Magnitude and determinants of anemia in the third trimester among pregnant women in Dodoma: An analytical cross-sectional study

Abstract

Despite widespread efforts, anemia during pregnancy remains a significant indirect cause of maternal mortality in Tanzania. During the third trimester, anemia is more critical as it can increase the risk of complications for the mother and the fetus. This study aimed to assess the prevalence of anemia in the third trimester, its determinants, facility preparedness, and preventive services provided. A facility-based analytical cross-sectional study was conducted in the Dodoma region. Through a systematic sampling, 396 pregnant women in their third trimester were studied. Data were collected using an interviewer-administered questionnaire, documentary review, hemoglobin measurement, and observation. Descriptive and inferential statistics using a binary logistic regression model were used to determine the predictors of anemia. The odds ratio and confidence interval were reported, and the significance level was set at P-value < .05. The prevalence of anemia among pregnant women in the third trimester was 33.3%. All facilities provided all the recommended anemia preventive services but were not well prepared for the provision of those services. The items that fell short in most of the facilities were FEFO and Mebendazole. There is a significant association between anemia and being in the age group of 15 to 24 years [adjusted odds ratio (AOR): 3.693, P = .025). The protective factors are being married (AOR: 0.408, P = .037), having college education level (AOR: 0.063, P = .00), more than 8 antenatal attendances visits (AOR: 0.311, P = .022), pregnancy interval of more than 2 years (AOR: 0.172, P < .001), having adequate knowledge (AOR: 0.392, P = .018) and positive attitude (AOR: 0.204, P = .015). The prevalence of anemia among pregnant women in the third trimester in Dodoma is alarming. All preventive services are provided, but not all facilities are prepared for the provision of anemia those services. Anaemia risk is increasing with the decrease of age. Protective factors vary from social, obstetric, knowledge, and attitude factors.

1. Introduction

Anemia in pregnancy refers to a low hemoglobin level (Hb) to <11 g/dL in the first and third trimesters, and <10.5 g/dL in the second trimester.^[1] Anemia in pregnancy affects 56 million pregnant women globally,^[2] with a prevalence of 57.1% as compared to nonpregnant women who have a prevalence of 40.0%.^[3] Africa has an anemia prevalence of 41.7%, with a bigger disparity among nations ranging from 23.36% in Rwanda to 57.10% in Tanzania.^[4] To facilitate early detection, management, and early referral, anemia in pregnancy has been classified as severe for the hemoglobin level of > 7.0 g/dL, moderate for the hemoglobin level of 7.0 to 9.9 g/dL, and mild anemia for the Hb level of 10.0 to 10.9 g/dL,^[1] and several approaches to solving maternal anemia such as early detection (hemoglobin testing) and clinical management prevention strategies have been identified.^[5]

There are several causes of anemia including blood cell loss, chronic or acute hemolytic anemia, hemoglobinopathies like sickle cell disease which reduces red blood cell production, parasitic infections, and nutritional deficiency.^[6,7] But the higher risk to pregnant women is due to the increased demand of the growing fetus, and the increased blood volume of up to 30% to 50% of the mother which occurs mostly during the third trimester.^[8] However, despite the haemodilution, iron deficiency has been the most common cause of anemia during pregnancy.^[6] Anemia in pregnancy especially in the third trimester has been linked to poor maternal outcomes like postpartum hemorrhage, blood transfusion, and preeclampsia, and poor neonatal birth outcomes which include low birth weight, preterm birth, and small for gestation age,^[9] and prolonged hospitalization.^[10]

Various factors were attributed to anemia in pregnancy which include being in the third trimester,^[11] rural residence,^[12] being illiterate,^[13] low pregnancy interval,^[14] multiparity,^[15] poor dietary diversity^[16] and nonadherence to anemia preventive services.^[17] However, it has been evidenced that anemia in pregnancy can be prevented during antenatal care (ANC). Studies have shown a significant reduction in the prevalence of anemia among pregnant women as the prevalence of 32.2% among 1205 pregnant at first booking, leading to its reduction at term as 736 of the 1205 anemia was corrected making a 69.9% prevention at term.^[18] The recommended services for the prevention of anemia at ANC in Tanzania include the provision of insecticide treated nets (ITN), intermittent preventive treatment for malaria (IPT), ferrous sulfate-C-folic acid (FEFO), and mebendazole.^[5]

