Table 2. Quality rating of included papers

First author	GlobalScore	Selection bias	Study design	Confounders	Blinding	Data collection methods	Withdrawals and drop outs
Tobe²⁹	weak	weak	mod	weak	weak	strong	mod
Kruis³⁰	mod	weak	strong	strong	mod	strong	mod
Vicens³¹	strong	strong	strong	strong	weak	strong	strong
Keeley³²	strong	mod	strong	strong	strong	strong	strong
Kristoffersen³³	mod	weak	strong	strong	strong	strong	strong
van Dijk-de Vries³⁴	weak	weak	strong	weak	mod	strong	mod
Racic³⁵	strong	strong	strong	strong	mod	strong	strong
Vicens³⁶	strong	strong	strong	strong	strong	strong	strong
Kristoffersen³⁷	mod	weak	strong	strong	strong	strong	strong
Zwar³⁸	weak	weak	strong	strong	strong	mod	strong
van Lieshout³⁹	mod	weak	strong	strong	mod	strong	strong
Vaillant-Roussel⁴⁰	mod	weak	strong	mod	mod	strong	mod
Griffiths⁴¹	strong	strong	strong	mod	mod	strong	mod
Eikelenboom⁴²	strong	strong	strong	strong	mod	strong	strong
Ramli⁴³	mod	strong	strong	strong	weak	strong	strong
Keeley⁴⁴	mod	weak	strong	strong	mod	strong	strong
Kristoffersen⁴⁵	weak	weak	strong	strong	weak	strong	strong
Baldeón⁴⁶	weak	weak	strong	weak	mod	strong	mod

mod = moderate.