Article Figures & Data
Figures
- Figure 1
Geographic Variation in New Uses of Antidepressants for Women Aged 46–65, per 1,000, the Netherlands 2009 and 2016
Notes: The maps plot the number of new antidepressant patients among women aged 46–65 in each four-digit postal code (2,001) that has inhabitants in all years 2006–2018 and that has at least one GP practice in all years 2009–2016. Prescribing measure plotted for 2009 and 2016 and postal codes are split into terciles of this measure of prescribing intensity. The black lines on the maps show the 2016 municipality boundaries.
- Figure 2
Antidepressant Usage Relative to Childbirth
Notes: Panel A is based on the sample of first births (N = 448,226) and splits the sample into no use before birth (no prescriptions in the 10–24 months prior to childbirth), “ever takers” (those with at least one prescription in the 10–24 months prior to childbirth, and “always takers” (those with prescriptions in more than 50 percent of the 10–24 months prior to childbirth). Panel B focuses on those with no use before birth (N = 427,475).
- Figure 3
Antidepressant Usage by Area-Level Prescribing Intensity
Notes: Panel A reports the proportion of individuals dispensing a script for antidepressants at a pharmacy in each month relative to childbirth for first births and never users (N = 427,475). It splits the sample into three groups depending on terciles of the instrument—new antidepressant patients in the woman’s postal code of residence in the year prior to giving birth (our instrument). Panel B plots residualized antidepressant use after controlling for the wide range of individual-level characteristics and postal code–level characteristics as mentioned in the empirical strategy.
- Figure 4
Balance Check: Do Individual Characteristics Predict the Instrument?
Notes: This figure plots the coefficients and 95 percent confidence intervals from a balance check in which we regress the value of the instrument for everyone (N = 427,475) on their individual-level characteristics (plotted), and include year of birth fixed effects, the four-digit postal code–level controls, and (481) area-level fixed effects. Standard errors are clustered at the area level.
Tables
(1) (2) (3) (4) (5) (6) (7) (1) New female patients aged 46–65 1.0000 (2) New female patients aged 15–45 0.3849 1.0000 (3) % population Dutch –0.3532 –0.1347 1.0000 (4) % low income HHs 0.4023 0.3175 –0.6113 1.0000 (5) % welfare recipients 0.4206 0.3990 –0.6020 0.8207 1.0000 (6) Number of GP practices 0.0873 0.0310 –0.3076 0.2461 0.1821 1.0000 (7) Total population 0.1457 0.0386 –0.4584 0.2372 0.2335 0.5878 1.0000 Notes: The measures of new antidepressant patients and total population at the four-digit postal code level are calculated by the authors, and measures for 2016 are used to calculate the correlations. The postal code characteristics (apart from mean number of GP practices) come from Statistics Netherlands (in Dutch: Kerncijfers per postcode) and are shown for 2016. The mean number of GP practices in each postal code is calculated over the 2009–2016 period using data obtained from NIVEL, Netherlands Institute for Health Services Research.
- Table 2
First Stage, Subsequent Antidepressant Use and Future Specialized Mental Healthcare
Antidepressant Use Specialized Mental Health Care Month 1–10 Month 11–20 Month 21–30 Month 1–10 Month 11–20 Month 21–30 (1) (2) (3) (4) (5) (6) Panel A1: OLS AD 1–10 mo after birth 0.573** 0.390** 0.401** 0.363** 0.279** (0.005) (0.005) (0.008) (0.008) (0.008) Panel A2: 2SLS New patients aged 46–65 0.197** (0.050) AD 1–10 mo. after birth 0.609** 0.687** 0.064 –0.005 0.193 (0.218) (0.248) (0.332) (0.342) (0.401) Weak IV 95% AR confidence set [0.111; 1.063] [0.221; 1.299] [−0.692; 0.689] [−0.919; 0.637] [−0.800; 1.107] Mean outcome 0.025 0.028 0.030 0.045 0.044 0.043 F-statistic first stage 15.23 15.23 15.23 16.28 13.23 10.74 Observations 427,475 427,475 427,475 254,489 214,617 174,249 Panel B1: OLS Top 25% Income AD 1–10 mo. after birth 0.612** 0.421** 0.438** 0.348** 0.250** (0.013) (0.013) (0.017) (0.018) (0.018) Panel B2: 2SLS Top 25% Income New patients aged 46–65 0.334** (0.071) AD 1–10 