Healthcare facility preparedness is critical for effective anemia-prevention services provision. Studies showed that poor facility preparedness, such as inadequate stock for essential anemia prevention, lack of guidelines, and trained staff can lead to ineffective services.^[19] Moreover, Rakanita et al, address infections, micronutrient deficiencies, and adherence issues can all be alleviated through effective anemia-prevention programs in a well-prepared healthcare facility.^[20]

Despite having adequate antenatal visits in Tanzania, such that 65% of women reported attending 4 times or more,^[3] pregnant women reported having lower hemoglobin levels, being more common in the third trimester by 48.8%,^[21] which jeopardizes women’s health and lead to significant poor neonatal and maternal outcomes. This has raised a concern about the types of services related to anemia-prevention women received during their ANC visits. In addition, it is not unimportant to identify whether the healthcare facilities in the ANC department are prepared enough to provide the services as required, because healthcare professionals have been reported elsewhere for failing to offer ANC services per the guidelines due to either a lack of continuous training or a shortage of commodities.^[22]

Studies have been undertaken in Dodoma-Tanzania to address the issue of anemia during pregnancy^[23] as well as the prevalence of postpartum anemia and its associated factors,^[24] still, there is limited information on studies focused on women in their third trimester, which is the most critical period for pregnant women and their new-borns. Existing researches address the subject in a fragmented manner, yielding an incomplete view of the conditions. Therefore, this study aimed to determine the prevalence and determinants of anemia in the third trimester, facility preparedness, and anemia preventive services received by pregnant women during ANC visits.

2. Materials and methods

2.1. Study area and setting

This study was conducted in urban and rural areas of the Dodoma region. Dodoma region is among 31 regions of the Tanzania mainland composed of 7 districts, with a population of 3,085,625, of which 1,572,865 are females.^[3] The main economic activities are agriculture and livestock keeping, with grapes being the major cash crop cultivated. Though the prevalence of anemia among pregnant women in Tanzania is known to be higher, (57%), the Dodoma region has a scarcity of recent information regarding anemia among pregnant women. However, a previous study reported a prevalence of 63.8% (Tatala et al, 2002). The fertility rate is 5.21% and the growth rate is 6.4%. The study settings were at the primary healthcare facility level (dispensaries, health centers) and secondary level (district hospitals). There are 477 healthcare facilities in the Dodoma region managed by various authorities including the Government, Faith-based organizations, and private ownership.^[25]

2.2. Study design and study population

The study adopted an analytical cross-sectional study design involving all pregnant women in the third trimester attending ANC visits in the selected healthcare facilities in the Dodoma region. Those who voluntarily consented to participate in the study were included, while women who reported having known chronic medical conditions like sickle cell, kidney diseases, cancer, ulcerative colitis, bleeding disorder, bleeding disorder, and infections such as malaria and hookworm were excluded because these infections can be a direct cause of anemia as a result of nutrient malabsorption and red blood cells destruction respectively, thus accurate prevalence without the confounding effect of such infections can be obtained.

2.3. Sample size estimation

The sample size was calculated using the Kish Leslie formula (Kish, 1998) $\mathrm{n}=\frac{Z^{2}P(1-P)}{d^2}$ where n = sample size, Z = standard normal deviation that corresponds to a 5% level of statistical significance, equivalent to 1.96, d = marginal error of the study, which is 0.05, P = 63.8% which is the prevalence of anemia in pregnancy in Dodoma.^[26] Therefore, the calculated sample size is 359. After adding 9% of the nonrespondent rate, the final sample size obtained is 395 participants.

