Abstract

Background

In 2012, there were 84.9 million unplanned pregnancies worldwide, with half of these ending in abortion [1]. Unplanned pregnancy can negatively affect both mothers’ and infants’ health [2]. Recent evidence suggests that women with unplanned pregnancies may defer receiving prenatal care, face higher risk of preterm delivery and suffer more obstetric complications [3–5]. In addition, infants born following unplanned pregnancies are more likely to suffer from congenital deformities and are at higher risk of not receiving good prenatal care and not being breastfed [3, 6]. Their birth weight may also be lower than that of planned new-born babies [7].

‘Unplanned pregnancy’ and ‘unintended pregnancy’ are used interchangeably in the literature, and measures of this outcome vary across studies. For example, several studies have used an item that simply asked participants whether their pregnancy was planned or unplanned (intended or unintended). However, research in the United States has estimated the ratio of unintended pregnancies using a timing-based measure from the National Survey of Family Growth, and United Kingdom-based research has estimated the proportion of unplanned pregnancies using the London Measure of Unplanned Pregnancy [8–10]. Because of these differences, direct comparison of the results of published studies is challenging.

To date, there has been little agreement regarding what a ‘planned’ or ‘unplanned’ pregnancy is or how pregnancy intention should be measured and interpreted [11–14]. Previous research has established that operationalising pregnancy intention as a dichotomous variable may be insufficient to assess a woman’s intention regarding pregnancy [11]. Data from several sources have indicted an incongruence between a pregnant woman’s intention and feelings about the current pregnancy [15–18] for example, Trussell et al. highlighted that most women who experienced a pregnancy because of contraceptive failure felt happy or very happy about it [15]. Several lines of evidence have also suggested that women often profess happiness about pregnancies classified as unplanned [9, 19].

Nevertheless, numerous studies have found unplanned pregnancy to be associated with the same key factors: Specifically, these are maternal age [20–27], paternal age [23], low educational attainment [20–23, 25, 26], low socioeconomic status [22, 23, 25], pregnancy history [21–26], employment status [22, 25], smoking status [20, 21, 26, 28, 29], alcohol consumption [21, 26], drug use [20, 21, 26], healthy behaviours [28] and depression [20, 23]. In particular, data from the third National Survey of Sexual Attitudes and Lifestyles [20] in Britain showed that nearly half of unplanned pregnancies (45.2%) occurred among participants aged 16–19years. Reproductive health items (i.e. having first intercourse at age 16years or younger, non-use of contraception at first sex), low educational level, harmful health behaviours (i.e. drug use, current smoking) and current depression were also found to be positively correlated with unplanned pregnancy in that study. An analysis of the 2014 Ghana Demographic and Health Survey reported other factors associated with unplanned pregnancy, including the number of pregnancies, contraception use, marital status, socioeconomic status and occupational status [22].

Published data on unplanned pregnancy in Japan are scarce. A study in Yamagata, Japan, found the prevalence of unplanned pregnancy to be 46.2% and reported associations between unplanned pregnancy and maternal age, age at initiation of sexual intercourse, age at first marriage and spousal age difference [30]. However, these results were based on data from approximately 20years ago and were obtained from participants in cervical and breast cancer screenings. Recently, another article reported that Japanese women still used effective contraceptive methods such as oral contraceptive pills (OCPs) relatively infrequently [31], but no single study has investigated how this affects the rate of unplanned pregnancies. Additionally, to our knowledge, no studies have specifically examined the effects of unplanned pregnancy on individuals and society in Japan.

Preventing unplanned pregnancies and improving maternal health care services are crucial for reducing maternal deaths, which is part of Sustainable Development Goal 3: ‘Reducing the global maternal mortality rate to less than 70 per 100,000 births’ by 2030 [32]. Drawing data from a community-based prospective cohort study, we investigated the prevalence of unplanned pregnancy and associated factors among women living in a rural area of Japan. The study aimed to contribute to the nascent body of research on unplanned pregnancy in Japan and to provide fresh insight into this major reproductive health issue to assist in the development of effective strategies for reducing the number of unplanned conceptions.

