AMJ Vol 9 No 2 June 2022.indd Althea Medical Journal. 2022;9(2) 61Althea Medical Journal. 2022;9(2):61–69 Trends in Determinants of Birth Interval based on Data from the Indonesian Demographic Health Survey in 2007, 2012, and 2017 Nedra Wati Zaly,1 Mugia Bayu Rahardja,2 Murtiningsih1 1Institute of Health and Technology PKP DKI Jakarta, Jakarta, Indonesia, 2National Research and Innovation Agency, Jakarta, Indonesia Abstract Background: Fertility is the main component of population dynamics that determines the population structure of a country. One method to measure fertility is to analyze the birth interval. This study aimed to identify determinants of birth interval based on the Indonesian Demographic and Health Survey (IDHS) data in 2007, 2012, and 2017. Methods: This study design was descriptive using IDHS data from 2007, 2012, and 2017 with a cross- sectional study design. Samples were mothers who gave birth to their last living child in the last five years. Results: In the IDHS 2007, the birth interval at risk (<24 months) was 11.8%, increased in the 2012 IDHS to 12.1%, and decreased in the 2017 IDHS to 9.8%. In the three-period IDHS, the same determinants of the birth interval were discovered, such as the age of the woman, duration of breastfeeding, use of contraception, the status of the child before the last child, mother’s education level, and wealth quintile. However, in 2007, the sex of the child before the last child and the number of surviving children also influenced the birth interval. In 2017 the number of children who were still alive also affected the birth interval. The most dominant factor influencing the occurrence of short birth intervals after controlling for other variables was the survival status of the previous child. Conclusion: The short birth interval is the most prevalent in the 2012 IDHS. Optimal birth interval has the potential to improve maternal, neonatal, and infant health. Keywords: Birth interval, female reproduction, the health of mother and child Correspondence: Ns. Nedra Wati Zaly, S.Kep., M.Kep, Institute of Health and Technology PKP DKI Jakarta, Jalan Raya PKP, Kelapa Dua Wetan, Ciracas, Jakarta, Indonesia, E-mail: nedrawati12@gmail.com Introduction Indonesia as one of the developing countries in Southeast Asia faces a big challenge, namely an excessive population, which is around 270 million people. Indonesia’s total fertility rate (TFR) has decreased from 2.6 in 2007 to 2.4 in 2017 based on the 2007 and 2017 IDHS data.1,2 Even though it has decreased in 10 years, this is still not in line with the Indonesian government plan. Fertility is one of the factors that play an important role in population growth.3 Fertility is the major component of population dynamics that determines the population structure of a country. Birth interval (time between two consecutive live births) is a more acceptable method of measuring fertility than other methods.4 Delayed marriage, higher education, smaller family, absence of child death experience, and living in food secured households were associated with the small number of children. Fertility was significantly higher among women with no child sex preference.5 Fertility behavior is usually influenced by the interaction of several biological, socioeconomic, and demographic factors. Birth interval is a good index of current fertility changes.3 The use of contraception is one of the factors that affect birth interval which has been supported through government programs. This is made clear by a study in Ethiopia that mentions the importance of contraceptive use factors in influencing birth interval. The short birth interval is still a concern for women due to religious factors, https://doi.org/10.15850/amj.v9n2.2643 Althea Medical Journal. 2022;9(2) 62 length of breastfeeding, unwanted pregnancy, and not using contraception. Increasing access and coverage of contraceptive use as well as involving religious leaders in family planning programs are important strategies that need attention.6 Factors that influence the family in determining the decision to adjust the birth interval include socioeconomic factors, partner psychology, partner age, and cultural factors. However, not all couples of childbearing age know the benefits of long-term pregnancy spacing, so there are still many couples of childbearing age who have children who are too close or too far apart.7 In addition, several studies conclude that the factors that affect birth interval include; gender of the last child, maternal age, length of time breastfeeding before conception and pregnancy, use of contraception, and place of delivery of