mo. after birth 0.612** 0.560* 0.095 0.245 0.661 (0.164) (0.221) (0.405) (0.422) (0.406) Weak IV 95% AR confidence set [0.272; 0.953] [0.145; 1.018] [−0.908; 0.937] [−1.051; 1.123] [−0.022; 1.825] Mean outcome 0.013 0.014 0.016 0.028 0.026 0.026 F-statistic first stage 21.88 21.88 21.88 11.86 8.33 12.68 Observations 106,868 106,868 106,868 72,432 59,442 46,670 Panel C1: OLS Bottom 75% Income AD 1–0 mo. after birth 0.566** 0.385** 0.393** 0.365** 0.283** (0.006) (0.006) (0.008) (0.009) (0.009) Panel C2: 2SLS Bottom 75% Income New patients aged 46–65 0.162** (0.060) AD 1–10 mo. after birth 0.646 0.750* 0.063 –0.088 –0.039 (0.334) (0.371) (0.472) (0.459) (0.612) Weak IV 95% AR confidence set [−0.314; 1.540] [0.052; 2.256] [−1.291; 1.137] [−1.680 0.867] [−4.703; 1.475] Mean outcome 0.029 0.033 0.035 0.051 0.050 0.049 F-statistic first stage 7.16 7.16 7.16 8.23 7.42 4.74 Observations 320,607 320,607 320,607 182,057 155,175 127,579 Notes: Panels B and C split the sample by household income (mean of the sum of maternal and paternal income in the two years prior to first birth). The antidepressant use dummies for months 1–10, 11–20, and 21–30 are equal to one if the woman was dispensed at least one script for antidepressants in these months. Specialized mental health care is a dummy variable for receipt of such care in the 1–10, 11–20, or 21–30 months after giving birth. All specifications contain area-level fixed effects, year of birth fixed effects, individual controls, and postal code–level controls. Robust standard errors are clustered at the area level and are shown in parentheses. Significance: *p < 0.05, **p < 0.01.
Positive Income Log Earnings Y1 Y2 Y3 Y1 Y2 Y3 (1) (2) (3) (4) (5) (6) Panel A1: OLS AD 1–10 mo, after birth –0.092** –0.101** –0.104** –0.141** –0.159** –0.205** (0.004) (0.004) (0.004) (0.012) (0.011) (0.011) Panel A2: 2SLS AD 1–10 mo. after birth –0.821* –0.089 –0.021 1.042 0.305 –0.416 (0.406) (0.420) (0.405) (1.213) (1.331) (1.329) Weak IV 95% AR confidence set [−1.907; −0.137] [−0.962; 0.868] [−0.862; 0.901] [−1.238; 4.283] [−2.462; 3.337] [−3.179; 2.347] Mean outcome 0.910 0.905 0.900 €30,590 €31,947 €33,457 F-statistic first stage 15.23 15.23 15.23 13.09 13.72 16.93 Observations 427,475 427,475 427,475 386,973 384,930 382,611 Panel B1: OLS Top 25% Income AD 1–10 mo. after birth –0.028** –0.034** –0.045** –0.063** –0.094** –0.146** (0.006) (0.007) (0.008) (0.022) (0.019) (0.023) Panel B2: 2SLS Top 25% Income AD 1–10 mo. after birth –0.179 0.329 0.021 0.172 –0.388 0.687 (0.269) (0.321) (0.303) (1.062) (1.213) (1.554) Weak IV 95% AR confidence set [−0.791; 0.327] [−0.276; 0.997] [−0.609; 0.591] [−1.825; 2.591] [−2.910; 2.374] [−2.236; 4.533] Mean outcome 0.978 0.972 0.968 €48,670 €50,504 €52,673 F-statistic first stage 21.88 21.88 21.88 18.89 16.39 14.11 Observations 106,868 106,868 106,868 104,146 103,464 102,926 Panel C1: OLS Bottom 75% Income AD 1–10 mo. after birth –0.096** –0.106** –0.107** –0.131** –0.151** –0.195** (0.004) (0.004) (0.004) (0.012) (0.013) (0.013) Panel C2: 2SLS Bottom 75% Income AD 1–10 mo. after birth –1.331* –0.473 –0.196 0.229 –0.321 –1.921 (0.665) (0.599) (0.599) (1.837) (1.845) (1.774) Weak IV 95% AR confidence set [−4.557; −0.344] [−2.428; 0.890] [−1.915; 1.403] [−5.773; 6.594] [−6.348; 4.245] [−7.363; 1.415] Mean outcome 0.887 0.883 0.878 €23,933 €25,125 €26,385 F-statistic first stage 7.16 7.16 7.16 5.64 6.84 9.71 Observations 320,607 320,607 320,607 282,827 281,466 279,685 Notes: Panels B and C split the sample by household income (mean of the sum of maternal and paternal income in the two years prior to first birth). Positive income is a dummy variable and indicates that an individual had income greater than zero in the full calendar years after the birth year (Y1–Y3). Log earnings conditional on employment (after an inflation correction) is also reported for the three years after the birth year (Y1–Y3). All specifications include area-level fixed effects, year of birth fixed effects, individual controls, and postal code–level controls. Robust standard errors are clustered at the area level and are shown in parentheses. Significance: *p < 0.05, **p < 0.01.