2.4. Sampling technique

A multistage sampling method was used to select districts, healthcare facilities, and study participants. Among 7 districts in the Dodoma region, a simple random sampling was used to select 2 districts. The selected districts were Dodoma City and Chamwino District Council. Then, healthcare facilities were selected from each district, where in the 2 selected districts, there is a total of 4 primary healthcare facilities and 2 secondary healthcare facilities. Based on the total sample size required for the study, a simple random sampling was used to select 2 primary and 1 secondary healthcare facility from each district making a total of 6 facilities. The secondary healthcare facilities selected are St Gema and Mvumi Hospital, while the selected primary healthcare facilities are Makole and Chamwino Health Centers, and Kikuyu and Manchali Dispensary. Thereafter, a proportionate sampling was used to select the number of participants in each facility using a formula n_i = (N_i/N_t)n,^[27] where n_i = proportional sample for each health facility, N_i = number of attendances in each health facility, N_t = total attendance in all health facilities, and n = calculated sample size. Within each selected facility, systematic random sampling was employed to select study participants using a formula of k = N/n,^[28] where n = total participants required per day and N = total estimated number of attendances per day. The first participant was randomly selected. The kth value differs for each facility, depending on the total number of attendances per facility.

2.5. Data collection methods and tools

The data were collected in antenatal clinics in each health facility from April 2024 to May 2024. Several data collection methods were used as follows:

2.5.1. An interviewer-administered questionnaire

Each participant was placed in a private room where she was interviewed by the principal investigator or research assistants who are nurses by profession. These research assistants received a few hours of orientation about data collection tools and procedures before the commencement of data collection. Study participants were questioned about their background characteristics, adherence to services received (e.g., FEFO)^[29] which is composed of 4 items asking whether a woman received the adequate number of tablets from the previous visit, the number of tablets she took in the previous month and the number of tablets she took in the previous week as directed. The response was confirmed by comparing the total tablets received previously versus the total tablets that have been consumed.

This method was also used to assess their dietary diversification^[30]; and their knowledge and attitude regarding anemia.^[31] All collected information was recorded in a paper-based questionnaire. The whole procedure took about 15 minutes to complete.

2.5.2. Documentary review

The antenatal card number 4 was used to collect data on women’s obstetric characteristics, including parity, gestation age at first visit, total number of prenatal visits, type of ANC services received related to anemia prevention, and HIV status. The checklist for documentary review was adopted from the Antenatal Guideline of 2018 which speculates the types of anemia preventive services to be given including mebendazole, ITN, FEFO, and IPT_p.^[32]

2.5.3. Observation

This technique was employed to gather data regarding facility preparedness for anemia prevention, whereby the investigator physically assessed the availability of all commodities and pharmaceuticals for anemia prevention. The observation checklist was adopted from Service Availability and Readiness Assessment (SARA).^[33] The tool has 15 items in a yes/no format, including items on the availability of anemia preventive services (4 items), availability of commodities/pharmaceutical for anemia prevention (4 items), e.g., ITN, FEFO, IPTp, and Mebendazole, commodity adequacy (4 items), and availability of training to staff and guidelines for anemia prevention and management (3 items).

2.5.4. Measurements

The hemoglobin level was measured using a finger-prick blood sample (capillary blood sample) by the lab technician to estimate the prevalence of anemia. The following items were assembled: prickers, a gallipot with dry swabs, an antiseptic solution, and a hemoque analyzer. The Haemoque machine used was hemochromax (model hemocromax Plus, from Korea). A participant was positioned in a comfortable position before having her ring fingers cleansed, dried, and poked. A microcuvette was attached to the oozing blood and inserted into the hemoque analyzer to obtain the results. The results were promptly recorded. All used supplies, including sharps and swabs, were placed in a safety box and discarded to an incinerator thereafter.

2.6. Measurement of variables

2.6.1. Dependent variable

Anaemia was measured by 1 item in a binary scale on women’s hemoglobin level, that is, those with a hemoglobin level of ≤ 10.9 g/dL were regarded as anemic and coded “0,” and those with a hemoglobin level of 11 g/dL and above were regarded as not anemic and coded “1.” The anemic status was further categorized as severe for the hemoglobin level of > 7.0 g/dL, moderate for the hemoglobin level of 7.0 to 9.9 g/dL, and mild anemia for the Hb level of 10.0 to 10.9 g/dL.^[1]

2.6.2. Independent variable

2.6.2.1. Social demographic characteristics

Marital status, occupation, place of residence, religion, educational level, age in years, and per capita income of 1 US dollar per day.