the last child.8 Birth intervals less than 24 months are considered short and intervals more than 5 years are considered long. Both short and long intervals are considered inadequate.9 Birth interval that is very short or too long directly affects the occurrence of risky pregnancies. Pregnancy with a distance that is too close can cause bleeding, anemia, and premature rupture of membranes, so couples of childbearing age need to be able to plan and determine the spacing of pregnancies.7 Besides being risky for the mother giving birth, the short birth interval is also a risk for the baby. The results of another study in Jambi10 stated that respondents who had a short birth interval had a 1.94 times risk of experiencing low birth weight (LBW) compared to respondents who had an optimal birth interval. On the other hand, the long birth intervals can also affect the occurrence of preeclampsia. The incidence of preeclampsia is more at risk at birth intervals <2 years and birth ≥5 years.11 This study aimed to identify determinants of birth interval based on Indonesian Demographic and Health Survey (IDHS) data in 2007, 2012, and 2017. Methods This study used the data collected in the Indonesian Demographic and Health Survey (IDHS). In 2007, 2012, and 2017 IDHS was conducted by the National Population and Family Planning Board (Badan Kependudukan dan Keluarga Berencana Nasional, BKKBN), Statistics Indonesia (Badan Pusat Statistik, BPS), and the Ministry of Health (Kementerian Althea Medical Journal June 2022 Kesehatan, Kemenkes) with a nationally representative sample. The sample frame of the IDHS was the master sample of Census Blocks (CBs) from the 2000 and 2010 Population Census. The frame for the household sample selection was the updated list of ordinary households in the selected census blocks. The IDHS sample was selected using a stratified, two-stage cluster design and CBs was the sampling unit for the first stage. For the 2017 IDHS, a representative sample of approximately 47,963 households from 1,970 CBs was selected. In the first stage, 1,970 CBs were selected from the list of enumeration areas based on the sampling frame. In the second stage, a complete listing of 25 households was carried out in each selected CBs. The analyzes presented in this study on birth interval in 2007, 2012, and 2017 IDHS was based on the 9,980, 10,703, and 10,446 women respectively who had at least one birth during a five-year period selected from the IDHS data. Variable observed for this study was the age of the woman at the time of delivery of the child before the last child, the sex of the child before the last child, duration of breastfeeding for the previous child, number of surviving children, contraceptive use, the status of the child before the last child, mother’s education level, place of residence, and wealth quintile. The data were analyzed in two phases. First, descriptive analysis was used to provide an overview of birth interval by an independent study. Bivariate analysis using the chi-square test to determine the significance of the relationship between birth interval and the independent variables studied. Second, the likelihood ratio test and Wald test were used in multivariate logistic regression analysis. Multivariate logistic regression analysis of the determinants of the birth interval was performed. The IDHS data collection was downloaded from the DHS website in compliance with the Access Policy (https://microdata.worldbank. org/index.php/catalog/3477). The procedures and questionnaires for IDHS were following the Standard DHS survey protocol approved by the ICF Institutional Review Board (IRB) and conforming with the U.S. Department of Health and Human Services regulations for the Protection of Human Subjects (45 CFR 46) (https://dhsprogram.com/methodology/ P r o t e c t i n g - t h e - P r i v a c y - o f - D H S -S u r v e y - Respondents.cfm). The ethical clearance of the 2017 IDHS has referred to the DHS-7 project which was an ethical review carried out by ICF IRB on March 11, 2015. Althea Medical Journal. 