Unmet need for postpartum family planning and influencing factors in China

To explore the rate and influencing factors of unmet need for postpartum family planning (PPFP) in China. We conducted a retrospective cohort study at 60 hospitals in 15 provinces that were in eastern, central, and western regions of China. Participants were women who delivered a live birth at the study sites with an interval of 13 to 24 months between delivery and interviews. We selected participants using cluster randomization approach, and the first 300 postpartum women who gave a birth at each study hospital after the initial month that had been selected were interviewed. Information on the women’s background characteristics, pregnancy history, time when menstruation and sexual activity resumed after childbirth, the adoption of contraceptive method, breastfeeding, and any pregnancy or pregnancy outcome after delivery were collected. We performed life-table analysis to estimate the rate of unmet need for PPFP and a 2-level logistic regression model to explore factors that influence unmet need for PPFP within the first 24 months postpartum. A total, 19,939 postpartum women were screened in this study, of which, 17,466 (87.6%) were eligible for this analysis. The rates of unmet needs for any FP methods were 23.9% (95% confidence interval [CI] 23.3–24.6%), 11.8% (95%CI 11.3–12.3%); 10.6% (95%CI 10.1–11.1%) at 6, 12, and 24 months postpartum; these rates for modern FP methods were 35.5% (95%CI 34.7–36.2%), 25.6% (95%CI 24.9–26.2%), and 24.6% (95%CI 23.9–25.2%), respectively. Results of 2-level logistic regression analysis showed that less-educated young women, those who had only one child or delivered by vaginal delivery at secondary hospitals, were associated with increased risk of unmet need for PPFP. Approximately 31% of women who had unmet need for PPFP reported a pregnancy during the first 24 months postpartum, which was significantly higher than the level for their counterparts (10.0%). The level of unmet need for PPFP in China was high, resulting in a high pregnancy rate within 24 months after delivery. Women’s age, education level, prior pregnancy and abortion histories, and delivery method were significantly associated with the risk of unmet need for PPFP. National PPFP guidelines that integrate PPFP services into prenatal and postnatal care are urgently needed and should be implemented throughout the country as soon as possible. PPFP services should promote the use of modern contraceptive methods.


Introduction
Sexual and reproductive health has been recognized as an important component of good health and well-being for human beings, and was made one of the targets of the United Nations Sustainable Development Goals for 2015. Nevertheless, the contraceptive prevalence rate (the percentage of women aged 15-49 who use any contraceptive method) increased only slightly from 47.7% in 2000 to 49.0% in 2020 (United Nations, 2020a, b). According to estimates in Kantorova et al. (2020), of the 1.9 billion women of reproductive age globally in 2019, 1.11 billion had a need for family planning (FP), and 270 million had an unmet need for modern methods. Unmet need for FP refers to women who want to delay or avoid a birth (or another birth) but are not using any family planning method (Bernstein & Edouard, 2007). Meeting FP needs continues to be an important public health concern both in developing and developed countries, and is essential for achieving Family Planning 2030 (Hardee & Jordan, 2021) and providing universal access to sexual and reproductive healthcare services, as set out in the United Nations Sustainable Development Goals (SDGs) (Kantorova et al., 2020).
Numerous studies have demonstrated that short inter-pregnancy intervals are associated with increased risks of adverse maternal and infant outcomes. For instance, compared with inter-pregnancy intervals of 18-23 months, intervals of less than 6 months have been associated with increased risks of preterm birth, low birth weight, and small size for gestational age (Conde-Agudelo et al., 2006). To achieve better maternal and child health outcomes, the World Health Organization (WHO) recommends an interval of at least 24 months after childbirth before becoming pregnant again (WHO, 2005), and suggests that postpartum women receive counseling on birth spacing and family planning (WHO, 2013). Promotion of postpartum family planning (PPFP) contributes to better health for mothers and children (Cleland et al., 2015). However, previous studies have shown that the levels of unmet need for FP varied among countries, e.g., ranging from 25 to 96% at postpartum day 42 in India, Pakistan, Zambia, Kenya, and Guatemala (Pasha et al., 2015). Analysis of Demographic and Health Surveys (DHS) data from 21 low-and middle-income countries conducted between 2005 and 2012 found that the overall prevalence of non-use of any method of PPFP by women 0-23 months postpartum was universally high, reaching a level of 61% (Moore et al., 2015). The average prevalence of women not using any modern contraceptive method 12 months after delivery was 48.5% in low-and middle-income 1 3 countries. The percentage was highest in South Asia and Southeast Asia (59.4%) and lowest (16.3%) in the Middle East and North Africa (Dev et al., 2019).
During last three decades, the Chinese government has made great efforts to promote contraceptive use in couples of reproductive age across the country. Several national surveys on reproductive health and FP have been conducted between 1997 and 2017, of which the 2017 China Fertility Survey is the most recent (Wei et al., 2020). According to the 2017 Survey, the overall contraceptive prevalence in China was 84.5%, but the survey did not make clear the level of unmet need (United Nations, 2020a, b). Haakenstad et al. (2022) published a comprehensive analysis that reported contraceptive prevalence and unmet need for any FP in 204 countries and territories in 2022, and based on the results, the prevalence of use of all contraceptive methods was 79.2% in China in 2019 and that of unmet need for any method was 2.9%. However, little is known about the unmet need for PPFP. A crosssectional study in 2015 of Shimen county, Hunan province showed that 66.4% of women had used some FP methods within 24 months after childbirth (Kang et al., 2018), which indicates that unmet need for PPFP would be high in China.
To date, no study has investigated the unmet need for PPFP. Our purpose in conducting this nationally representative retrospective cohort study among postpartum women in China during 2017 and 2018 was to understand levels of postpartum contraceptive use and unmet need for PPFP, and the levels of unintended pregnancy and induced abortion in the country. The levels of unmet need for any method and unmet need for modern methods, and associated factors that influence the existence of unmet need are presented in this paper. To our knowledge, this is the first study reporting national levels of unmet need for PPFP in China.