2.6.2.2. Obstetric characteristics

Parity, pregnancy interval, gestation age at first visit, and total number of ANC attendance.

2.6.2.3. Healthcare facility characteristics

This includes the level of healthcare facility (HCF), which could either be of a lower level (e.g., dispensary) or a higher level/type (e.g., district hospital), ownership (privately owned or government owned), location (urban or rural).

2.6.2.4. Anaemia prevention services received

Services received were measured by 4 items in a binary scale (yes/no) on whether a woman has received all the required services necessary for anemia prevention in each ANC visit. These services were FEFO, IPT, mebendazole, and ITN. One point was given for every service received in each visit, and 0 points for service not received. The total score was calculated for all visits. As per the service provision plan set by Reproductive, Maternal New-born Child and Adolescent Health (RMNCAH), if 85% of women received IPT, FEFO, and mebendazole, and if 80% received ITN, then it is termed as there are adequate services, if it is less than that, there is inadequate.^[34]

2.6.2.5. Health facility preparedness

The preparedness of healthcare facilities was evaluated using 15 indicators in a binary scale on whether the healthcare facility has anemia preventive services, commodity, e.g., pharmaceuticals, commodity adequacy which was measured by comparing the number of participants attended per month versus the stock received per month, and training and guideline. One point was given for the available items and 0 for the unavailable. The total score was 15, and a facility has to have all items to be considered as prepared.

2.7. Analysis

Data were entered into SPSS software version 25 and checked for completeness before the analysis. Descriptive analysis was used to summarize the background characteristics of women, anemia-prevention services received, and facility preparedness, and the results were presented in frequency and percentages. A chi-square test and a binary logistic regression model were used to determine the predictors of anemia. crude odds ratio, adjusted odds ratio (AOR), and a 95% confidence interval were reported and a P-value of < .05 was considered statistically significant.

2.8. Ethical consideration

Before conducting the study, ethical clearance was sought from the university of Dodoma research committee board. A permission for research conduct was obtained from the president’s office, regional administration, and local government. Participants rights were taken into consideration as a brief explanation of the research study was given and written informed consent was obtained. As for participants who were below 18 years of age, the consent was obtained from the parent or a husband who escorted the woman. The option of agreeing or disagreeing on the study or quitting at any time of the study without threats or cohesion was explained prior to the study. Privacy and confidentiality of participant information were ensured, the questionnaires were coded instead of using patients’ names, and the interviews were conducted in a private area. And those who were found to be anemic were connected to the ANC unit in-charge for further management.

3. Results

3.1. Social demographic and obstetric characteristics of the respondents

The response rate in this study is 94%. The age of participants ranged from 17 to 44 years, with a mean age of 27 years (SD 6.16). The majority of participants were from urban areas, 329 (83.608%), while 280 (70.71%) were married, and 138 (34.85%) had a primary education level. Moreover, 204 (51.1%) had 2 to 3 pregnancies, and 115 (29.04%) delivered at least once, whereby the pregnancy interval for most of the participants 167 (60.29%) was more than 2 years (Table 1).

Table 1.

Social demographic and obstetric characteristics of study participants (N = 396).

Variable	Frequency (n)	Percent (%)
Social demographic
Age group
15 to 24	144	36.36
25 to 29	103	26.01
30 to 34	79	19.95
35+	70	17.68
Marital status
Married	280	70.71
Not married	116	29.29
Education level
Never attended school	39	9.85
Primary education	138	34.85
Secondary education	131	33.08
Collage/university	88	22.22
Occupation
Employed	60	15.15
Self-employed	151	38.13
Peasant/farmer	72	18.18
Housewife	113	28.54
Income
Below 2500	82	30.04
Above 2500	191	69.96
Place of residence
Urban	329	83.08
Rural	67	16.92
Religion
Muslim	92	23.23
Christian	301	76.01
Pagan	3	0.76
Obstetric characteristics
Gravidity
Prime	119	30.05
2 to 3	204	51.52
4+	73	18.43
Parity
Nullipara	125	31.57
Primipara	115	29.04
Multipara	156	39.39
Pregnancy interval
Below 2 yr	110	39.71
Above 2 yr	167	60.29
1st ANC gestation
Below 12 weeks	61	15.4
Above 12 weeks	335	84.6
Visits
1 to 3	151	38.13
4 to 6	142	35.86
7+	103	26.01
HCF characteristics
Facility type
Hospital	62	15.66
Health center	269	67.93
Dispensary	65	16.41
Facility location
Urban	330	83.33
Rural	66	16.67
Facility ownership
Government/public	334	84.34
Private	62	15.66

