93 Patient cost-sharing, mental health care and inequalities introduced in that year. As expected, the coefficient for the year of 2012 is the smallest, reflecting the combined effect of the mandatory deductible (220 euros) and the copayments. Figure A3. Standard regression discontinuity coefficients for each year Notes: Coefficients are estimated based on local linear regression discontinuities using the same set of controls as in model (1) and also excluding observations between months 217 and 227. The red horizontal dashed line represents the average of the coefficients for the low deductible period (2009-2011). Third, we explore the assumption of time-invariant manipulation effects – which will be differenced out by the difference-in-discontinuity approach – by looking at whether age-trends of mental health care use up to age 18 – i.e. before becoming eligible for the deductible – are parallel in the low versus high deductible periods. In Table A1, coefficients Age and Age2 refer to the trend in mental health care for the low deductible period, and coefficients Age x High deductible and Age2 x High deductible refer to the difference between the high and low deductible period. For both genders, the insignificance of the latter coefficients suggests that our findings are not driven by time-varying anticipatory behaviour. 3
RkJQdWJsaXNoZXIy MjY0ODMw