97 Patient cost-sharing, mental health care and inequalities APPENDIX G. SENSITIVITY ANALYSES Table A3. Impact of the deductible increase on mental health care use when transiting to adulthood: different age bandwidths, and placebo discontinuity at age 16 Female Male Age 18 x High deductible Observations Age 18 x High deductible Observations Age bandwiths 17-19 years old (main specification; pp) -1.1 (-1.8 , -0.4)*** 981,814 -0.3 ( -0.7, 0.1 ) 1,022,663 16-20 years old (pp) -0.8 (-1.2 , -0.4)*** 1,925,229 -0.4 (-0.6, -0.1)** 2,000,323 15-21 years old (pp) -0.8 (-1.1 , -0.4)*** 2,862,766 -0.4 (-0.6, -0.2)*** 2,973,573 Placebo discontinuity At 16 years old (pp) 0.1 (-0.3 , 0.6) 976,324 -0.2 (-0.6, 0.2) 1,020,190 Notes: pp – percentage points. *p-value<0.05 **p-value<0.01 ***p-value<0.001, (Confidence Intervals at 95% level). Quadratic age polynomials for females and linear age-trends for males. Models include year fixed effects and controls for postcode, migratory background and income quartile at the year of the 17th birthday. Observations are in person-years. Table A4. Impact of the deductible increase on mental health care use when transiting to adulthood: model specification Female Male Age 18 x High deductible Observations Age 18 x High deductible Observations Main specification (pp) -1.1 (-1.8 , -0.4)** 981,814 -0.3 (-0.7 , 0.1) 1,022,663 Without controls (pp) -1.1 (-1.8 , -0.4)** 981,814 -0.3 (-0.6 , 0.1) 1,022,663 With postcode-year control (pp) -1.1 (-1.8 , -0.4)** 981,814 -0.3 (-0.7 , 0.1) 1,022,663 Non-linear model (OR) 0.86 (0.79 , 0.94)*** 981,814 0.95 (0.89 , 1.01) 1,022,663 Notes: pp – percentage change. OR – Odds ratio *p-value<0.05 **p-value<0.01 ***p-value<0.001, (Confidence Intervals at 95% level). Quadratic age polynomials for female and linear age-trends for male. Except if otherwise mentioned models include year fixed effects and controls for postcode, migratory background and income quartile at the year of the 17th birthday. Observation are in person-years. 3
RkJQdWJsaXNoZXIy MjY0ODMw