Knowledge level and constructs of the theory of planned behaviour (TPB) to the practice of unsafe abortion among postnatal mothers attending Mkonze health center, Dodoma Region, Tanzania

Background Unsafe abortion is now a global agenda because 45% of all global abortions are unsafe, and 97% are occurring in developing countries. In Tanzania, one million reproductive-aged women face unplanned pregnancies per year, and 39% end up with abortion. About 16% of maternal deaths are reported per year in Tanzania, and unsafe abortion takes the second position. There are several efforts to prevent and intervene unsafe abortions, such as equipping healthcare facilities across all levels of healthcare, approval of Misoprostol use, establishment of comprehensive post-abortion care (PAC), revising policy guidelines and standards, provision of emergency contraceptives, and capacity building of healthcare providers. There is little documentation about how the constructs of the theory of planned behaviour, knowledge, and sociodemographics influence the practice of abortion. Objectives To assess the association of knowledge level, sociodemographic characteristics, and constructs of the theory of planned behaviour (TPB) to the practice of unsafe abortion among postnatal mothers at Mkonze Health Center in the Dodoma region. Methodology It is an analytical cross-sectional study design conducted in Dodoma-Tanzania and involved 206 postnatal women. A validated questionnaire was used and analysis was performed in the Statistical Package for the Social Sciences (SPSS), through descriptive and inferential statistics. Results The practice of unsafe abortion in the current study is 28/206 (13.6%), influenced bytheir lower educational level and being single women. It was found that the majority had adequate knowledge of unsafe abortion 129 (62.6%), positive attitude 130 (63.1%), good subjective norms 113 (54.9%), and positive perceived behavioral control111 (53.9%). Knowledge, attitude, subjective norms, and perceived behavioral control were not significantly associated with the practice of unsafe abortion. Conclusion The majority of the respondents had high knowledge, attitudes, subjective norms, and perceived behavior control on unsafe abortion. This is an indicator that the implemented initiatives are effective. Maintaining the ongoing effort and improving strategies are promising to mitigate the burden of unsafe abortion. Future research needs to find out hidden factors associated with attitude and how health beliefs might influence someone’s attitude towards unsafe abortion. Supplementary Information The online version contains supplementary material available at 10.1186/s12889-024-18921-z.