Study participants
We conducted an institutionally based retrospective cohort study at 60 public hospitals. First, we divided all of mainland China's 31 provinces into three macro-geographic regions: eastern, central, and western, and then selected five provinces from each region taking into account the north-south distribution of their geographic locations. In each of the 15 provinces, a senior physician was appointed as provincial coordinator responsible for defining a sampling frame and in charge of quality control of data collection for each province. Second, using a stratified randomization approach, a statistician randomly selected 4 hospitals from 10 to 12 candidates that were recommended by the provincial coordinator in each province, including two secondary hospitals (one general and one maternal & children's hospital) and two tertiary hospitals (one general and one maternal & children's hospital). A month between July 2015 and June 2016 was randomly assigned to each participating hospital as the initial month of investigation, allowing an interval of 13-24 months between delivery and interviews for all participants. The first 300 women who delivered a live birth after the initial month that had been selected were interviewed by trained investigators. The target size of the sample was 18,000, which was calculated based on the prevalence of unintended pregnancy among postpartum women in a given region. Detailed information on sample size calculation and procedures for selecting participants has been published elsewhere (Che et al., 2021).

Data collection
A structured questionnaire administered by face-to-face or telephone interviews were used for data collection. Two junior doctors and/or nurses were trained for data collection work at each participating hospital. Data was collected for demographic characteristics, pregnancy history, time when menstruation and sexual activity resumed after childbirth, the adoption of contraceptive method, and any pregnancy or pregnancy outcome after childbirth. The field work started in July 2017 and ended in January 2018.

Definition of unmet need for FP
Unmet need for FP is typically defined as the percentage of women of reproductive age who want to prevent or delay getting pregnant for two years, but are not using any method of contraception, either modern or traditional (Festin, 2020). It should be noted that the UN SDG indicator refers to unmet need for modern methods only. The broad definition including all methods has been widely used by researchers, for example, in a study by Pasha et al. (2015) and in Haakenstad's study (Haakenstad et al., 2022) mentioned above. Our study presents the rate of unmet need for both modern contraceptive methods and all contraceptive methods. To determine the level of unmet need for any FP, this paper divides postpartum women not using any FP method by all postpartum women, and multiplies the result by 100%. For modern FP, we divide postpartum women not using a modern FP method by all postpartum women, and multiply the result by 100%. Modern contraceptive methods include male and female condoms, oral contraceptive pills, injectables, intrauterine devices, sub-dermal implants, male and female sterilizations, and vaginal rings. Traditional methods include the rhythm/calendar method, and withdrawal.

Statistical analysis
We used EpiData 3.1 for data entry, and SPSS 25.0 and STATA 15.0 for data analysis. Frequencies, percentages, means, and cross-tabulations were used to summarize descriptive statistics. Life-table analysis was used to estimate the cumulative rate of unmet need for PPFP within the first 24 months postpartum. We assumed that a woman would not change her method once she began using a contraceptive method. We used 2-level logistic regressions (level-1: individual woman, level-2: hospital) to explore factors that influence unmet needs for PPFP within the first 24 months postpartum. We assessed the model fit using likelihood ratio (LR) test, which compared the LR of the data under the full model (including all predictors) against the LR of the data under empty model (including none predictor). A P value for the full model fit statistic that is < 0.05 would compel us to reject the empty model.

