Table 2.
Factors associated with an ADHD diagnosis (results of two-level multivariable logistic regression analysis with approximately six million children and adolescents (level 1) residing in 402 districts (level 2). *The models adjusted for all variables in the table. The MOR and PCV were calculated using the formulas proposed by Merlo et al.18.
| Variables | Total sample (n = 402 districts) | Rural areas with a low population density (n = 96 districts) | Rural areas with population concentrations (n = 103 districts) | Urban districts (n = 136 districts) | Big urban municipalities (n = 66 districts) |
|---|---|---|---|---|---|
| Fixed effects at the individual level | AOR & 95% CI | AOR & 95% CI | AOR & 95% CI | AOR & 95% CI | AOR & 95% CI |
| Sex: boys vs. girls | 3.10 (3.07–3.13) | 3.09 (3.02–3.16) | 3.15 (3.08–3.22) | 3.11 (3.06–3.15) | 3.06 (3.01–3.12) |
| Age | |||||
| 6 vs. 5 years | 1.70 (1.65–1.76) | 1.65 (1.53–1.78) | 1.67 (1.55–1.80) | 1.75 (1.66–1.84) | 1.69 (1.59–1.80) |
| 7 vs. 5 years | 3.04 (2.95–3.13) | 2.73 (2.54–2.93) | 2.86 (2.66–3.06) | 3.23 (3.09–3.39) | 3.08 (2.91–3.26) |
| 8 vs. 5 years | 4.61 (4.48–4.74) | 3.97 (3.71–4.25) | 4.41 (4.13–4.72) | 4.92 (4.70–5.15) | 4.71 (4.46–4.97) |
| 9 vs. 5 years | 5.67 (5.52–5.83) | 4.84 (4.53–5.18) | 5.61 (5.26–5.99) | 6.03 (5.77–6.31) | 5.75 (5.46–6.06) |
| 10 vs. 5 years | 6.15 (5.98–6.32) | 5.35 (5.01–5.72) | 6.00 (5.62–6.40) | 6.54 (6.26–6.84) | 6.22 (5.91–6.56) |
| 11 vs. 5 years | 6.21 (6.04–6.38) | 5.32 (4.98–5.69) | 6.15 (5.76–6.57) | 6.64 (6.35–6.94) | 6.22 (5.90–6.56) |
| 12 vs. 5 years | 6.08 (5.91–6.24) | 5.26 (4.92–5.61) | 6.18 (5.79–6.59) | 6.45 (6.17–6.75) | 6.01 (5.70–6.34) |
| 13 vs. 5 years | 5.86 (5.70–6.02) | 5.08 (4.75–5.42) | 5.91 (5.53–6.31) | 6.23 (5.96–6.51) | 5.80 (5.51–6.13) |
| 14 vs. 5 years | 5.32 (5.18–5.47) | 4.55 (4.26–4.87) | 5.54 (5.19–5.91) | 5.57 (5.32–5.83) | 5.34 (5.06–5.63) |
| Fixed effects at the district level | |||||
| Place of residence | |||||
| rural areas with a low population density | 1.14 (1.00–1.30) | — | — | — | — |
| rural areas with population concentrations | 1.07 (0.94–1.21) | — | — | — | — |
| urban districts | 1.12 (0.99–1.25) | — | — | — | — |
| big urban municipalities | reference | ||||
| Socioeconomic deprivation index | |||||
| lowest deprivation | reference | reference | reference | reference | reference |
| low deprivation | 0.97 (0.88–1.08) | 1.47 (0.74–2.91) | 0.95 (0.76–1.19) | 0.91 (0.79–1.05) | 1.10 (0.87–1.40) |
| medium deprivation | 1.01 (0.91–1.13) | 1.48 (0.76–2.89) | 0.98 (0.79–1.23) | 0.95 (0.82–1.11) | 0.95 (0.74–1.21) |
| high deprivation | 1.12 (1.00–1.26) | 1.62 (0.83–3.15) | 1.14 (0.91–1.42) | 1.09 (0.93–1.28) | 1.03 (0.79–1.34) |