Introduction
Abortion is a global issue that has been taking the attention of different cultures, religions, and social, psychological, and country laws [1].Abortion has become a major issue across the country's boundaries whereby some countries' policies accept it in special scenarios such as life-threatening conditions and emergencies while others such as United States and Canada allow women to abort depending on their reasons such as medical, psychology, rape, financial hardship, incest, and fetal abnormalities [2].Out of 210 million pregnancies per year, 46 million pregnancies end in abortion [3].
Unsafe abortion is now a global agenda because 45% of all global abortions are unsafe, with 97% occurring in developing countries and 3% in developed countries [4].Out of 55.7 million abortions that occurred globally between 2010 and 2014, 25.1 million of these were unsafe every year [5].It is approximated that there are 20 million unsafe abortions annually and one out of 250 end to death and maternal disabilities [6].This is supported by another study reporting that 13% of unsafe abortions end in maternal death each year [7].While a huge burden of unsafe abortion is in Sub-Saharan countries, East Africa accounts for 75% of unsafe abortions [8].
Different reasons contribute to the decision of unsafe abortion among many women and adolescent girls, including unpreparedness for pregnancy, poor access to contraceptives, education level, and sexual education, economic issue, rape, marital status, victim of incest, support from partner and parents, society norms, religion, stigma, health care system and laws of the countries [9].All these have contributed to unsafe abortion among women in countries where there is no accessibility of safe abortion due to restrictive laws of abortion [10].The rate of unsafe abortion is increasing proportionally especially in restrictive countries due to the influence of demographics, low community awareness, socioeconomic and cultural factors such as financial issues, and peer groups, to continue their education, secure their future aspiration, relationship status, stigma, risk measuring, and availability and accessibility of methods of abortion such as misoprostol [11].According to research done in Zambia, 77 women out of 362 had unsafe abortions whereby most of them fall into the group of unmarried, living in high density, having a history of unplanned pregnancy and a history of miscarriage [12].Generally, the higher proportions of unsafe abortion are confined to countries with highly restrictive abortion laws [13,14], and countries with difficulty accessing health facilities [15].Abortion is the termination or expulsion of the fetus before 20 weeks of gestation [16].Abortion can be spontaneous or induced, whereby it is induced abortion when there is termination of unwanted pregnancy by use of external methods (medically or by dilation and curettage) [17].On the other side, spontaneous abortion refers to a miscarriage or purely accidental abortion [18].When the abortion is carried out using a method recommended by WHO, appropriate to the pregnancy duration, and by someone with the necessary skills, it is considered a safe abortion [19].Meanwhile, unsafe abortion refers to a procedure of pregnancy termination either by persons lacking the necessary skills or in an environment that does not conform to minimal medical standards or both [7].
In Tanzania, one million reproductive-aged women face unplanned pregnancies per year, and 39% end up with abortion [20].About 16% of maternal deaths are reported per year in Tanzania, and unsafe abortion takes the second position [21].Several efforts are observed in Tanzania to combat unsafe abortion by improving postabortion care and provision of emergency contraceptives and safe abortion services to women underwent incest and rape [8].In addition, the efforts of tackling unsafe abortion in the country have been through identification and prioritizing unsafe abortion services to zones with the highest rates of abortion, promotion of men's involvement in family planning, equipping healthcare facilities across all levels of the healthcare system to provide basic post-abortion care, implementation of national Road Map Strategic Plan, approval of Misoprostol, and building on a comprehensive post-abortion care (PAC) training program [22].Moreover, there have been new and revised policy guidelines and standards to guide the extension of post-abortion care, capacity building of healthcare providers, developing networks, and platforms to promote understanding of abortion issues [23].
It is reported that one-third of hospital admissions due to pregnancy complications are due to unsafe abortion while one-quarter lead to maternal death [20].Unsafe abortion in Tanzania is due to unintended pregnancy and inaccessibility of contraceptives [24].Most studies in Tanzania regarding abortion have focused on incidence, abortion-seeking practices, and service delivery [21].There is little documentation about how the constructs of the theory of planned behavior and knowledge influence abortion practice.Therefore, the study aimed to assess the association of knowledge level, sociodemographic characteristics, and constructs of the theory of planned behaviour (TPB) to the practice of unsafe abortion among postnatal mothers at the Mkonze health center in Dodoma region.
The theory of planned behavior (TPB) which is a cognitive theory by Azjen (1985), states that an individual's decision to engage in a specific behavior can be predicated by their intention to engage in that behaviour [25].The theory has four constructs; attitude, subjective norms, perceived behavior control, and intention [26].The specific objectives of this study are; ① To determine the knowledge about unsafe abortion among postnatal mothers at Mkonze health center in Dodoma regional ② To assess the attitude towards unsafe abortion among postnatal mothers at Mkonze health center in Dodoma regional ③ To predict subjective norms leading to unsafe abortion among postnatal mothers at Mkonze health center in Dodoma regional ④ To determine perceived behavior control about unsafe abortion among postnatal mother at Mkonze health center in Dodoma regional and ⑤ To evaluate the intention practice towards unsafe abortion among postnatal mother at Mkonze health center in Dodoma regional.

Study design
An analytical cross-sectional study design was employed to use the construsts of the theory of planned behavior and knowledge to assess the intentional practice of unsafe abortion among postnatal mothers at the Mkonze health center in Dodoma region.

Study area
The study was carried out in Dodoma City at Mkonze Health Center.Dodoma city was chosen as the study area because it has the highest magnitude of teenage pregnancies (29-39%) compared to other regions of Tanzania [27,28].Moreover, it is a marked region with relatively low utilization of contraceptives [29].

Study population
The study population included all postnatal women attending or admitted to Mkonze Health Centre.Participants were excluded if had sickness to impair their responses, could not comprehend the instructions, and were not willing to participate because of self-reported tiredness and busy.

Sample size calculation
The sample size was calculated using the sample size formula: n = (Z² * P * (1 -P)) / E².Where: n = Sample size, Z = Standard normal deviation (for a 95% confidence interval, Z = 1.96),P = Proportion of target population.The prevalence of 16% for unsafe abortion in Tanzania [22].E = Marginal error on the quantity to be estimated (5%).Therefore, the calculated sample size was 206.

Sampling technique
Probability sampling especially systematic random sampling was used in the study.The postnatal ward at the health center had a daily recorded number of admitted postnatal women in the register.The required number of participants per day was 6 postnatal women, therefore, the nth number was calculated by dividing the number of required postnatal women over the total admitted number of postnatal women of a particular day.The participants were then selected at the interval of the nth number in the register.