Methods

The data used in this study were drawn from Project Koshu, which is an ongoing prospective cohort study for which recruitment began in 1988. Koshu City is a rural area in Yamanashi Prefecture, Japan, with a population of 32,188 people as of 1 October 2014, which represents about 3.8% of the total population of Yamanashi Prefecture (840,139). Expectant mothers were enrolled in Project Koshu at the Koshu City Office.

All participants were informed in writing of the aim of Project Koshu. They then provided their data from the Maternal and Child Health (MCH) Handbook of Japan, along with their signatures. Participation was voluntary. We analysed the data anonymously. This study was approved by the Ethics Review Board of the School of Medicine, University of Yamanashi (approval number: 1398) in accordance with the provisions of the Declaration of Helsinki and the ‘Guidelines Concerning Epidemiological Research’ of Japan. A detailed description of the project has been published elsewhere [33–35].

The Strengthening the Reporting of Observational Studies in Epidemiology guidelines [36] were used to ensure appropriate reporting of our study’s design, conduct and findings (Supplementary Material 1). In the present study, we analysed cross-sectional data on pregnant women participating in Project Koshu from 2011 to 2016. All variables were obtained through a self-report questionnaire included in the MCH Handbook. The MCH Handbook is a national programme in Japan that involves a 49-page booklet in which pregnancy, delivery, child development and health care information is recorded to inform all health care workers and other relevant service providers [37].

Outcome and exposures

Pregnancy intention, which was the primary outcome of our research, was assessed with the question ‘Is this pregnancy planned?’ The response options were: (1) yes; (2) no and (3) I cannot say either way. We categorized this outcome as a dichotomous variable: If a participant answered yes, her pregnancy was considered planned; if she chose response (2) or (3), the pregnancy was considered unplanned.

Despite the lack of a generally accepted definition of pregnancy intention, there is no doubt about the detrimental effects of unplanned pregnancy on maternal and child health. Thus, to avoid inadvertently omitting any such vulnerable pregnancies, we kept respondents who said I cannot say either way in the analysis, classifying them as having unplanned pregnancies.

There were certainly drawbacks to this proposed definition. First, it would have been desirable to have further empirical justification for including the pregnancies of participants who felt unsure about whether the pregnancy was planned in the group of unplanned pregnancies, but the data required for this were not available in our study. Additionally, our use of a single item to assess pregnancy intention was potentially problematic. To address this potential problem, we used a statistical method to assess the robustness of our results, as described in the following subsection.

The independent variables, chosen on the basis of a review of the relevant literature, were maternal age, paternal age, family structure, number of pregnancies (defined as the number of previous pregnancies +1), maternal employment status, smoking status, drinking status, having a balanced diet and currently being depressed. These covariates were obtained through the self-report questionnaire included in the MCH Handbook. The categorization of all covariates is shown in Table 1.

Table 1

Prevalence of unplanned pregnancy and baseline characteristics among 932 participants, 2011–2016

Mean±standard deviation or number (%)
Maternal age, years	31.3±5.2
Paternal age, years	33.4±6.3
Pregnancy intention
Planned	550 (59.0)
Unplanned	382 (41.0)
Family structure
Nuclear family	620 (66.5)
Extended family	286 (30.7)
Other	26 (2.8)
Number of pregnancies
1–2	633 (67.9)
≥3	299 (32.1)
Employment status
Employed	639 (68.6)
Unemployed	293 (31.4)
Smoking status
Never smoked	647 (69.4)
Ex-smoker	244 (26.2)
Current smoker	41 (4.4)
Drinking status
Never drank	479 (51.4)
Ex-drinker	412 (44.2)
Current drinker	41 (4.4)
Balanced diet
No	431 (46.2)
Yes	501 (53.8)
Currently depressed
No	519 (55.7)
Yes	413 (44.3)

Open in a separate window

Results

Table ​Table11 shows the overall characteristics of the 932 pregnant women who completed the questionnaire when they registered their pregnancies at the City Office in Koshu City, Yamanashi Prefecture, Japan, from 2011 to 2016. The mean age of participating women was 31.3years and that of their partners was 33.4years. Few women (2.8%) reported that their family structure was ‘other’ (i.e. something other than a nuclear or extended family). Approximately two-thirds of the participants had been pregnant once or twice (including the current pregnancy), and 68.6% were employed. The majority of the respondents reported healthy lifestyle behaviours, including never smoking (69.4%), never consuming alcohol (51.4%) and having a balanced diet (53.8%). However, 413 women (44.3%) reported that they were currently experiencing depressed mood.