Background characteristics of participants
A total of 19,939 women who delivered a live birth in the study hospitals between July 2015 and June 2016 were contacted. After 2473 (12.4%) were excluded from this study because of their fertility intentions or for other reasons, a total of 17,466 (87.6%) postpartum women were included in this analysis, including 5685 from eastern, 5839 from central, and 5942 from western China. The average age was 29.89 ± 4.63 years for all participants, and 30.13 ± 4.41, 30.21 ± 4.47, and 29.37 ± 4.93 years for women from the eastern, central, and western regions, respectively. Participants from eastern region were more likely to be better-educated with white collar jobs, while those from western region were more likely to have multiple gravidities, births and abortions. Chi-square tests show that the distributions of women's characteristics were statistically significant between geographic regions (P < 0.001) ( Table 1).

Unmet need for PPFP
Results of life-table analyses show that the cumulative rates of unmet need for any FP and unmet need for modern FP at 3 months after delivery were 57.2% and 63.8%, respectively. These rates declined to 23.9% and 35.5% at 6 months, to 11.8% and 25.6% at 12 months, and further to 10.6% and 24.6% at 24 months after childbirth ( Table 2). As shown in Figs. 1 and 2, the rates declined rapidly during the first 6 months after delivery, but remained stable with little change between postpartum months 12-24. The pattern was similar in each of the three geographic regions, except that the levels of unmet need for both modern method and any method in the central region were slightly higher than in the other two regions beginning approximately 6 months after delivery.

Factors influencing unmet need for PPFP
Two-level logistic regression models were used to explore the factors that influence unmet need for PPFP during the 24 months after childbirth. As shown in Table 3, unmet need for any PPFP methods was significantly associated with geographic region, hospital level, maternal age and education, gravidity, number of children, delivery method, and length of time of exclusive breastfeeding. For instance, postpartum women from the central region who had two or more children and who only breastfed their child for longer than 6 months were significantly more likely to have unmet need for FP than their counterparts. Participants, who delivered at tertiary hospitals, were aged 25-34 and better-educated, whose gravidity was ≥ 2, and who gave birth via cesarean section were significantly more likely to use a PPFP method, compared to their counterparts.
The factors influencing the unmet need for modern PPFP methods were very similar to those for any PPFP method. The one exception was for women who had a history of abortion before delivery; these women were significantly more likely to have an unmet need for modern PPFP methods than women with no abortion history (Table 3). 1 3

Association of unmet need for PPFP with subsequent pregnancy after childbirth
Since women who have unmet need for PPFP are at a greater risk of unintended pregnancy after childbirth, we compared the rates of subsequent pregnancy during the first 24 months postpartum for women who had unmet need for PPFP and for those who did not. Approximately 12% (2133) of participants reported a pregnancy during the 24 months postpartum period after childbirth. These participants reported that more than 80% of the pregnancies were unintended. As shown in Table 4, the pregnancy rates at 6, 12, and 24 months postpartum were 19.8%, 30.1% and 30.9%, respectively, for women with unmet need for any contraceptive method. All of these levels were significantly higher than the levels for their counterparts during the same time segments (9.8%, 9.8%, and 10.0%). Similarly, pregnancy rates during the 24 months postpartum period were significantly higher for women who had unmet need for modern PPFP method than for those who did not (Table 4).

Unmet need for PPFP and subsequent pregnancy
Because family planning is concerned with both a woman's use of contraception and her fertility preferences, unmet need for family planning can be viewed from a rights-based perspective. Meeting the unmet needs of reproductive aged women represents what is, perhaps, the most critical challenge for FP efforts to attain maximum results with minimal efforts (Bernstein & Edouard, 2007). Estimation of unmet need is a surrogate marker for deficiencies in FP services and, thus, highlights the scope of immediate action (Pasha et al., 2015). The postpartum period is a critical time and should be the target of efforts to address high unmet FP needs and reduce the risks of closely spaced pregnancies (Gaffield et al., 2014). There are already unmet reproductive health care needs in China, and some argue that, in the current situation characterized by low-fertility as the norm, the country needs improved FP programs more than ever. Strengthened FP services can become an integral component of the Healthy China 2030 strategy by ensuring that women meet their reproductive health care needs and, in meeting those needs, realize their reproductive rights as set out in UN SDGs (Gu, 2021). Our study found the rates of unmet need for all FP methods and modern methods at 3 months postpartum were 57.2%, and 63.8%, respectively. These unmet needs declined to 23.9% and 35.5% at 6 months postpartum, 11.8% and 25.6% at 12 months postpartum and finally to 10.6% and 24.6% at 24 months postpartum. However, more than half of the participating women (56%) had resumed sexual activity by the time they reached 3 months postpartum, and 99% of the participants had resumed their sexual activities by 12 months postpartum (Fan et al., 2022). Of the participants who had two or more children, the average birth interval between their two most recently born children was 65 months (Zhang et al., 2022). A recent survey in Tianjin reported a similar birth interval (4.9 years) among postpartum women (Zhang, 2020). An investigation in Jinan showed that a duration of 3 to 5 years is the ideal birth interval for women of reproductive age (Xu et al., 2020). This study found a high level of unmet need for PPFP and a high proportion of postpartum women using less effective traditional contraceptive methods, findings that indicate a substantial number of participants were exposed to the risk of unintended pregnancy during the postpartum period.
There are several reasons that may explain these results. Kang et al. (2018) showed that many health care facilities simply ignored the provision of PPFP services, despite service providers' acknowledgement of the importance of PPFP for improving maternal and infant health. We found in our qualitative study that departments of obstetrics at participating hospitals did not routinely provide FP services. Instead, FP services were provided only by family planning departments, making the provision of PPFP services to women during the perinatal period problematic. Furthermore, there was little or no incentive for obstetricians to provide PPFP services. Finally, postnatal care focuses on the postpartum rehabilitation of mothers and how to take care of babies. PPFP are neglected by both service providers and postpartum women. Approximately 12% of all participants in this study reported an unintended pregnancy within 2 years after childbirth (Che et al., 2021). These findings confirmed yet again the urgent need to improve the quality of PPFP services throughout China. Service providers should counsel women and their partners about the benefits of modern contraceptive methods through the entire childbearing period from conception to postpartum.