Data collection procedure and data collection tools
Data were collected from Mkonze Health Centre from 25th May to 30th June 2023 by a principal investigator with an assistant.The questionnaire which was used in the data collection was developed and validated by a principal investigator.The development started with a literature review to identify items for each variable, followed by content validity where the experts were invited to rate how well was the tool.The tool was pre-tested to 20% of the actual sample size, and internal consistent reliability was determined.Through principal component analysis (PCA), the weak items were modified, making the final tool to have consisted six parts, (1) socio-demographic information, (2) attitude (3) practice (4) knowledge (5) perceived behavior control, and (6) subjective norms among postnatal mothers.Different parts contained varying numbers of items related to the study's context.For instance, the practice toward unsafe abortion among postnatal mothers was assessed using 8 questions, with response 1 for Yes and 2 for No, the knowledge toward unsafe abortion among postnatal mothers was assessed using 13 questions having responses of 1-Yes and 2-No.Attitude toward unsafe abortion among postnatal mothers was assessed using five items on a 5-Likert scale (1-strongly disagree,2-disagree, 3-neutral, 4-agree, 5-strongly agree).The higher the score (strongly agree) the more positive attitude of the participants.Subjective norms toward unsafe abortion among postnatal mothers were assessed using five items having a response of 1 for Yes and 2 for No. Participants who responded as "Yes" in each item, were considered to get the correct answer.Perceived behavior control toward unsafe abortion among postnatal mothers was assessed using three items 5-Likert scale (1-strongly disagree, 2-disagree, 3-neutral, 4-agree, and 5-strongly agree).The increased number indicates a correct answer, with the maximum (strongly agree).
The questionnaire was translated from English to Swahili's native language for the convenience of participants.

Data analysis
Regarding measures and data processing, knowledge, subjective norms, and practices of unsafe abortion were measured through binary responses of Yes/No, but attitude and perceived behavioral control were measured through a 5-Likert scale.In each variable, the average of items were computed and the scores were categorized.The cut-off point for each variable was 50%, the above 50%, the higher score, and vice versa.Data were entered and analyzed in the Statistical Package for the Social Sciences.Descriptive statistics summarize the data in the form of frequency, percentage, and mean.A cross-tabulation was carried out to look association between sociodemographic data and factors.A P-value of less than 0.05 was considered statistically significant.

The practice of unsafe abortion among the participants
The majority of the participants about two-thirds seemed not to have had an abortion in the course of their lives 161 (78.2%) and those who had abortions were about one-third 45 (21.8%) of the total participants.Of those who had abortions, 17 (38%) had safe abortions and 28 (62.2%) had an unsafe abortion.Therefore, the practice of unsafe abortion in the current study is 28/206 (13.6%).The highly useful method for unsafe was medication 89 (43.2%) and home remedies 84 (40.8%).Three-quarters did not practice unsafe abortion in the past two years 177 (85.9%) and those who practiced it were 29 (14.1%).Reasons for unsafe abortion are highly due to the advice from friends and parents 129 (62.6%), rape 97 (47.1%) pregnancies rejected by their partners 89 (43.1%), and those in need to continue with studies 88 (42.7%).From those who practiced unsafe abortion, the high frequency of having an unsafe abortion was once in a life lifetime 26 (12.6%).Refer to Table 2.A total score was computed whereas a score above 4 was regarded as good practice toward unsafe abortion and a score below 4 was regarded as poor practice toward unsafe abortion.By summing up the scores of practices, it was found that the majority of participants had practiced unsafe abortion 147 (71.4%) and 59 (28.6%) had practiced safe abortion.

Knowledge of unsafe abortion among participants
Most of the participants had heard of safe 150 (72.8%), and unsafe abortion 150 (72.8%).Participants agreed that unsafe abortion is a problem in our current society 116 (56.3%).Regarding the source of information on unsafe abortion, most of the participants got information through friends 132 (64.1%), social media 116 (56.3%), health professionals 112 (54.4%), and relatives/family 114 (55.3%).It was found that 158 (76.6%) of participants had never experienced unwanted pregnancy in their lives and further responded that in case they become pregnant unwillingly they would continue and give birth 176

The association knowledge level and practice of unsafe abortion
Among postnatal women who had adequate knowledge, 94 (72.9%) had unsafe abortion.While, those who had inadequate knowledge, 53 (68.8%) had unsafe abortions.Generally, the result has shown that there was no significant association between knowledge level and practice of unsafe abortion (χ 2 = 0.385; P = 0.535).