2022;9(2) 63Nedra Wati Zaly et al.: Trends in Determinants of Birth Interval based on Data from the Indonesian Demographic Health Survey in 2007, 2012, and 2017 Results The results showed that in the 2007 IDHS, the risk birth interval (<24 months) was 11.8%, increased in the 2012 IDHS by 12.1%, and decreased in the 2017 IDHS by 9.8%. The independent variables that affect birth interval in the 2007 IDHS, 2012, and 2017 are shown in Table 1. The age of women at risk (age <20 years and >35 years) decreased starting from the 2007 IDHS by 23.9%, in 2012 by 22%, and in 2017 by 18.2%. The gender of the previous child was more male in the 2007 IDHS, 2012, and 2017. Infants who were not breastfed in the 2007 IDHS were 11.6%, in 2012 it was 7.5%, and in 2017 it was 4.7%. The number of living children (>2 children) in the 2007, 2012, and 2017 IDHS were almost the same, namely 50.8%, 49.5%, and 48.9%. Similarly, the survival status of children was almost the same at 94.1%, 94.6%, and 95.7%. The use of contraception fluctuated from the 2007 IDHS, 2012 to the 2017 IDHS from 91.7%, 91%, and 93.6%. Women who did not go to school decreased from the 2007 IDHS, 2012 to 2017 by 4.5%, 3.5%, and 1.7%, respectively. Most of the respondents lived in rural areas in the 2007 IDHS, 2012, and 2017, namely 40.8%, 54.8%, and 51.3%. Furthermore, the wealth quintile fluctuated from 2007 by 23.3%, increased in 2012 to 30.6%, and in 2017 decreased to 27.9%. In the IDHS 2017, the factors influencing the occurrence of birth intervals were the age of women, duration of breastfeeding, number of children born alive, use of contraception, Table 1 Distribution of Research Variables from the IDHS Data for the Year 2007, 2012 and 2017* Variable IDHS 2007 IDHS 2012 IDHS 2017 n % n % n % Birth Interval ≥24 months <24 months 8,807 1,173 88.2 11.8 9,403 1,300 87.9 12.1 9,424 1,022 90.2 9.8 Age of woman at birth the child before the last child 20–35 years <20 and >35 years 7,592 2,388 76.1 23.9 8,350 2,353 78.0 22.0 8,548 1,898 81.8 18.2 Child’s gender before the youngest child Male Female 5,181 4,799 51.9 48.1 5,584 5,119 52.2 47.8 5,434 5,012 52.0 48.0 Breastfeeding duration Not breastfed ≤18 months >18 months 1,153 6,560 2,267 11.6 65.7 22.7 805 8,150 1,748 7.5 76.1 16.3 495 6,113 3,838 4.7 58.5 36.7 Number of children still alive ≤2 children >2 children 4,908 5,072 49.2 50.8 5,408 5,295 50.5 49.5 5,342 5,104 51.1 48.9 Ever used contraceptive Yes No 9,152 828 91.7 8.3 9,737 966 91.0 9.0 9,782 664 93.6 6.4 Child status before the last child Still living Death 9,389 591 94.1 5.9 10,125 578 94.6 5.4 9,999 447 95.7 4.3 Education level No education Primary Secondary Higher 452 4,602 4,299 628 4.5 46.1 43.1 6.3 375 3,849 5,334 1,145 3.5 36.0 49.8 10.7 182 3,097 5,620 1,547 1.7 29.6 53.8 14.8 Place of residence Urban Rural 4,047 5,933 40.6 59.4 4,835 5,868 45.2 54.8 5,092 5,354 48.7 51.3 Household wealth index The poorest Poor Middle Rich The richest 2,324 1,967 1,966 1,866 1,857 23.3 19.7 19.7 18.7 18.6 3,270 2,094 1,917 1,776 1,646 30.6 19.6 17.9 16.6 15.4 2,917 2.040 1,908 1,831 1,750 27.9 19.5 18.3 17.5 16.8 9,980 100.0 10,703 100.0 10,446 100.0 Note: IDHS= Indonesian Demographic and Health Survey, *IDHS secondary data, 2007, 2012, and 2017, processed Althea Medical Journal. 2022;9(2) 64 previous child’s life status, women’s education, and wealth quintile (Table 2). There were 7 factors affected birth interval in the IDHS 2017. Meanwhile, in the 2012 IDHS, the factors affecting the occurrence of birth interval were the age of women, duration of breastfeeding, contraceptive use, previous child’s life status, women’s education, and wealth quintile. There were 6 factors affected birth interval in the 2012 IDHS. In contrast to what happened in the Althea Medical Journal June 2022 Table 2 Relationship between Birth Interval and Observed Independent Variables* Variable IDHS 2007 (n=9,980) p- value IDHS 2012 (n=10,703) p- value IDHS 2017 (n=10,446) p- value≥24 months <24 months ≥24 months <24 months ≥24 months <24 months Age of woman at birth the child before the last child 20–35 years <20 and >35 years 6,767 2,039 89.1 85.4 824 349 10.9 14.6 0.000 7,445 1,958 89.2 83.2 905 395 10.8 16.8 0.000 7,740 1,684 90.5 88.7 808 214 9.5 11.3 0.016 Child’s gender before youngest child Male Female 4,534 4,273 87.5 89.0 647 526 12.5 11.0 0.018 4,902 4,501 87.8 87.9 682 618 12.2 12.1 0.824 4,924 4,500 90.6 89.8 510 512 9.4 10.2 0.154 Breastfeeding duration Not breastfed ≤18 months >18 months 978 5,875 1,953 84.8 89.6 86.2 175 685 314 15.2 10.4 13.8 0.000 684 7,166 1,553 85.0 87.9 88.8 121 984 195 15.0 12.1 11.2 0.019 435 5,475 3,514 87.9 89.6 91.6 60 638 324 12.1 10.4 8.4 0.001 Number of children still alive ≤2 children >2 children 4,289 4,517 87.4 89.1 619 555 12.6 10.9 0.010 4,733 4,670 87.5 88.2 675 625 12.5 11.8 0.283 4,856 4,568 90.9 89.5 486 536 9.1 10.5 0.016 Ever used contraceptive Yes No 8,139 668 88.9 80.6 1013 160 11.1 19.4 0.000 8,667 736 89.0 76.2 1,070 230 11.0 23.8 0.000 8,881 543 90.8 81.8 901 121 9.2 18.2 0.000 Child status before the last- child Still living Death 8,408 398 89.6 67.4 980 193 10.4 32.6 0.000 9,011 392 89.0 67.8 1,114 186 11.0 32.2 0.000 9,100 324 91.0 