in those with BMI≥95th% vs. those with BMI <95 % (RR 1.96, CI: 1.04–3.69) after adjusting for covariates. Risk of incident CM was 2.5 X higher in those with BMI≥95 % compared with those with BMI <95 % (RR 2.43, CI: 1.23–4.80). After adjusting for covariates, risk of CM did not change. Risk of incident CTTH was not increased in those with BMI≥95 % compared with those with BMI <95 % (RR 0.5, CI: 0.12–2.16). |
Included episodic headache participants in non-CDH groups. Excluded those who began study with CDH. |
| [36] | Retro, CS, Clin Trial | Source: Ped Neuro Clinic in Israel Size: 181 BMI≥5th – <85th%: 109 BMI≥85th – <95th%: 48 BMI≥95th%: 24 |
Total HA participants (Any Type): 181 Total EM/CM and p-EM/p-CM combined: 81 Total TTH/CTTH and p-TTH/p-CTTH: 87 Total Unknown: 13 |
56 % F | NR | 10.1 (4–18) | ICHD-2 | M-BMI | Crude prevalence: 60 % of those with BMI≥85th– <95th% and 62.5 % of those with BMI≥95th% had migraines (EM/CM) compared with 34 % of those with BMI≥ 5th% – <85th%, (P=0 .01). The odds of definitive or probable EM/CM in general was over 2 fold greater in those with BMI≥ 85th – <95th% (OR 2.4, 95th% CI: 1.2–4.7) compared with those with BMI≥5th% – <85th%. In girls it was 3–5 fold greater if they were overweight or obese (overweight: OR 3.01, 95th CI: 1.24–7.3; obese: OR 4.93, 95th% CI: 1.46–8.61). CDH reported to be more prevalent in those with BMI≥85th% (23 %) compared with those with BMI ≥5th% – <85th% participants (12 %, P<.01). |
Developmental delay seizures, psychiatric disorders, hydrocephalus, or systemic disease excluded. Did not separate episodic from chronic fully |
Notes:
1.# participants in categories calculated based on % reported.
2.Numbers for EM, p-M, TTH, and CDH were calculated from reported percentages of initial participants of 913.
3.Rel BMI calculated with following formula: (participants BMI) × 100/(50th percentile BMI for participant's age and sex).
4.Based on International Obesity Task Force, 2000, Cole TJ et al. establishing a standard definition for child overweight and obesity worldwide: International survey. BMJ 2000;320:1240–43. Overweight is defined as the BMI with respect to age and sex that correlates to BMI of 25 at 18 y (adult cutoff in widest use for overweight). Obesity is defined as the BMI with respect to age and sex that correlates to BMI of 30 at 18-y-old (adult cut-off in widest use for obesity).
5.Based on Taiwan Department of Health, Chen W, Chang MH. New growth charts for Taiwanese children and adolescents based on World Health Organization standards and health-related physical fitness. Pediatr Neonatal. 2010;51:69–79.
BMI body mass index, CDH chronic daily headache, CI confidence interval, Clin clinical, CM chronic migraine, CS cross-sectional, CTTH chronic tension type headache, Dx diagnosis, EM episodic migraine, F female, Gen general, GP general population, HA headache, Hisp Hispanic, ICHD-2 international classification headache disorders 2, Long longitudinal, M measured, MA migraine with aura, Mig migraine, MO migraine without aura, Mod modified, Neuro neurology, NR not reported, OR odds ratio P probable, Ped pediatric, Pro prospective, Rel relative, Retro retrospective, SR self-reported, TTH tension-type headache, US United States