| highest deprivation | 1.00 (0.90–1.12) | 1.42 (0.72–2.79) | 1.11 (0.88–1.41) | 0.95 (0.78–1.17) | 0.89 (0.72–1.11) |
| Proportion of individuals with a foreign citizenship | |||||
| lowest quintile | 1.29 (1.10–1.51) | 0.78 (0.54–1.14) | 1.71 (0.89–3.31) | 1.55 (1.20–2.00) | 1.32 (0.95–1.84) |
| low quintile | 1.24 (1.07–1.44) | 0.84 (0.56–1.25) | 1.68 (0.88–3.21) | 1.42 (1.12–1.80) | 0.96 (0.72–1.28) |
| middle quintile | 1.22 (1.05–1.40) | 0.80 (0.53–1.20) | 1.63 (0.86–3.09) | 1.43 (1.13–1.81) | 1.04 (0.82–1.33) |
| high quintile | 1.10 (0.96–1.26) | — | 1.52 (0.77–2.99) | 1.16 (0.94–1.45) | 1.14 (0.94–1.39) |
| highest quintile | reference | reference | reference | reference | reference |
| Density of paediatric psychiatrists (per 100,000 citizens) | |||||
| lowest quintile (<0.61) | reference | reference | reference | reference | reference |
| low quintile (0.61–0.90) | 1.02 (0.93–1.12) | 1.12 (0.96–1.32) | 1.04 (0.88–1.23) | 0.92 (0.78–1.07) | 0.93 (0.71–1.22) |
| middle quintile (0.91–1.20) | 1.23 (0.91–1.13) | 0.98 (0.79–1.21) | 1.32 (1.13–1.54) | 1.24 (1.09–1.41) | 1.16 (0.90–1.50) |
| high quintile (1.21–1.80) | 1.22 (1.00–1.26) | 1.08 (0.83–1.40) | 1.17 (0.92–1.50) | 1.19 (1.01–1.39) | 1.21 (0.86–1.69) |
| highest quintile (>1.80) | 1.22 (0.90–1.12) | 1.22 (0.99–1.50) | 1.20 (0.99–1.45) | 1.40 (1.15–1.71) | 1.03 (0.79–1.36) |
| Density of paediatricians (per 100,000 citizens) | |||||
| lowest quintile (<5.40) | reference | reference | reference | reference | reference |
| low quintile (5.40–6.20) | 0.94 (0.86–1.03) | 1.02 (0.85–1.22) | 0.86 (0.73–1.01) | 0.93 (0.78–1.07) | 0.83 (0.43–1.61) |
| middle quintile (6.21–7.10) | 0.96 (0.87–1.06) | 1.08 (0.89–1.32) | 0.82 (0.68–0.98) | 1.24 (1.09–1.41) | 0.97 (0.48–2.00) |
| high quintile (7.11–8.50) | 1.06 (0.95–1.19) | 1.17 (0.85–1.63) | 1.03 (0.83–1.28) | 1.19 (1.01–1.39) | 0.77 (0.40–1.47) |
| highest quintile (>8.50) | 1.12 (1.01–1.24) | 1.06 (0.82–1.37) | 1.08 (0.88–1.33) | 1.22 (1.01–1.47) | 0.90 (0.47–1.71) |
| Random effects | |||||
| Variance (SE) from the empty model | 0.107 (0.0077) | 0.098 (0.0147) | 0.099 (1.0143) | 0.111 (0.0137) | 0.108 (0.0192) |
| Variance (SE) from the final model | 0.092 (0.0067) | 0.089 (0.0133) | 0.081 (0.0116) | 0.070 (0.0087) | 0.080 (0.0143) |
| MOR (95% Crl) from the empty model | 1.36 (1.33–1.39) | 1.35 (1.29–1.40) | 1.35 (1.29–1.40) | 1.37 (1.32–1.42) | 1.37 (1.29–1.44) |
| MOR (95% Crl) from the final model | 1.33 (1.30–1.36) | 1.33 (1.27–1.38) | 1.31 (1.26–1.36) | 1.29 (1.24–1.32) | 1.31 (1.24–1.37) |
| PCV | −14% | −10% | −19% | −37% | −25% |
AOR, adjusted odds ratio; CI, confidence intervals; Crl, credible interval; MOR, median odds ratio; PCV, proportional change in variance; SE, standard error.