72.5 899 123 9.0 27.5 0.000 Education level No education Primary Secondary Higher 394 4,148 3,757 508 87.3 90.1 87.4 80.9 58 454 542 120 12.7 9.9 12.6 19.1 0.000 298 3,452 4,710 943 79.5 89.7 88.3 82.4 77 397 624 202 20.5 10.3 11.7 17.6 0.000 150 2,861 5,089 1,324 82.4 92.4 90.6 85.6 32 236 531 223 17.6 7.6 9.4 14.4 0.000 Place of residence Urban Rural 3,578 5,229 88.4 88.1 469 704 11.6 11.9 0.673 4,262 5,141 88.1 87.6 573 727 11.9 12.4 0.396 4,600 4,824 90.3 90.1 492 530 9.7 9.9 0.684 Household wealth index The poorest Poor Middle Rich The richest 1,969 1,734 1,764 1,695 1,644 84.7 88.2 89.7 90.9 88.5 355 233 202 171 213 15.3 11.8 10.3 9.1 11.5 0.000 2,750 1,863 1,719 1,602 1,469 84.1 89.0 89.7 90.2 89.2 520 231 198 174 177 15.9 11.0 10.3 9.8 10.8 0.000 2,528 1,866 1,775 1,670 1,585 86.7 91.5 93.0 91.2 90.6 389 174 133 161 165 13.3 8.5 7.0 8.8 9.4 0.000 Total 8,807 88.2 1,173 11.8 9,403 87.9 1,300 12.1 9,424 90.2 1,022 9.8 Note: IDHS= Indonesian Demographic and Health Survey, *IDHS secondary data, 2007, 2012, and 2017, processed Althea Medical Journal. 2022;9(2) 65 2007 IDHS, those that affected the occurrence of birth intervals were the age of the woman, the gender of the child, the number of children born alive, the duration of breastfeeding, the use of contraception, the previous life status of the child, women’s education, and wealth quintile. There were 8 factors affected birth interval in the 2007 IDHS. In the 2007–2017 range in the IDHS data for 2007, 2012, and 2017, the factors that influenced birth interval were the woman’s age, duration of breastfeeding, contraceptive use, previous child’s life status, education level, and household wealth index. The number of children born alive (parity) was significantly affected by the birth interval in the 2007 and 2017 IDHS, while the child’s gender factor significantly affected on birth interval only in the 2007 IDHS. The results of the multivariate analysis of the factors that influenced short birth interval can be seen in Table 3. The factor that did not affect the birth interval in the 3 IDHS periods was the gender of the previous child. In the 2007 IDHS, the factors that significantly influenced birth interval were the woman’s age, duration of breastfeeding, contraceptive use, previous child’s life status, education Table 3 Odds Ratio Based on Binary Logistic Regression for the Effect of Observed Independent Variables on a Birth Interval, IDHS 2007, 2012 and 2017* Variable IDHS 2007 IDHS 2012 IDHS 2017 OR p-value CI 95% OR p-value CI 95% OR p-value CI 95% Age of woman at birth the child before the last child 20–35 years <20 and >35 years 1.27 1.00 0.013 1.05–1.53 0.89 1.00 0.194 0.76–1.06 0.77 1.00 0.002 0.65–0.91 Child’s gender before the youngest child Male Female 1.13 1.00 0.054 0.99–1.28 0.97 1.00 0.596 0.86–1.09 0.89 1.00 0.897 0.79–1.02 Breastfeeding duration Not breastfed ≤18 months >18 months 1.16 1.24 1.00 0.041 0.000 1.04–1.37 1.10–1.38 1.55 1.11 1.00 0.001 0.207 1.21–1.99 0.94–1.32 1.37 1.21 1.00 0.043 0.010 1.01–1.85 1.05–1.39 Number of children still alive ≤2 children >2 children 1.08 1.00 0.283 0.94–1.23 0.95 1.00 0.407 0.84–1.07 0.73 1.00 0.000 0.64–0.84 Ever used contraceptive Yes No 0.55 1.00 0.000 0.45–0.67 0.48 1.00 0.000 0.40–0.57 0.57 1.00 0.000 0.45–0.70 Child status before the last child Still living Death 0.21 1.00 0.000 0.16–0.27 0.28 1.00 0.000 0.23–0.36 0.25 1.00 0.000 0.20–0.32 Education level No education Primary Secondary Higher 0.31 0.30 0.51 1.00 0.000 0.000 0.000 0.21–0.46 0.24–0.39 0.41–0.65 0.45 0.32 0.49 1.00 0.000 0.000 0.000 0.32–0.64 0.26–0.39 0.40–0.59 0.54 0.30 0.50 1.00 0.007 0.000 0.000 0.35–0.85 0.24–0.38 0.42–0.60 Place of residence Urban Rural 1.15 1.00 0.075 0.98–1.34 1.21 1.00 0.009 1.05–1.39 1.24 1.00 0.006 1.06–1.45 Household wealth index The poorest Poor Middle Rich The richest 2.28 1.69 1.31 0.99 1.00 0.000 0.000 0.019 0.957 1.78–2.90 1.33–2.14 1.05–1.65 0.79–1.25 2.36 1.52 1.30 1.09 1.00 0.000 0.000 0.026 0.468 1.87–2.97 1.21–1.93 1.03–1.64 0.87–1.37 2.42 1.35 1.01 1.15 1.00 0.000 0.019 0.943 0.246 1.89–3.09 1.05–1.73 0.78–1.30 0.91–1.46 Note: IDHS= Indonesian Demographic and Health Survey, *IDHS secondary data, 2007, 2012, and 2017, processed Nedra Wati Zaly et al.: Trends in Determinants of Birth Interval based on Data from the Indonesian Demographic Health Survey in 2007, 2012, and 2017 Althea Medical Journal. 