“…There were 139 studies involving potentially relevant published data, and 56 were retained after screening titles and abstracts. And 29 studies were excluded due to those reasons: (i) 9 studies were reviews about depression and osteoporosis or BMD (Bab & Yirmiya, 2010;Carlone et al 2015;Cizza et al 2009;Gold & Solimeo, 2006;Ilias et al 2006;Williams et al 2009); (ii) 7 studies assessed antidepressant medications and osteoporosis (Diem, Blackwell, Stone, Yaffe, Haney, et al, 2007;Haney et al 2007;Williams et al 2008;Aydin et al 2011;Rizzoli et al 2012;Diem et al 2013;Bruyère & Reginster, 2014); (iii) 3 studies were not a case-control design (Coelho et al 1999;Jacka et al 2005;Lunsford et al 2014); (iv) 1 study did not measure BMD levels (Tolea et al 2007); (v) 5 studies reported osteoporosis with normalized BMD value or T-score or Z-score without raw data (Erez et al 2012;Furlan et al 2005;Govender et al 2010;Kurmanji et al 2010;Lourenço et al 2014); (vi) 4 studies were meta-analyses up to 2009 (Cizza et al 2010;Wu et al, 2009Wu et al, , 2010Yirmiya & Bab, 2009); and (vii) 1 study was a follow-up study using duplicated population (Schweiger et al 2000). Finally, there were 26 studies included in our meta-analysis from January 1990 to January 2019 (Schweiger et al 1994;Michelson et al 1996;Amsterdam & Hooper, 1998;Reginster et al 1999;Whooley et al 1999;Schweiger et al 2000;Robbins et al, 2001;…”