In response to the question, ‘Is this pregnancy planned?’, 59% of the participants chose yes. This meant that 41% of the pregnancies were unplanned (Table ​(Table1).1). In addition, most of the women (85%) felt happy about being pregnant; only 1.2% selected do not feel anything, and 35 participants (3.8%) reported their feelings as other (Supplementary Material 2).

The associations between pregnancy intention and the independent variables were tested using logistic regression analyses. We found no evidence of multicollinearity among the predictors: All VIF values were around 1, except for the values for maternal age and paternal age, which were 2.16 and 2.04, respectively (Supplementary Material 3). Table 2 shows that, in the univariate analyses, there were significant associations between unplanned pregnancy and younger maternal age, paternal age, ‘other’ family structure, three or more pregnancies and currently smoking. Having a balanced diet and currently being depressed were also independent predictors of unplanned pregnancy. We did not find any statistically significant associations between pregnancy intention and women’s employment or drinking status. The multivariable model fitted using a stepwise approach showed that younger maternal age, ‘other’ family structure, three or more pregnancies, currently smoking, having a balanced diet and currently being depressed were factors associated with unplanned pregnancy. These results are summarised in Table ​Table22.

Table 2

Risk factors for unplanned pregnancy among Japanese mothers

	Crude analysis OR (95% CI)	p-value	Multivariable analysis OR (95% CI)	p-value
Maternal age
+1year	0.95 (0.93–0.97)	< 0.001	0.94 (0.92–0.97)	< 0.001
Paternal age
+1year	0.96 (0.94–0.98)	< 0.001	–	–
Family structure
Nuclear family	Ref	–	Ref	–
Extended family	1.11 (0.84–1.48)	0.46	1.01 (0.75–1.37)	0.93
Other	3.47 (1.48–8.10)	0.004	2.76 (1.12–6.76)	0.027
Number of pregnancies
1–2	Ref	–	Ref	–
≥3	1.71 (1.29–2.25)	< 0.001	2.26 (1.66–3.08)	< 0.001
Employment status
Employed	Ref	–	–	–
Unemployed	1.08 (0.82–1.43)	0.58
Smoking status
Never smoked	Ref	–	Ref	–
Ex-smoker	1.13 (0.84–1.53)	0.41	1.06 (0.77–1.45)	0.72
Current smoker	3.81 (1.91–7.61)	< 0.001	2.60 (1.26–5.35)	0.01
Drinking status
Never drank	Ref	–	–	–
Ex-drinker	0.82 (0.62–1.07)	0.14
Current drinker	1.14 (0.6–2.17)	0.68
Balanced diet
No	Ref	–	Ref	–
Yes	0.55 (0.42–0.72)	< 0.001	0.62 (0.47–0.83)	< 0.001
Currently depressed
No	Ref	–	Ref	–
Yes	1.74 (1.34–2.26)	< 0.001	1.63 (1.24–2.16)	< 0.001

Open in a separate window

Multivariable model fitted using stepwise regression. OR odds ratio; CI confidence interval; Ref reference

Table ​Table33 shows the results of the multivariate logistic regression analyses with two different definitions of unplanned pregnancy. Notably, the main analysis and the sensitivity analysis revealed almost the same risk factors for unplanned pregnancy. Therefore, our results can be considered robust.