2022;9(2) 66 level, and household wealth index. In contrast to the 2012 IDHS, the factors that significantly influenced birth interval were duration of breastfeeding, contraceptive use, previous child’s life status, education level, place of residence, and household wealth index. Meanwhile, in the 2017 IDHS, the factors that significantly influenced birth interval were the woman’s age, duration of breastfeeding, contraceptive use, previous child’s life status, education level, place of residence, and household wealth index. Discussion Birth interval is the period between the previous child’s birth date and the last child’s birth date.12 Optimal birth interval is beneficial for the health of the mother and baby for the better. Short birth interval (space between births less than 24 months) can be detrimental to the health of mother and child.13 The ideal birth spacing is more than 24 months. Reproductive organs can prepare to get pregnant again at that time.14 Therefore, it is very important to pay attention to the distance because it is closely related to the reproductive health of women and the health of children born. The risk of low birth weight (LBW) babies will also be reduced by keeping the distance between births.8 A healthy reproductive age range is a safe age for women to experience pregnancy and childbirth, which is 20 to 35 years.15 Meanwhile, age <20 years and >35 years are the ages that are at risk for women to experience pregnancy and childbirth. This is because, after <20 years, a woman’s uterus and pelvis have not yet grown to their adult size. Likewise, at the age of >35 years, maternal diseases such as hypertension, diabetes mellitus, obesity, and aging of the uterine organs (cell degeneration) are easy to occur. All of these are things that increase the risk for women who experience pregnancy and childbirth at that age.16 The age of the mother referred to in this study is the age of the mother at the time of giving birth to the child before the last child. This is because this study used secondary data from the IDHS, so the birth spacing data were obtained by comparing the data of the last child with the previous child. The results of the bivariate analysis in this study found that maternal age significantly affected birth intervals in the 2007 IDHS, 2012, and 2017. Meanwhile, the results of multivariate analysis showed that women aged 20–35 years had a higher chance of experiencing short birth spacing than women aged <20 years and >35 years old. The effect of women’s age with short birth interval is probably due to the current trends of marriage. The delay in marriage causes the reproductive age to be shorter. Therefore, women who marry late will speed up the birth gap. This is confirmed by research which explains that women who experience social pressure due to late marriage will try to accelerate the birth rate (by shortening the birth distance).16 This is also in line with research in Ethiopia which states that women who marry late are closely related to the birth interval. Women who marry late are getting older, so they should have children immediately after marriage and plan to have the desired number of children before the end of their reproductive period.17 Our study has shown that the duration of breastfeeding affects birth intervals. Women who do not breastfeed have a higher risk of experiencing shorter birth intervals compared to women who breastfeed >18 months (OR:1.16; OR:1.55; OR 1.37). This proves that the duration of breastfeeding in the previous child shows a relationship with a short birth interval. Three studies found an association between shorter birth intervals and shorter breastfeeding duration. One study found shorter birth intervals without breastfeeding compared to exclusive or mixed breastfeeding.18 In line with research in Iran19 that the duration of breastfeeding is an independent factor of birth interval. This shows that breastfeeding duration of more than 2 years is more likely to have a longer birth interval than a breastfeeding duration of less than 6 months.19 It can be concluded that breastfeeding duration can prolong the birth spacing period by slowing down female fertility. Couples need to set the ideal pregnancy distance, which is >2 years (Qur’an Surah Al Baqoroh verse 233), and thus, a mother should breastfeed her children for 2 full years because breastfeeding is a natural way to space out pregnancy up to 2 years. The T-interval that is too close results in the inability of families to take good care of their children, it is hoped that mothers will breastfeed their babies for up to 18 to 24 months so that the baby’s nutrition is fulfilled.20 Birth interval is also affected by contraceptive use. Couples who do not use contraception have a higher chance of experiencing shorter birth intervals than couples who use contraception (OR 1.82 in the IDHS year 2007; OR 2.08 in the IDHS Althea Medical Journal June 2022 Althea Medical Journal. 