Results

Background characteristics of study participants

A total of 18,045 women who delivered a live birth between 12 and 24 months immediately preceding were interviewed for the present study. Of these, 6005, 6010, and 6030 were from the eastern, central, and western regions of China, respectively. Participants ranged in age from 16 to 50 years, with a mean age of 29.9 ± 4.6 years. More young mothers (aged 20–24) were interviewed in the western (15%) region relative to the eastern (9%) and central (7%) regions. Less than half (48%) of the participants possessed an undergraduate education, about 22% had completed high school, and 22% had finished middle school. There were more less-educated women (middle school or below) in the western (33%) region than in the eastern (23%) or central (18%) regions. Han nationality accounted for 88% of the study participants, although the proportion of minorities was significantly higher in the western (12%) region relative to the eastern and central regions (5%). Nearly half (48%) of all participants were interviewed during the 13–18 months postpartum; this proportion was higher in the eastern (52%) region compared with the central (44%) and western (48%) regions. Baseline demographics and characteristics of participants are shown in supplementary Table S1.
Contraceptive rates, unmet need, and modifications in contraceptive method use during the first 24months postpartum
The cumulative 6-,12-, and 24-month contraception rates for participants using any method were 73.9% (95% CI 73.3-74.5), 85.7% (95% CI 85.2-86.2), and 86.8% (95% CI 86.3-87.3), respectively; while the rates for modern methods were 62.7% (95% CI 62.0-63.4), 72.4% (95% CI 71.8-73.1), and 73.2% (95% CI 72.6-73.9), respectively. Among women who resumed sexual intercourse after childbirth, the 6-, 12-, and 24-month values for unmet need for postpartum contraception were 28.7% (95% CI 28.0-29.4), 25.6% (95% CI 25.0-26.2) and 23.8% (95% CI 23.2-24.4), respectively. The level of contraceptive use was highest in the west (6-24 months: 77.8-88.6%) , followed by the east (6-24 months: 72.6-87.7%), and lowest in the central region (6-24 months: 71.4-84.0%) during the 24 months postpartum. In contrast, the percentage of unmet need for postpartum contraception was highest in the central region (6-24 months: 30.3-25.5%) and lowest in the east (6-24 months: 27.2-22.4%). (see Supplementary material, Table S2).
Condoms were dominant postpartum contraceptive method initially used after childbirth, accounting for approximately 2/3 of study participants, followed by traditional methods (14%), IUDs (2%), and hormonal methods (1%). More than 10% (12%) of participants did not use any methods. The variety of methods for first choice showed a similar pattern betweens regions. (see Supplementary material, Table S3).
A considerable number of initial contraceptive users changed their first choices to current methods. Of the 12,097 initial condom users, 10% switched to IUDs, 2% used traditional methods, and 4% became non-users; approximately 46% of traditional-method users converted to condoms, 13% changed to IUDs, and 8% became non-users; 7% of users replaced IUDs with condoms, 4% removed IUDs and became non-method users; 30% of combined oral contraceptive (COC) users changed to condoms, 12% turned to IUDs, and 3% to traditional methods. More than half (52%) of emergency contraceptive users turned to condom use and 19% to IUDs. (see Supplementary material, Table S4).

Pregnancy and unintended pregnancy during the first 24 months postpartum

Our study encompassed a total of 2186 pregnancies after index births. Of these, 1693 (77%) were unintended. More than half (52%, 1141) of the pregnancies were conceived within the first 12 months after delivery, and 83% (949/1141) of these were unintended.
Results of life-table analyses showed that the overall cumulative 6-, 12-, and 24-month pregnancy rates after childbirth were 1.7%, 6.3%, and 18.5%, respectively. The rates were 1.5%, 5.6%, and 15.5% in the east; 1.1%, 5.0%, and 17.8% in the central region; and 2.3%, 8.4%, and 21.8% in the west, respectively (Table 1). As shown in Figure 1, the curve for the western region increased faster than that for the other two regions. Our log-rank test results indicated that pregnancy rate was significantly higher in the west than in either of the other two regions during the first two years postpartum (allP <0.001), while the rates for the eastern and central regions were essentially similar (P = 0.329).
The overall cumulative 6-,12-, and 24-month unintended pregnancy rates were 1.4%, 5.3%, and 13.6%, respectively. The regional rates were 1.4%, 4.8%, and 11.8% in the east; 1.0%, 4.2%, and 13.3% in the central region; and 2.0%, 6.9%, and 15.6 in the west, respectively (Table 1). The curve reflecting the unintended pregnancy rate rose faster in the west than in the other two regions during the 24 months postpartum (Figure 2, log-rank test, P <0.001), and as the lines for the eastern and central regions overlapped, we observed no statistical difference in the rates between the latter two regions during the 24 months after delivery (log-rank test, P =0.183).

Abortion during the first 24 months postpartum

Of the 1141 pregnancies conceived within 12 months of a delivery, 708 (62%) ended in abortion, accounting for 75% (708/949) of unintended pregnancies during the postpartum period. However, only 8.3% (16/192) of the pregnancies not reported as unintended pregnancies ended in abortion. By 24 months postpartum, 59% (1286/2186) of all pregnancies and 76% (1286/1693) of unintended pregnancies ended in abortion. Only 10% (50/493) of the non-unintended pregnancies were terminated by abortive procedures.
Results of life-table analyses showed that the overall cumulative 6-, 12-, and 24-month rates of induced abortion were 1.1%, 4.0%, and 10.8%, respectively (Table 1). In addition, the cumulative 6-, 12-, and 24-month unintended pregnancy rates were 1.0%, 4.0%, and 10.2% in the east; 0.7%, 3.0%, and 10.0% in the central region; and 1.4%, 5.1%, and 12.1% in the west, respectively.
The cumulative rate for induced abortions increased fastest in the west and slowest in the central region during the 24 months postpartum (Figure 3). Women in the west were significantly more likely to undergo an abortion relative to their counterparts in the eastern (P = 0.008) and central regions (P <0.001), and women in the east were significantly more likely to have a pregnancy terminated by abortion compared to those in the west (P = 0.003) during the first two years after childbirth.