Table 3

Multivariate analysis results for two definitions of unplanned pregnancy

	Main analysis(a)	p-value	Sensitivity analysis(b)	p-value
Maternal age
+1year	0.94 (0.92–0.97)	< 0.001	0.95 (0.93–0.98)	< 0.001
Family structure
Nuclear family	Ref	–	Ref	–
Extended family	1.01 (0.75–1.37)	0.93	1.03 (0.77–1.38)	0.86
Other	2.76 (1.12–6.76)	0.027	3.03 (1.21–7.60)	0.018
Number of pregnancies
1–2	Ref	–	Ref	–
≥3	2.26 (1.66–3.08)	< 0.001	2.19 (1.62–2.97)	< 0.001
Smoking status
Never smoked	Ref	–	Ref	–
Ex-smoker	1.06 (0.77–1.45)	0.72	1.02 (0.75–1.39)	0.90
Current smoker	2.60 (1.26–5.35)	0.01	2.17 (1.06–4.45)	0.035
Balanced diet
No	Ref	–	Ref	–
Yes	0.62 (0.47–0.83)	< 0.001	0.67 (0.51–0.88)	0.004
Currently depressed
No	Ref	–	Ref	–
Yes	1.63 (1.24–2.16)	< 0.001	1.73 (1.32–2.28)	< 0.001

Open in a separate window

Pregnancy intention is measured using a single item: ‘Is this pregnancy planned?’ in (^a) and by a combination of that item and a variable gauging women’s feelings about the pregnancy in (^b). Details of the definition are described in the Methods section.

The data are presented as odds ratios with 95% confidence intervals. Ref reference

Discussion

Reducing the number of unplanned pregnancies is a critical goal of family planning programmes. To achieve this goal, health policy-makers need to be provided with information about the prevalence of and factors associated with unplanned pregnancy. To obtain that information, we conducted a secondary analysis of baseline data from Project Koshu, a cohort study conducted in a rural area of Japan from 1998 to 2016. Of the 932 women who completed the Project Koshu questionnaire from 2011 to 2016, 41% reported that their pregnancies were unplanned. We found that young women, women who experienced three or more pregnancies (including the current pregnancy), women whose family structure was neither nuclear nor extended, current smokers, women with an unbalanced diet and women who were currently depressed had a higher odds of experiencing unplanned pregnancy.

Conclusions

This study showed that the prevalence of unplanned pregnancy in the study setting was relatively high (41%), demonstrating the need for more attention to this issue. Family planning programmes should be widely and easily accessible to women of reproductive age, especially those at high risk for unplanned pregnancy. In particular, our findings suggest that there is a need for better sex education and increased knowledge about contraceptive methods for young people. In addition, multigravidas and women without family support may need individualised instruction and counselling to increase the effectiveness and consistency of use of contraceptive methods. Finally, mental health care and interventions for health behaviour modification should be implemented from preconception throughout pregnancy to prevent unplanned pregnancy and related negative outcomes.

Supplementary information

Acknowledgements

Abbreviations

Authors’ contributions

HY and STH conceived the study. HY, ZY and STH contributed to the study design. STH analysed the data. STH, SH, RK, YA, OT, HY, RS and ZY interpreted the data and contributed to the discussion. STH wrote the first draft of the manuscript and revised it. ZY was responsible for data integrity. ZY, YA, RK, TO, SH and HY obtained funding. All authors have agreed with the results and conclusion of the study and approved the final version of the manuscript.

Funding

This work was supported by Grants-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan (grant numbers: KAKENHI 18H03037 and JP18K17376), and the Ministry of Health, Labour and Welfare of Japan (grant number: H28-sukoyaka-ippan-001).

The funding sources had no role in the study design, data collection, data analysis, data interpretation or writing of the report.

Availability of data and materials

The data sets analysed in the current study are available from author Zentaro Yamagata ([email protected]) upon request.

Consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

Footnotes

Contributor Information

Son Trung Huynh, Email: [email protected].

Hiroshi Yokomichi, Email: pj.ca.ihsanamay@ihcimokoyh.

Yuka Akiyama, Email: pj.ca.ihsanamay@aakuy.

Reiji Kojima, Email: pj.ca.ihsanamay@ramijok.

Sayaka Horiuchi, Email: moc.liamg@ihcuirohakayas.

Tadao Ooka, Email: pj.ca.ihsanamay@akohot.

Ryoji Shinohara, Email: pj.ca.ihsanamay@arahonihsr.

Zentaro Yamagata, Email: pj.ca.ihsanamay@tgmynez.

References