2022;9(2) 67 year 2012; OR 1.75 in the IDHS year 2017). This is in line with a study in Ethiopia6 that stated the importance of contraceptive use factors in influencing birth interval. Short birth interval is still a concern for Ethiopian women due to factors such as religion, length of breastfeeding, unwanted pregnancies and not using contraception.6 Important strategies that need to be considered are the coverage of contraceptive use, increasing access, and involvement of religious leaders in family planning programs.6 In addition, the factor that most influences birth interval is the use of contraceptive methods.12 The results of another study also stated that women who did not use any contraceptive method were more likely to have shorter birth intervals than women who used all forms of contraception.21 The survival status of children born previously also affects the occurrence of birth intervals. The short birth interval was 4 times more likely to occur in women who had a previous child death than in women whose children had previously lived (OR 4.76 in the IDHS year 2007, OR 3.57 in the IDHS year 2012, and OR 4 in the IDHS year 2017). This shows that the previous child’s survival factor is the most dominant influencing birth interval. When a baby/child dies, the mother will stop breastfeeding (unless the children are twins), and the mother will be at risk of getting pregnant more quickly than if she still has babies.22 This is also to the research that says that short birth interval tends to occur. in subsequent births in couples who experience child death. Spouses tend to want to make efforts to have children immediately to replace lost children.19 The results of the study prove that birth interval can be influenced by a woman’s education level. Short birth interval is more at risk for women who graduate from college than women who do not go to school. In 2007, the possibility of short births in women who graduated from college was 3 times (AOR= 3.22) than women who did not attend school. Furthermore, it decreased in 2012 the risk of 2 times (AOR = 2.22) and decreased again in 2017 the risk of 1 time (AOR= 1.85) experienced short birth interval in women who graduated from college compared to women who did not attend school. The short birth interval that occurs in college graduates is probably due to a delay in marriage. Higher education women tend to marry late, limiting their reproductive years and the number of children.21 This is not in line with research related to women’s education in Bangladesh19 which states that women without formal education tend to have a short birth in subsequent births. Likewise, research in Ethiopia12 shows that education level affects short birth intervals. The short birth interval is 3 times more likely to occur in women without formal education compared to women with formal education.12 Women with low formal education are at risk for experiencing short birth intervals. Mothers’ low education is related to their lack of information and knowledge about various things including how to maintain reproductive health. This is also explained in another study which stated that low formal education affects women’s understanding and/or knowledge about contraception and birth interval as well as behavior seeking information on reproductive health.16 Unlike what happened in Indonesia, it was found that women who did not attend school were at risk of this to happen. The short birth interval was lower than female university graduates (AOR=0.31; AOR=0.45; AOR=0.54). The possibility of this happening could be due to the successful family planning program carried out by the Indonesian government. This can be seen from the increasing percentage of women aged 15–49 years who use family planning tools/methods from the 1991–2017 IDHS.2 In addition, half of the women have been exposed to information about family planning by the media.23 This study found that women from the lowest wealth quintile are twice as likely to experience short birth intervals as women from the top wealth quintile. This statement is in line with other studies which state that women with low-income levels have shorter pregnancy intervals than women with higher education. This is related to women with low incomes choosing not to use contraception so that unwanted pregnancies occur.24 The results of this study indicate that the most dominant factor influencing the occurrence of short birth intervals after controlling for other variables is the survival status of the previous child. The risk of short- distance births is 4 times for women who experience child death compared to women whose previous children are still alive. This is in line with research in Manipur, India25 which showed that the child’s previous survival status was the most important in determining the pattern of child spacing for social and biological reasons. Socially, couples who have experienced the loss of a child as an infant avoid contraception with the motivation to Nedra Wati Zaly et al.: Trends in Determinants of Birth Interval based on Data from the Indonesian Demographic Health Survey in 2007, 2012, and 2017 Althea Medical Journal. 