Father Lives at Same Address Father Lives in Same Municipality Second Birth Y1 Y2 Y3 Y1 Y2 Y3 <30 mo (1) (2) (3) (4) (5) (6) (7) Panel A1: OLS AD 1–10 mo. after birth –0.019** –0.033** –0.042** –0.011** –0.018** –0.021** –0.137** (0.003) (0.003) (0.003) (0.002) (0.002) (0.003) (0.004) Panel A2: 2SLS AD 1–10 mo. after birth 0.148 0.309 0.512 0.201 0.086 0.174 –0.256 (0.307) (0.351) (0.382) (0.270) (0.270) (0.305) (0.752) Weak IV 95% AR confidence set [−0.428; 0.907] [−0.351; 1.246] [−0.131; 1.608] [−0.308; 0.869] [−0.475; 0.702] [−0.400; 0.930] [−1.819; 1.456] Mean outcome 0.906 0.896 0.884 0.934 0.928 0.922 0.393 F-statistic first stage 15.23 15.23 15.23 15.23 15.23 15.23 15.23 Observations 427,475 427,475 427,475 427,475 427,475 427,475 427,475 Panel B1: OLS Top 25% Income AD 1–10 mo. after birth –0.014** –0.025** –0.036** –0.003 –0.012* –0.019** –0.205** (0.005) (0.006) (0.007) (0.004) (0.005) (0.005) (0.012) Panel B2: 2SLS Top 25% Income AD 1–10 mo. after birth 0.205 0.094 0.331 0.443 0.151 0.383 –1.143 (0.293) (0.270) (0.321) (0.257) (0.218) (0.255) (0.926) Weak IV 95% AR confidence set [−0.346; 0.871] [−0.468; 0.656] [−0.273; 1.063] [0.010; 1.079] [−0.258; 0.647] [−0.045; 1.013] [−3.435; 0.415] Mean outcome 0.969 0.969 0.964 0.982 0.981 0.979 0.477 F-statistic first stage 21.88 21.88 21.88 21.88 21.88 21.88 21.88 Observations 106,868 106,868 106,868 106,868 106,868 106,868 106,868 Panel C1: OLS Bottom 75% Income AD 1–10 mo. after birth –0.018** –0.032** –0.041** –0.011** –0.018** –0.019** –0.123** (0.003) (0.004) (0.004) (0.003) (0.003) (0.003) (0.005) Panel C2: 2SLS Bottom 75% Income AD 1–10 mo. after birth –0.022 0.252 0.405 –0.023 –0.064 –0.090 –0.108 (0.444) (0.527) (0.562) (0.391) (0.411) (0.457) (1.036) Weak IV 95% AR confidence set [−1.209; 1.252] [−0.843; 2.181] [−0.652; 2.797] [−1.068; 1.100] [−1.244; 1.035] [−1.401; 1.131] [−2.671; 3.274] Mean outcome 0.884 0.872 0.857 0.918 0.911 0.903 0.365 F-statistic first stage 7.16 7.16 7.16 7.16 7.16 7.16 7.16 Observations 320,607 320,607 320,607 320,607 320,607 320,607 320,607 Notes: Panels B and C split the sample by household income (mean of the sum of maternal and paternal income in the two years prior to the first birth). In the first three columns, the outcome variable is equal to one when the father and the child live at the same address. In Columns 4–6 the outcome variable is equal to one if the father and child live in the same municipality. Column 7 is a dummy variable that is equal to one if the mother had a second child within 30 months. All specifications include area-level fixed effects, year of birth fixed effects, individual controls, and postal code–level controls. Robust standard errors are clustered at the area level and shown in in parentheses. Significance: *p < 0.05, **p < 0.01.