Attitudes toward unsafe abortion among participants
Participants strongly agreed on the following items; unsafe abortion is an unhealthy procedure for a woman 101 (49%), and unsafe abortion can cause complications to a woman 91 (44.2%).The remaining three items indicate that the majority of participants agreed; that unsafe abortion can cause death to a woman 66 (32%), unsafe abortion can lead to sepsis to a woman 68 (33%), and unsafe abortion is a bad thing 75 (36.4%).Refer to Table 5.A total score was computed whereas a score above 10 was a positive attitude toward unsafe abortion while a score below 10 was regarded as a negative attitude toward unsafe abortion.The majority of participants, 130 (63.1%) had positive attitudes toward unsafe

The association attitude and practice of unsafe abortion
The results show that the attitude towards unsafe abortion was not significantly associated with the practice χ 2 = 1.066 ; P = 0.302.

Subjective norms towards unsafe abortion
Participants answered correctly on the following items: in your family and society is unsafe abortion viewed as a sin 123 (59.7%),When someone is known she had an abortion she is viewed as a killer 129 (62.6%), and when someone is known she had an abortion she is viewed as bad person 120 (58.3%).However, in two items, participants were incorrectly; when someone is known she had an abortion should be stigmatized 118 (57.3%) and do social norms a source of information about unsafe abortion 113 (54.9%).Refer to Table 7.A total score was computed whereas a score above 2 was regarded as good subjective norms toward unsafe abortion while a score below 2 was regarded as poor subjective norms.Through summation of scores, 113 (54.9%) had good subjective norms compared to 93 (45%) who had poor subjective norms.Refer to Fig. 3.

The association of subjective norms and the practice of unsafe abortion
Among participants with good subjective norms, 65 (69.9%) had unsafe abortion.Those who had poor subjective norms, 82 (72.6%) had unsafe abortions.Generally, the results indicate that subjective norms towards unsafe abortion were not significantly associated with the practice (χ 2 = 0.178 ; P = 0.673).

Perceived behavioral control
Participants agreed on items stating that I cannot practice unsafe abortion because it is illegal in Tanzania 70 (34%) while many participants 58 (28.2%) disagreed on the item "I'm confident that I cannot practice unsafe abortion".Participants were neutral 54 (26.2%) and others agreed 54 (26.2%) when responding to the item "Regardless I get unwanted pregnancy I cannot practice unsafe   abortion".Refer to Table 9.A total score was computed whereas a score above 6 was regarded as positive perceived behavior control toward unsafe abortion and a score below 6 was regarded as negative perceived behavior control toward unsafe abortion.After the summation of all scores, most of the participants 111 (53.9%) had positive perceived behavioral control while 95 (46.1%) had negative perceived behavioral control.

The association of perceived behavioral control and practice of unsafe abortion
The results show that perceived behavioral control towards unsafe abortion was not significantly associated with the practice χ 2 = 0.466; P = 0.495).

The association of sociodemographic characteristics and practice of abortion
The result indicates that only two sociodemographic characteristics (Marital status and educational level) were significantly associated with the safe abortion.In such a way respondents who were divorced had high good practice 16 (57.1%)compared to other participants who belonged to other marital statuses (χ 2 = 13.515;P = 0.004) and participants who had secondary educational level had safe abortion 73 (79.3%) than other participants with different educational levels (χ 2 = 10.146;P = 0.017).Other sociodemographic characteristics were significantly association with safe abortion; age (χ 2 = 8.751P = 0.119), place of residence (χ 2 = 0.029; P = 0.866), and occupation (χ 2 = 4.663; P = 0.198).Refer to Table 11.

Interaction of knowledge, attitude, and perceived behavioral control
Knowledge level of unsafe abortion was not significantly associated with attitude (χ2 = 3.657; P = 0.056).Refer to Supplementary Table (1) In the same way, there was no significant association between knowledge level and perceived behavioral control (χ2 = 3.537; P = 0.06).Refer to Supplementary Table (2) Meanwhile, there was a significant association between attitude and perceived behavioral control in such a way that 61 (55%) participants who had a positive attitude towards unsafe abortion had positive perceived behavioral control (χ2 = 3.657; P = 0.056).Refer to supplement Table 3.