3.2. Prevalence and severity of anemia in the third trimester

The Hb level readings showed that the maximum level was 19.6 g/dL and the minimum level was 7.7 g/dL, with a mean Hb level of 11.6 g/dL and a standard deviation of 1.519 g/dL. The results revealed that 132 (33.3%) pregnant women had an Hb level of ≤ 10.9 g/dL. thus, were termed anemic. Based on World Health Organization categorization, this study indicates that, out of 132 pregnant women who were anemic, 123 (93.2%) had mild anemia, i.e., a hemoglobin level of 9 to 10.9 g/dL, 9 (6.8%) had moderate anemia, i.e., a hemoglobin level of 7.0 to 8.9 g/dL while none had severe anemia.

3.3. Types and adequacy of preventive services received

Results of this study showed that the majority of women received ITN 358 (90.4%), while 327 (82.58%) received FEFO (Fig. 1). According to one of the 3 plans set by RAMNCA, based on the target population in a month, the coverage should be 80% for ITN, and 85% for IPT, FEFO, and Mebendazole. Therefore, the service provision for ITN and IPTp was adequate as the established target was met, while for mebendazole and FEFO, the provision was inadequate, as the target of 85% was not met (Fig. 2).

Figure 1.

Proportion of pregnant women who received Anaemia preventive services (ITN, IPT, Mebendazole, and FEFO). FEFO = ferrous sulfate-C-folic acid, IPT = intermittent preventive treatment for malaria, ITN = insecticide treated nets.

Figure 2.

Adequacy level of anaemia preventive services provided.

3.4. Healthcare facility preparedness in anemia prevention

Six healthcare facilities were assessed for the availability of services, training and guidelines, availability of commodities, and the adequacy of the commodities for 3 consecutive months. The study found that in all healthcare facilities, 6 (100%) provide anemia-prevention services, and half of them have IPT and FEEFO, 3 (50%). As for the adequacy of these commodities, results showed that all 6 facilities had adequate ITN (100%), and less than half had adequate FEFO and mebendazole 2 (33.3%) (Table 2). The general preparedness score showed that all healthcare facilities were not prepared for the provision of anemia preventive services.

Table 2.

Item analysis for anemia preventive services provided in healthcare facilities (N = 6).

Item	Yes, n (%)	No, n (%)
Service provision
Does the facility provide Intermittent preventive treatment in pregnancy?	6 (100.0)	–
Does the facility provide ITN?	6 (100.0)	–
Does the facility provide FEFO?	6 (100.0)	–
Does the facility provide Mebendazole/albendazole?	6 (100.0)	–
availability of the commodities
Does the facility have FEFO for the past consecutive 3 mo?	3 (50.0)	3 (50.0)
Does the facility have mebendazole for consecutive 3 mo?	4 (66.7)	2 (33.3)
Does the facility have IPT for the consecutive 3 3 mo?	3 (50.0)	3 (50.0)
Does the facility have 1TN for the past consecutive 3 mo?	6 (100.0)	–
Adequate commodities
Does the facility have an adequate number of FEFOs compared to the number of attendees per month?	2 (33.3)	4 (66.7)
Does the facility have an adequate number of mebendazole tabs compared to the number of attendees per month?	2 (33.3)	4 (66.7)
Does the facility have an adequate number of IPTs compared to the number of attendees per month?	5 (83.3)	1 (16.7)
Does the facility have adequate ITN compared to the number of attendees per month?	6 (100.0)	0 (0.0)
Staff training and guideline availability
Are there any National ANC guidelines?	4 (66.7)	2 (33.3)
Do the providers utilize the provided guideline?	2 (33.3)	4 (66.7)
Is there any training for staff in ANC units in the last 2 yrs?	5 (83.3)	1 (16.7)

ANC = antenatal care, FEFO = ferrous sulfate-C-folic acid, IPT = intermittent preventive treatment for malaria, ITN = insecticide treated nets.