2022;9(2) 68 have another child as a substitute. Biologically, infant mortality interferes with breastfeeding which contributes to the return of fertility, and the absence of contraception can increase the likelihood of subsequent early conception.25 This study also provides evidence that child mortality will have an impact on shorter delivery intervals for subsequent children. Child mortality also affects other factors, namely breastfeeding and contraceptive use. Women who will stop breastfeeding will be fertile again soon and are at risk of pregnancy if they do not use contraception. This is because lactational amenorrhea arising from breastfeeding can prolong the birth interval.25 The relationship between child mortality and the birth interval is more related to maternal behavior, such as discontinuation of contraceptive use and cessation of breastfeeding. Mothers who stop breastfeeding result the mother fertility returning quickly.22 In general, the increase in infant mortality rate can be caused by babies born with short birth intervals, maternal age who are too young or too old, or babies born to mothers with high parity.2 Maternal age and high parity are at risk of death in children. Due to the death of a previous child, couples tend to try to get pregnant again to replace immediately the child who has died. This results in a short birth interval with the occurrence of child deaths.19 A mother’s education affects the occurrence of short birth spacing. Mothers without formal education tend to have a short birth interval with subsequent births.19 When a child dies in a mother without formal education, the mother will likely try as soon as possible to be able to obtain a substitute child. This incident can be caused by the lack of understanding of mothers about reproductive health. In contrast, mothers with college graduates are likely to experience short birth intervals due to delays in marriage so that couples will shorten their reproductive period. This statement is in line with research that states that study-oriented factors are one of the causes of women getting married late.26 When a baby or child dies, it is very likely that mothers with college graduates will immediately plan to have another child due to their limited reproductive age. This study has several limitatations, since the study used secondary data, researchers had limited direct access to respondents, and researchers can only process existing data. The sample in this study were married women who had at least 2 children, and did not exclude women who had abortions or miscarriages, so there might be an underrepresentation of birth interval. In conclusion, the largest percentage of short birth interval is in the 2012 IDHS data (12.1%). In the 2007 IDHS, the factors that significantly influence birth interval are the woman’s age, duration of breastfeeding, contraceptive use, previous child’s life status, education level, and household wealth index. In contrast to the 2012 IDHS, factors that have a significant effect on birth spacing are duration of breastfeeding, contraceptive use, previous child’s life status, education level, place of residence, and household wealth index. Meanwhile, in the 2017 IDHS, the factors that significantly affect birth interval are the woman’s age, duration of breastfeeding, contraceptive use, previous child’s life status, education level, place of residence, and household wealth index. Optimal birth interval has the potential to improve maternal, neonatal, and infant health. Birth interval is expected to be a basic priority for policymakers to improve maternal and child health. Child mortality is the most dominant factor influencing the occurrence of short birth intervals after being controlled by other variables. Efforts that can be made to improve maternal and child health are to develope strategies to reduce neonatal, infant, child, and perinatal mortality. This study recommends increasing awareness about optimal birth interval and increasing the use of modern contraception, improving maternal and child health services, and providing health education to mothers regarding the care of babies to toddlers who are still very susceptible to illness. References 1. National Population and Family Planning Board (BKKBN), Statistics Indonesia (BPS), Ministry of Health (Kemenkes), ICF. Indonesia demographic and health survey 2012. Jakarta, Indonesia: BKKBN, BPS, Kemenkes, and ICF; 2013. 