Factors influencing unmet need for PPFP
The results of our study showed that less-educated young women, those who had only one child, and those who gave a birth by vaginal delivery were associated with increased risk of unmet need for PPFP. These results were expected because less-educated young women and those who had only one child were less likely than their counterparts to have knowledge of contraceptives and experience in their use. Women who had cesarean delivery were more likely to have been advised by care providers about the risks of untended pregnancy during the postpartum period and encouraged to use contraceptives before or at the time when sexual activities resumed after childbirth. Our study found that women who breastfed exclusively for more than 6 months were more likely to have unmet need for any FP methods; however, exclusive breastfeeding was not associated with unmet need for modern methods. This result indicates that women who breastfed exclusively for more than 6 months were more likely to use traditional methods during the postpartum period. These women may have insufficient knowledge of the lactation amenorrhea method, which is effective for up to 6 months after delivery, as long as monthly bleeding has not returned; it is a complete or almost complete substitute for breastfeeding (Peterson et al., 2000). Women who had a history of abortion before delivery were significantly more likely to have unmet need for modern FP methods; however, this association was not statistically significant for all FP methods. This finding indicates that women with an abortion history were less likely to use modern contraceptive methods than their counterparts; and the fact that these women were less likely to use modern contraceptive methods is the reason why they were at greater risk of induced abortion (Luo et al., 2021). Women who delivered at tertiary hospitals were less likely to have unmet need for all methods, but this association was not the case for modern methods. The quality of technical services at tertiary hospitals is generally considered to be better than at secondary hospitals. The reason for the difference of unmet need for all methods and modern methods between tertiary and secondary hospitals was unclear. We did not identify any previous studies that explored factors influencing the unmet need for PPFP in China. However, factors associated with the use of PPFP methods were explored in some studies. For instance, Kang et al. (2018) reported that postpartum contraceptive utilization was associated with women's education level, breastfeeding status, and fertility intentions for a next child, and delivery method. These findings were largely similar to ours. A number of studies conducted in other countries (Dev et al., 2019;Embafrash & Mekonnen, 2019;Mehata et al., 2014;Thiongo et al., 2022;Wilopo et al., 2017) showed that unmet need for PPFP was associated with geographical region, maternal age, education and occupation, parity, and knowledge of contraceptive methods. Our results agree with most of the studies.

Strength and limitations
This is the first nationally representative study of unmet need for PPFP in China. The levels of unmet need for all contraceptive methods and for modern methods during the first 24 months postpartum were presented. This study provides a snapshot of the status of unmet need for PPFP, and can serve as a valuable aid for policy makers and service providers to develop PPFP service guides and national programs to improve the quality of PPFP services throughout China.
This study has several limitations. We did not gather data on women's birth intentions after delivery, and this means the study may overestimate the level of unmet need for PPFP. We believe the effect of this on estimates is limited, because, as has been demonstrated in previous investigations, most Chinese couples intending to have another child plan a next pregnancy two years after a delivery. In addition, we did not query individual woman about their knowledge of contraceptives or their experience using particular methods. We also did not collect data on the characteristics of male partners to include in data analyses. More research on unmet need for PPFP is needed in China, especially in light of China's introduction of a three-child population policy in 2021.

Conclusion
The levels of unmet need for all FP methods and for modern FP methods were high in China. Postpartum women having unmet need for PPFP were more likely to become pregnant again during the 24 months after childbirth. Less-educated young women who had only one child by vaginal delivery and a history of abortion were more likely to have unmet need for PPFP. National PPFP guidelines that integrate PPFP services into prenatal and postnatal careare urgentlyneeded andshould be implemented throughout the country as soon as possible. PPFP services should promote the use of modern contraceptive methods.