Table 1 General studies, effect of completion, Cox regressions for transitions. The coefficients are hazard ratio estimates with lower and upper 95% confidence limits. Reference non-completion

Work to Unemp

0.530 (0.519, 0.542)

0.608 (0.594, 0.622)

0.607 (0.590, 0.624)

Work to Sick leave

0.471 (0.462, 0.481)

0.570 (0.558, 0.582)

0.565 (0.551, 0.580)

Work to Education

2.174 (2.101, 2.249)

1.905 (1.840, 1.973)

1.693 (1.625, 1.763)

Unemp to Work

1.281 (1.256, 1.306)

1.191 (1.167, 1.216)

1.201 (1.165, 1.238)

Unemp to Sick leave

0.637 (0.606, 0.670)

0.736 (0.697, 0.778)

0.731 (0.688, 0.776)

Unemp to Education

3.277 (3.103, 3.462)

2.593 (2.450, 2.743)

2.214 (2.075, 2.361)

Unemp to Disability

0.454 (0.342, 0.603)

0.606 (0.441, 0.834)

0.554 (0.395, 0.777)

Sick leave to Work

1.401 (1.371, 1.432)

1.298 (1.269, 1.328)

1.314 (1.262, 1.369)

Sick leave to Unemp

0.859 (0.827, 0.893)

0.891 (0.855, 0.927)

0.903 (0.864, 0.945)

Sick leave to Education

4.881 (4.088, 5.828)

3.600 (3.005, 4.314)

3.317 (2.705, 4.068)

Sick leave to Disability

0.743 (0.561, 0.984)

0.929 (0.681, 1.267)

0.940 (0.685, 1.291)

Education to Work

1.115 (1.080, 1.150)

1.123 (1.089, 1.160)

1.136 (1.092, 1.181)

Education to Unemp

0.867 (0.817, 0.919)

0.898 (0.846, 0.953)

0.883 (0.820, 0.951)

Education to Sick leave

0.784 (0.654, 0.940)

0.837 (0.697, 1.006)

0.858 (0.693, 1.062)