Knowledge of unsafe abortion among the participants
The current study found that most of the participants 129 (62.6%) had adequate knowledge of unsafe abortion.This indicates the effectiveness of the ongoing national approaches for promoting public understanding of unsafe abortion.Further, it is because the majority were aged 30-34 and probably had been exposed to previous health education from antenatal care visits.The adequate knowledge of unsafe abortion (91%) is also reported in the previous study [30].

Attitude on unsafe abortion among participants
In this study, two-thirds of the participants 130 (63.1%) had positive attitudes.health beliefs (perceived susceptibility, perceived severity, perceived benefits) might be a reason for a positive attitude.For instance, when a woman feels susceptible to sepsis after an unsafe abortion might help her to avoid practicing unsafe abortion.Same  Fig. 2 Attitude towards unsafe abortion among study participants way, the woman will avoid unsafe abortion if believes that she may end up with complications and unfertile.The previous study supports that health beliefs influence attitude [31].Moreover, another study pinpoints that health beliefs and attitudes have a significant association with lifestyle change and risks in primary care [32].

Subjective norms on unsafe abortion among participants
From the results of our study, most participants had good subjective norms toward unsafe abortion 113 (54.9%) because the participants encounter pressure from a society that considers unsafe abortion sin, the doer of unsafe abortion as a killer, and anyone involved in unsafe abortion needs to be stigmatized.The finding is aligned with previous findings by Connell [33].

Perceived behavior control among participants
From the results of our study, the majority 111 (53.9%) had positive perceived behavior control, which has been influenced by the positive attitude participants possessed.
It is surprising to find out that the current study finding is contrary to the theory of planned behaviour since the theory is silent on whether attitude influences perceived behavioral control.

The practice of unsafe abortion among participants
Most participants 147 (71.4%) were found to practice unsafe abortion due to marital status and educational level.Since most of the participants in this study were single women and abortion in Tanzania is illegal, if a single woman becomes pregnant while still having a plan to develop a career, unsafe abortion is the alternative.Meanwhile, if a single woman becomes pregnant but the pregnancy is rejected by a partner, this triggers the woman to perform an unsafe abortion.Regarding education level, participants with lower educational levels or those who have never attended the school practice  Fig. 3 Subjective norms towards unsafe abortion among study participants unsafe abortion.The school is the place with health clubs, teaching, and guiding students in all health-related matters with unsafe abortion inclusively.The more a woman advances her education level, the more she is exposed to teachings on unsafe abortion.Out of health clubs, the curriculum may contain courses with the content of unsafe abortion.The finding is consistent with the previous study reported that unsafe abortion is higher among less educated women [34].

Study limitation
The study was carried out in a single region of Tanzania and at a single healthcare facility which may limit the generalization of the current findings.It seems that most of the important factors for unsafe abortion were left out as none of the independent variables in the current study were significantly associated with the practice of unsafe abortion.

Conclusion
Most of the respondents had high knowledge, attitudes, subjective norms, and perceived behavior control on unsafe abortion.This is an indicator that the implemented initiatives are effective.Maintaining the ongoing effort and improving strategies are promising to mitigate the burden of unsafe abortion.Future research needs to find out hidden factors associated with attitude and how health beliefs might influence someone's attitude towards unsafe abortion.Even though the theory does not support the attitude to influence perceived behavioral control, extensive studies need to be conducted to confirm the association between these two variables.

Fig. 1
Fig. 1 Knowledge level of unsafe abortion among study participants

Table 1
Sociodemographic characteristics of respondents (85.4%).Most women knew the gestation age when abortion is unsafe to be 3 months of pregnancy 126 (61.2%).On the signs and symptoms of unsafe abortion, participants agree that the following are signs and symptoms of unsafe abortion; abdominal pain 148 (71.8%), vaginal infection 152 (73.8%), vaginal bleeding 165 (80.1%),

Table 2
Practices of abortion among study participants

Table 3
Knowledge of unsafe abortion among participants abortion compared to those with negative attitudes 76 (36.9%).Refer to Fig.2.

Table 4
Association between sociodemographic characteristics and knowledge level among respondents

Table 5
Attitude on unsafe abortion among study participants

Table 6
The association between sociodemographics and attitude of participants

Table 7
Subjective norms on unsafe abortion among study participants

Table 8
The association between sociodemographics and subjective norms of participants

Table 9
Perceived behavior control on unsafe abortion among study participants

Table 10
The association between sociodemographic and perceived behavioral control of participants

Table 11
The association of sociodemographic characteristics and practice of unsafe abortion