Totally, 3 701 prim iparousw om en in Shanghaiw ere follow ed up for 15 m onths after delivery to assess their initialand subsequent changes in choice of contraceptive m ethods. Results show ed that 15 m onths after delivery, 95.46of w om en had al- ready used a contraceptive m ethod. The leading firstchoice of contraceptive m ethods am ong urban w om en w asthe condom (50.72) or IUD(29.09), w hile the leading firstchoice am ong ruralw om en w asIUD(56.65) or the condom (30.60). Within 15 m onths after delivery, w om en w hose first choice had been the condom tended to sw itch usually to an IUD. Butfew er w om en w hosefirstm ethod w asan IUD sw itched to other m ethods. Atthe 15th m onth follow -up, 94.54of w om en w ere all using a m ethod, w ith 60.76of urban w om en and 71.07of rural w om en using IUDs. Multinom inal logistic regression analysis suggests that w om en’s characteristics influ- ence the choice of m ethods. Urban, w ell-educated w om en w ere m ore likely to choose condom and rhythm , and lesslikely to chooseOC. Older w om en tended to use condom and rhythm m ore often than younger w om en. Breastfeeding w om en w ere lesslikely to use OC. Wom en w ho received recom m endations regarding postpartum contraception from fam ily planning w orkers w ere m ore likely to use IUD. It w as concluded that fam ily planning w orkers should increase w om en’s aw areness of the advantages and disadvantagesof differentm ethods, dispelmythsand rum orsaboutsom em ethodsand improve w om en’sknow ledge of inform ed choice. Totally, 3 701 prim iparousw om en in Shanghaiw ere follow ed up for 15 m onths after delivery to assess their initialand subsequent changes in choice of contraceptive m ethods. Results show that that 15 m on ths after delivery, 95.46of w om en had The leading first choice of contraceptive m ethods am ong urban w om w asthe condom (50.72 ) or IUD (29.09 ), w hile the leading first choice am ong rural w om w as IUD (56.65 ) or the condom (30.60). Within 15 m onths after delivery, w om en w hose first choice had been the condom tended to sw itch usually to an IUD. Butfew er w om en w hosefirstm ethod w asan IUD sw itched to other m ethods. At the 15th m onth follow-up, 94.54 ° of w om en w ere all using am ethod, w ith 60.76 ° of urban w om and 71.07 ° of rural w om using IUDs. Multinom inal logistic regression analysis suggests that w om en’s characteristics influ- ence the choice of m ethods. Urban, w ell-educated w om en w ere m ore likely to choose condom and rhythm, and less likely to choose OC. Older w om en tended to use condom and rhythm m ore often than younger w om. ho received recom m endations regarding postpartum contraception from fam ily planning w orkers w ere m ore likely to use IUD. It w as permit that fam ily planning w orkers should increase w om en’s aw areness of the advantages and disadvantages of different m ethods, dispelmythsand rum orsaboutsom em ethodsand improve w om en skipnow of inform ed choice.