- Table 5
Robustness Results: First Stage, Antidepressant Use and Mother’s Labor Market Outcomes
Antidepressant Use Mother Positive Income Month 1–10 Month 11–20 Month 21–30 Y1 Y2 Y3 (1) (2) (3) (4) (5) (6) Panel A1: OLS—Postal Codes with 3 or Fewer GP Practices AD 1–10 mo. after birth 0.577** 0.388** –0.090** –0.098** –0.101** (0.006) (0.006) (0.005) (0.005) (0.005) Panel A2: 2SLS—Postal Codes with 3 or Fewer GP Practices New patients aged 46–65 0.265** (0.059) AD 1–10 mos after birth 0.694** 0.712** –0.501 0.122 0.204 (0.192) (0.216) (0.325) (0.364) (0.351) Weak IV 95% AR confidence set [0.296; 1.093] [0.306; 1.247] [−1.241; 0.110] [−0.563; 0.950] [−0.457; 1.003] Mean outcome 0.025 0.028 0.030 0.911 0.907 0.901 F-statistic first stage 20.29 20.29 20.29 20.29 20.29 20.29 Observations 275,160 275,160 275,160 275,160 275,160 275,160 Panel B1: OLS—Rural Areas AD 1–10 mo. after birth 0.565** 0.381** –0.067** –0.073** –0.089** (0.010) (0.009) (0.007) (0.007) (0.007) Panel B2: 2SLS—Rural Areas New patients aged 46–65 0.314** (0.087) AD 1–10 mo. after birth 0.776** 1.025** –0.712 –0.791 –0.711 (0.231) (0.300) (0.419) (0.479) (0.458) Weak IV 95% AR confidence set [0.343; 1.393] [0.520; 1.945] [−2.000; −0.007] [−2.262; 0.016] [−2.115; 0.059] Mean outcome 0.024 0.027 0.029 0.927 0.922 0.915 F-statistic first stage 13.12 13.12 13.12 13.12 13.12 13.12 Observations 110,809 110,809 110,809 110,809 110,809 110,809 Panel C1: OLS—First and Second Births 0.576** 0.404** –0.082** –0.092** –0.098** (0.004) (0.004) (0.003) (0.003) (0.003) Panel C2: 2SLS – First and Second Births New patients aged 46–65 0.165** (0.036) AD 1–10 mo. after birth 0.577** 0.686** –0.713 –0.169 –0.210 (0.188) (0.239) (0.376) (0.365) (0.351) Weak IV 95% AR confidence set [0.187; 0.967] [0.236; 1.230] [−1.643; −0.006] [−0.927; 0.589] [−0.940; 0.450] Mean outcome 0.025 0.029 0.033 0.901 0.898 0.893 F-statistic first stage 20.95 20.95 20.95 20.95 20.95 20.95 Observations 768,740 768,740 768,740 768,740 768,740 768,740 Notes: Panel A focuses on 1,548 postal codes with three or fewer GP practices. Panel B focuses on rural areas (urbanicity categories four and five). Panel C includes both first and second births. The antidepressant use dummies for month 1 to 10, 11 to 20, and 21 to 30 are one if the woman was dispensed at least one antidepressant script in these months. Positive income is a dummy and indicates that an individual had income greater than zero in the full calendar years after birth (Y1–Y3). All specifications include area-level (municipality) fixed effects, individual controls. and postal code–level controls. Robust standard errors are clustered at the area level and shown in parentheses. Significance: *p < 0.05, **p < 0.01.
Free alternate access to The Journal of Human Resources supplementary materials is available at https://uwpress.wisc.edu/journals/journals/jhr-supplementary.html
Jump to section
Related Articles
Cited By...
- No citing articles found.