2. National Population and Family Planning Board (BKKBN), Statistics Indonesia (BPS), Ministry of Health (Kemenkes), ICF. Indonesia demographic and health survey 2017. Jakarta, Indonesia: BKKBN, BPS, Kemenkes, and ICF; 2018. 3. Singh, SA. Human fertility behavior through birth interval models: overview. Am J Theor Appl Stat. 2016;5(3):132–7. 4. Nath DC, Leonetti DL, Steele MS. Analysis of birth intervals in a non-contracepting Indian population: an evolutionary ecological approach. J Biosoc Sci. Althea Medical Journal June 2022 Althea Medical Journal. 2022;9(2) 69 2000;32(3):343–54. 5. Mekonnen W, Worku A. Determinants of fertility in rural Ethiopia: The case of Butajira Demographic Surveillance System (DSS). BMC Public Health. 2011;11:782. 6. Gebrehiwot SW, Abera G, Tesfay K, Tilahun W. Short birth interval and associated factors among women of childbearing age in northern Ethiopia, 2016. BMC Women’s Health. 2019;19(1):1–9. 7. Laili U, Masruroh N. Penentuan jarak kehamilan pada pasangan usia subur. J Kesehat Al-Irsyad. 2018;XI(2):52–7. 8. Ejigu AG, Yismaw AE, Limenih MA. The effect of sex of the last child on short birth interval practice: The case of northern Ethiopian pregnant women. BMC Res Notes. 2019;12(1):4–9. 9. Barbosa R, Alves MTSSB, Nathasje I, Chagas D, Simões VF, Silva L. Factors associated with inadequate birth intervals in the BRISA Birth Cohort, Brazil. Rev Bras Ginecol Obstet. 2020;42(2):67–73. 10. Kurniasari L. The correlation of parity, birth spacing and preeclampsia history toward LBW incidence at RSIA Annisa Jambi in 2017. Sci J. 2018;7(1):53–7. 11. Fatkhiyah N, Khodiyah M. Determinan maternal kejadian preeklampsia (Studi kasus di kabupaten Tegal, Jawa Tengah). J Keperawatan Soedirman. 2016;11(1):53– 61. 12. Hailu D, Gulte T. Determinants of short interbirth interval among reproductive age mothers in Arba Minch District, Ethiopia. Int J Reprod Med. 2016;2016:1–17 13. Ajayi AI, Somefun OD. Patterns and determinants of short and long birth intervals among women in selected sub- Saharan African countries. Medicine (Baltimore). 2020;99(19):e20118. 14. Monita F, Suhaimi D, Ernalia Y. Hubungan usia, jarak kelahiran, dan kadar hemoglobin ibu hamil dengan kejadian berat bayi lahir rendah di RSUD Arifin Achmad Provinsi Riau. Jom FK. 2016;3:1–5. 15. Dorland, Newman. Kamus Kedokteran Dorland. 31st ed. Jakarta: EGC; 2010. 16. Muluneh AA, Kassa ZY, Siyoum M, Gebretsadik A, Woldeyes Y, Tenaw Z. Determinants of sub-optimal birth spacing in Gedeo Zone, South Ethiopia: a case-control study. Int J Women’s Health. 2020;12:549–56 17. Chernet AG, Shebeshi DS, Banbeta A. Determinant of time-to-first birth interval after marriage among Ethiopian women. BMC Women’s Health. 2019;19(1):1–6. 18. Pimentel J, Ansari U, Omer K, Gidado Y, Baba MC, Andersson N, et al. Factors associated with short birth interval in low- and middle-income countries: a systematic review. BMC Pregnancy Childbirth. 2020;20(1):1–17. 19. Khan JR, Bari W, Latif AHMM. Trend of determinants of birth interval dynamics in Bangladesh. BMC Public Health. 2016;16(1):1–11. 20. Adriani M. Pengantar gizi masyarakat. 6th Ed. Jakarta: Kencana Indonesia; 2016. 21. Nausheen S, Bhura M, Hackett K, Hussain I, Shaikh Z, Rizvi A, et al. Determinants of short birth intervals among married women: a cross-sectional study in Karachi, Pakistan. BMJ Open. 2021;11(4):1–10. 22. Montgomery, MR. Cohen B. From death to birth: mortality decline and reproductive change. Washington, D.C.: National Academy Press; 1998. 23. Arsyad SS. Determinan fertilitas di Indonesia. J Kependud Indones. 2016;11(1):1–14. 24. Wulandari DF, History A, Wulandari DF, Malang K, Java E. Literature review: factors influence birth interval from couples in childbearing age. Matern Neonat Health J. 2021;0825(2016):20–36. 25. Singh SN, Singh N, Narendra R. Demographic and sosio-economic determinants of birth interval dynamics in Manipur: a survival analysis. Online J Health Allied Sci. 2010;9(4):3. 26. Maqhfirah D. Makna hidup pada wanita yang terlambat menikah. J Divers. 2018;4(2):109–18. Nedra Wati Zaly et al.: Trends in Determinants of Birth Interval based on Data from the Indonesian Demographic Health Survey in 2007, 2012, and 2017