78 Chapter 3 -35.9%, -17.1%; panel A of Table 4). Increased treatment cessation in the high deductible period occurred regardless of treatment intensity or use of medicines for mental health disorders. Panel B of Table 4 suggests that young adults receiving treatment of low and medium intensity had a higher cessation in relative terms (around -26% for females and between -32% to -40% for males). Nevertheless, we also find a sizeable higher discontinuation for those treated above the median intensity (-13% to -18% for females and around -22% to -23% for males). Table 3. Difference in mental health care use at the transition to adulthood between high and low deductible periods, overall and by household income subgroups. Age 18 x High deductible (pp) Baseline use Relative change Observations Panel A: Overall use Female -1.1 (-1.8 , -0.4)*** 8.1% -13.6% (-22.1%, -5.2%)*** 981,814 Male -0.3 (-0.7 , 0.1) 5.6% -5.3% ( -11.8%, 1.2%) 1,022,663 Panel B:Income quartiles a Female 1st -1.7 (-3.1 , -0.2)** 8.8% -18.9% ( -35.4%, -2.3%)** 244,767 2nd -1.9 (-3.3 , -0.5)*** 9.0% -21.3% ( -36.7%, -5.9%)*** 246,632 3rd -0.2 (-1.5 , 1.1) 7.7% -2.9% ( -20.0%, 14.1%) 246,093 4th -0.6 (-1.9 , 0.7) 6.9% -8.9% ( -28.0%, 10.2%) 244,322 Male 1st -0.3 (-1.2 , 0.5) 6.8% -5.0% ( -17.0%, 7.1%) 253,763 2nd -0.4 (-1.2 , 0.3) 5.3% -8.4% ( -22.3%, 5.5%) 256,682 3rd -0.4 (-1.1 , 0.3) 5.4% -7.2% ( -20.2%, 5.8%) 256,333 4th 0.0 (-0.7 , 0.6) 4.9% -0.7% ( -14.3%, 12.9%) 255,885 Notes: pp – percentage points, *p-value<0.05, **p-value<0.01, ***p-value<0.001, (Confidence Intervals at 95% level). a First quartile corresponds to the lowest income group. Baseline use refers to the average mental health care use across study years at 228 months of age, corresponding to the first moment in which young adults pay the full-year deductible. Best fitting age polynomial is quadratic for female and linear for male. Estimates for income display main coefficients for each quartile, obtained by repeating the estimation of the interacted model with each quartile as reference category. Hence, statistical significance reported in the table concerns differences of the main term from zero. Within each model the interaction terms for the remaining quartiles are not significantly different from the main term (p-value for the interaction coefficient > 0.05). All models include year dummies and control for migratory background and postcode. The baseline model (row Male and Female) also controls for income quartiles at the year of the 17th birthday. Observations are in person-years. Panel C of Table 4 presents the treatment cessation by the number of psychotropic drug classes used at age 17. Males using one psychotropic drug class were most likely to discontinue care in the high versus low deductible period (-46.5%, CI 95%: -62.5%, -30.3%) with similar discontinuation of care in relative terms for the remaining subgroups of both genders. When restricting to single drug class users (panel D of Table 4), we find the largest discontinuation of services for those using agents for ADHD at age 17. In
RkJQdWJsaXNoZXIy MjY0ODMw