The prevalence and factors associated with premature birth among post-delivery women in the Mbeya region: An analytical cross-sectional study

In Tanzania, there was an increase of prematurity rate from 11.4% in 2014 up to 16.60% in 2016 (1). This is a hospital based analytical cross-sectional study which involved biopsychosocial model, which focused on identifying prevalence and associated factors for preterm births among post-delivery women in Mbeya region, one of Tanzania regions. This study involved hospitals in Mbeya urban, Mbeya rural, Chunya, Kyela, Mbarali, Rungwe, Busokelo and Tukuyu districts, where the prevalence of preterm births in Mbeya found to be 39.1%. The study pointed out that factors associated with preterm births were child spacing of <24months (AOR=3.058; 95% CI = 1.026-9.116: p-value 0.045), non-effective use of malaria prophylaxis during pregnancy (AOR=5.418; 95% CI =1. p-value 0.008), twin pregnancy (AOR=4.657; 95% CI =2.112-10.223, p-value < 0.001), violence during pregnancy (AOR=2.059; p-value 0.048), lack of social support (AOR=1.993; p-value 0.022) and use of pica during pregnancy (AOR=1.880; p-value 0.029). The study outcome revealed that the prevalence of preterm births in Mbeya Region is even higher. Therefore, to minimize or eliminate the problem a deliberate effort to come up with strategies to improve family planning, applications of antimalaria prophylaxis, stop the use of pica and violence during pregnancy was highly recommended.


Introduction
Premature birth has recently gained global attention due to the adverse impact if not well 19 addressed. Of 15million premature births in a year globally, over 84% occur at 32-36 weeks of 20 gestation and categorized as late preterm. About 5% only fall into the extremely preterm birth of . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2022. ; https://doi.org/10.1101/2022.11.14.22282321 doi: medRxiv preprint less than 28 weeks category and the other 10% births occur at 28-32 weeks of gestation and categorized as very preterm (1). 23 Preterm birth rates are estimated to be 11.1% worldwide of which South Asia and Sub-Saharan 24 Africa accounts for 60% of all premature births (2). Low-income countries have an average of 25 12.2% of preterm birth whereas 9.4% and 9.3% correspondingly for middle and higher-income 26 countries (3). Tanzania a sub-Saharan country has been reported to be among the top 10 27 countries with the highest number of preterm babies in which 336,000 babies are born too early 28 each year (2). In Tanzania, preterm is regarded to be the primary factor in one out of every four 29 newborn deaths, while prematurity-related conditions are thought to account for 23% of all 30 newborn deaths (8).

31
The Mbeya Region in Tanzania where this study focused in determining the true prevalence and 32 factors of preterm delivery had 31% of neonatal deaths caused by prematurity (16). Little was 33 known about the prevalence and factors associated with premature birth in this region. Therefore, 34 this study aimed to identify prevalence and factors associated with preterm birth in Mbeya 35 Region, Tanzania.

36
The literature indicates that globally, an estimated 15 million babies are born too early every year 37 which is more than 1 in 10 babies (3). Premature births account for 11.1% of the world's live births, 38 ranging from 5% in European countries and 18% in some of African countries, making majority 39 of about 60% of preterm birth in South Asia and sub-Saharan Africa (2). In the poorest countries, 40 on average, 12% of babies are born premature, compared to 9% in higher-income countries (12).

41
Sub-Saharan Africa is the most affected in southern region with prematurity rate of 11.19% 42 followed by eastern region with prematurity rate of 7.34% then western region with prematurity . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2022. ; https://doi.org/10.1101/2022.11.14.22282321 doi: medRxiv preprint rate of 3.24% and central region with prematurity rate of 2.06% (13). The major possible causes 44 of the problem have been stated to be infections, chronic maternal illness such as hypertension and 45 diabetic mellitus and genetic inspirational, a prior history of preterm birth, underweight, obesity, 46 diabetes, hypertension, smoking, infection, maternal age of either under 17 years or over 40 years, genetics, multi-fetal pregnancy either twins, triplets, or higher and pregnancies spaced too closely 48 together (1).

49
Through this study, the objective was achieved indicating the preterm births in Mbeya region to 50 be 39.1% and associated factors being child spacing, poor application of malaria prophylaxis, twin 51 pregnancies, violence, lack of social support and use of pica during pregnancy.

54
A census method was used to select all district and referral hospitals and a simple random 55 sampling with lottery replacement technique was used to select study participants. A structured 56 questionnaire was used for data collection. Frequencies, percentages, chi-square analysis, and 57 binary logistic regression methods were used for data analysis using SPSS version 25 to compare 58 and explain the prevalence and preterm delivery contributing variables. A variable in the final 59 model was judged statistically significant when it had a 95% Confidence Interval (CI) and a p-  participate. This study did not involve postnatal mothers who had stillbirths and those who did 77 not give consent to participate. Also, postnatal mothers who were extremely ill or whose infants 78 were seriously ill and unsuitable for interviews were not included.

79
Sample size and sampling procedure 80 Estimated minimum sample size (N) for this study was calculated by using the formula;

81
(1) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint  is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

131
The overall prevalence of preterm birth in Mbeya region was found to be 39.1%. This is higher 132 compared to the global rate which is 11.1% and the prevalence of preterm birth in Tanzania 133 which is 16.6%. The elevated prevalence in this study may be due to high proportion of preterm . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2022. ; https://doi.org/10.1101/2022.11.14.22282321 doi: medRxiv preprint birth which was found among participants who were not properly using malarial prophylaxis 135 (58.6%), child space of less than 24 months (66.7%) and twin pregnancy (62.2%) of which were 136 strongly statistically significant associated with preterm birth.

138
The study revealed that the prevalence of preterm births in Mbeya Region is even higher. It was 139 noted that women who are more likely to deliver preterm babies were those with less than 2years 140 child spacing, poor adherence to anti-malaria prophylaxis, those with multiple pregnancies, those 141 who are lacking enough social support, experiencing violence and those who are using pica 142 during pregnancy. To ensure full term delivery, a deliberate effort to come up with strategies to 143 improve family planning use and complete antimalarial prophylaxis is highly recommended.

144
Public health awareness campaigns, family planning to promote ideal pregnancy spacing, and 145 health education to the society will help to minimize/eliminate prematurity and its implications. 146 However, this study methodology was limited because it could not demonstrate a causal link 147 between causes and effects, therefore it was suggested that a cohort study which is the best 148 effective method for establishing the occurrence and natural history of illnesses to be conducted 149 in Mbeya region to help examining numerous outcomes of a single exposure.  . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2022. ; https://doi.org/10.1101/2022.11.14.22282321 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2022. ; https://doi.org/10.1101/2022.11.14.22282321 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint The copyright holder for this this version posted November 18, 2022. ; https://doi.org/10.1101/2022.11.14.22282321 doi: medRxiv preprint