This study was conducted from June to August 2024 in public hospitals in East Gojjam. The zone is found in the Amhara Region of Ethiopia, bordered to the south by the Oromia Region, to the west by West Gojjam, to the north by Debub Gondar, and to the east by Debub Wollo. The zone had an estimated population of 2,719,118 in 2020 (24). Based on the 2019 East Gojjam Zone Administration Office report, the zone had an estimated 91,634 women of reproductive age. This zone has 10 public hospitals, including one general hospital, eight primary hospitals, and one comprehensive specialized hospital, as well as 104 health centers and 406 health posts (25).

An institution-based cross-sectional study design was utilized.

All pregnant women who had antenatal care follow-ups and were available at the time of data collection were included in this study, while pregnant women with pre-existing type-1 or type-2 diabetes and those who were seriously sick or unable to communicate during data collection were excluded.

The assumption of the single population proportion formula was utilized to compute the sample size.n=(Zα/2)2p(1−p)d2where n = the desired sample size, Zα/2 = the critical value corresponding to a 95% confidence interval (Zα/2 = 1.96), d = the margin of error (5%), and p = the estimated population proportion, which was 48% from the study conducted in Oromia, Ethiopia (26).

N=(1.96)2×0.48×(1−0.0.48)(0.05)2=383.5∼384.

After considering a 10% non-response rate, the final sample size (Nf) = ni × (1/1-non-response rate); thus, Nf = 384 × (1/1–0.1) = 427.

A sample of five public hospitals was selected using a simple random sampling technique from the 10 public hospitals located in the East Gojjam Zone. The antenatal care registration book of each hospital was used to proportionally allocate the calculated sample size and determine the sampling fraction (k) (calculated using the population size divided by the sample size). The first mother was chosen using a simple random sampling technique among mothers who had an antenatal care follow-up on the day of data collection. Then, a systematic random sampling technique was used to select participants until the required sample size was achieved.

The data were collected using a face-to-face interviewer-administered questionnaire adapted from previously published articles (26, 27). The tool included the patient's background information and obstetric and medical history, as well as 15 questions assessing awareness of GDM, including its risk factors, diagnosis, treatment, and complications. The questions on risk factors assessed the patient’s awareness of the risk of GDM in patients with obesity prior to pregnancy, excessive weight gain during the present pregnancy, history of diabetes mellitus during previous pregnancies, and family history of diabetes.

Awareness of the course of GDM and its consequences to the unborn baby and mother was assessed by questions on whether GDM usually disappeared after delivery and whether women with GDM and children born to these women were at an increased risk for future T2DM and obesity.

To assess the patient’s awareness of the screening and diagnosis of GDM, questions on the type of test used and the timing of the test during pregnancy were asked. The options for the type of test used were urine test, blood test, blood test after a glucose load, and I do n't know. For the timing of the test, the options given were 12–16 weeks (3–4 months), 24–28 weeks (6–7 months), during delivery, and I do n't know. The answer 24–28 weeks (6–7 months) was considered the correct answer.

The patient’s awareness of the treatment for GDM was assessed using a question with the following options: diet and exercise, oral antidiabetic drugs, insulin injections, and I do n't know. Diet and exercise, insulin injections, and oral antidiabetic drugs were considered the correct response. Each correct response was given a score of 1 and each woman was scored out of a total of 15.

The tool was prepared in English and then translated into Amharic (the local language). It was then translated back into English to check its consistency. The face and content validity of the Amharic version questionnaires were assessed by four experts to control the threat to the validity of the data from the proposed instrument. In addition, the tool was pretested with 22 pregnant women who met the study criteria and was revised by experienced academic researchers. A well-structured tool, consisting of a chart review and interviewer-administered questions, was utilized to collect the data. The tool included sociodemographic information, obstetric and medical-related data, lifestyle-related data, and information on awareness of GDM. Five BSc midwives and five MSc professionals were recruited as data collectors and supervisors, respectively.

A 2-day training session was provided to both data collectors and supervisors by the principal investigator about the objective of the study, the data collection tool, procedure, and how to fill out the questionnaire. The tool was pretested at Fenote Selam public hospital on 5% of the sample size to ensure the consistency and completeness of the questionnaire. Data collectors were supervised throughout the course of the data collection period. Then, the overall process was coordinated and controlled by the principal investigator. The principal investigator, supervisors, and data collectors had a discussion meeting after data collection to ensure completeness. Codes were given to the questionnaires during data collection. Furthermore, the collected data were entered into EpiData version 4.2 to minimize data entry errors and kept in the form of a file in a secure place. The results of the study were used only for the study's purpose.

The collected data were entered into EpiData version 4.2. It was then exported to SPSS version 25 for analysis. Descriptive statistics, such as frequency and summary statistics, were used to describe the characteristics of the study participants. A binary logistic regression model was used to determine the factors associated with the outcome variable. The model's fitness was assessed using the Hosmer–Lemeshow goodness-of-fit test for the adjusted model [chi-square (χ²) = 8.93, p = 0.35], which indicated a good fit. Multicollinearity among the independent variables was evaluated using variance inflation factors (VIFs), and no significant multicollinearity was detected. In the bivariable logistic regression, all explanatory variables with a p-value of 0.25 or less were considered for the multivariable logistic regression analysis. An adjusted odds ratio (AOR) with its corresponding 95% confidence intervals was used to indicate the association between the dependent and independent variables, and a p-value less than 0.05 indicated statistical significance.

Ethical clearance was obtained from the Institutional Ethical Review Board of the College of Medicine and Health Sciences, Debre Markos University (Approval No.: CMHS/R/C/Ser/D/315/01/16). Responsible officials and managers at each hospital were informed and permission was obtained. The participants (legal guardian/next of kin) provided written informed consent to participate in this study, and were informed that they had the right to withdraw from the study at any time.

The awareness score was determined based on the participants’ correct answers, with 1 point given for each correct answer and 0 for incorrect responses. It was measured using 15 awareness questions and categorized as “good awareness of GDM” (>9 out of 15), “fair awareness of GDM” (6–9 out of 15), and “poor awareness of GDM” (0–5 out of 15). Finally, pregnant women with good or fair knowledge were considered to be aware of GDM, and those with poor knowledge were considered to be unaware (21, 23).

In total, 423 (99.1%) respondents participated in this study. The mean maternal age of the participants was 26.64 (SD 4.96) years, of which 63.6% were ≥25 years. Almost all of the respondents (421, 99.5%) were married. Approximately 30% of the study participants had attended college or higher education, while 56 (13.2%) had not received any formal education. Over half (51.8%) of the study participants were housewives (Table 1).

More than three-fifths (63.1%) of the women were multigravida; 393 (92.9%) of the women who participated had experienced a planned and wanted pregnancy. Approximately 12% (52, 12.3%) of the participants had a history of a bad obstetric complication, with abortions accounting for 71.2% of these cases, followed by early neonatal death and stillbirth in 17.3% and 11.5%, respectively. Approximately half (214, 50.5%) of the study participants were in their second trimester of pregnancy (Table 2).

Of the study participants, 8 (1.9%) had chronic medical disease, among which HIV accounted for 50% of these cases, followed by hepatitis and psychosis in the remaining 50%. All HIV-positive participants received antiretroviral therapy. Approximately 10% of the participants had a family history of chronic medical disease, with diabetes mellitus and hypertension being the most commonly noted conditions (Table 3).

Participants were grouped into three categories based on their scores, with “good awareness of GDM” defined as a score of >9 out of 15, “fair awareness” as a score of 6–9 out of 15, and “poor awareness” as a score of 0–5 out of 15. Ultimately, women with good or fair awareness were considered to be aware of GDM, while those with poor knowledge were deemed to be unaware. Accordingly, the research revealed that only 27.0% (95% CI: 0.23–0.31) of the pregnant women were aware of GDM (Figure 1).

Various sources of information about GDM were identified, with friends being the most common at 53.2%, followed by family members at 17.7% and healthcare professionals at 15.4% (Figure 2).

In the bivariable logistic regression model, we considered various factors for further analysis, including maternal age, residence, religion, occupation, income, parity, receipt of tetanus toxoid injection and iron supplementation, pregnancy status, family history of medical disease, obstetrical history, past medical conditions, and partner involvement (all with p < 0.25).

In the final multivariable logistic regression model, several factors emerged as significantly associated with awareness of GDM. Women who were Muslim were 2.9 times more likely to have good awareness compared to Orthodox women (CI 1.34, 6.24; p = 0.007). Partner involvement also played a role; women who did not have an involved partner were less likely to have good awareness compared to those with involved partners (AOR: 0.58; CI 0.35, 0.95; p = 0.03). A family history of chronic medical disease (diabetes, hypertension, or both) increased the likelihood of good GDM awareness by 5.2 times (CI 2.40, 11.25; p < 0.001). Additionally, women with mistimed but wanted pregnancies were 3.4 times more likely to have good GDM awareness compared to those with planned and wanted pregnancies (CI 1.40, 8.10; p = 0.007) (Table 4).