3.5. Factors associated with anemia among pregnant women in the third trimester

A binary logistic regression model was used to assess factors associated with anemia among pregnant women in the third trimester. Initially, a chi-square test was performed, and all variables with a P-value of < .2 were entered into the regression model. Then all variables with a P-value of < .05 in a bivariable logistic regression model were fitted in the final multivariable model to establish the sole effect of each variable on anemia.

Results showed that there is a significant association between anemia and being in the age group of 15 to 24 years (AOR: 3.69, P-value: .025), 25 to 29 years (AOR: 3.65, P-value: 0.03), and 30 to 34 years (AOR: 3.3, P-value: 0.027) compared to being more than 35 years of age. Moreover, women who were attending ANC at the dispensary level were 4 times more likely to have anemia in their third trimester compared to those who were attending at the hospital level (AOR: 4.2, P-value: .04). Furthermore, the protective factors are being married (AOR: 0.40, P-value: .037), having a college education level (AOR: 0.06, P-value: .00), having ≥ 8 antenatal attendances visits (AOR: 0.31, P-value: .022), having pregnancy interval of more than 2 years (AOR: 0.1, P-value < .001), adherence to preventive services (AOR: 0.365, P-value: .016), having adequate knowledge (AOR: 0.39, P-value: .018) and having a positive attitude towards preventive services (AOR: 0.20, P-value: .015) compared to their counterparts (Table 3).

Table 3.

Bivariate and multivariable binary logistic regression analysis of the factors associated with anemia in the third trimester (N = 396).

Variable	Not anemic, n (%)	Anemic, n (%)	Bivariate analysis			Multivariate analysis
			COR	95% CI	P-value	AOR	95% CI	P-value
Age group
15 to 24	81 (56.25)	63 (43.75)	3.11	(1.58–6.09)	.001	3.69	(1.16–11.73)	.03
25 to 29	67 (65.05)	36 (34.95)	2.15	(1.05–4.38)	.04	3.65	(1.18–11.34)	.04
30 to 34	60 (75.95)	19 (24.05)	1.27	(0.58–2.77)	.55	3.32	(1.07–10.35)	.03
35+	56 (80.00)	14 (20.00)	Reference			Reference
Marital status
Married	209 (74.64)	71 (25.36)	0.31	(0.20–0.48)	<.001	0.41	(0.18–0.95)	.04
Not Married	55 (47.41)	61 (52.59)	Reference			Reference
Education level
Never attended school	10 (25.64)	29 (74.36)	Reference			Reference
Primary education	84 (60.87)	54 (39.13)	0.22	(0.10–0.49)	.00	0.21	(0.06–0.72)	.01
Secondary education	98 (74.81)	33 (25.19)	0.12	(0.05–0.26)	<.00	0.11	(0.03–0.40)	.00
Collage/university	72 (81.82)	16 (18.18)	0.08	(0.03–0.19)	<.00	0.06	(0.01–0.32)	.00
Occupation
Employed	43 (71.67)	17 (28.33)	0.75	(0.38–1.48)	.41	–	–	–
Self-employed	105 (69.54)	46 (30.46)	0.83	(0.49–1.40)	.49	–	–	–
Peasant/farmer	42 (58.33)	30 (41.67)	1.36	(0.74–2.49)	.33	–	–	–
Housewife	74 (65.49)	39 (34.51)	Reference
Income
Below 2500	51 (62.20)	31 (37.80)	Reference			–	–	–
Above 2500	132 (69.11)	59 (30.89)	0.735	(0.43–1.26)	.27	–	–	–
Place of residence
Urban	226 (68.69)	103 (31.31)	Reference			–	–	–
Rural	38 (56.72)	29 (43.28)	1.675	(0.98–2.86)	.06	–	–	–
Facility type attended
Hospital	50 (80.65)	12 (19.35)	Reference			Reference
Health center	184 (68.40)	85 (31.60)	1.93	(0.98–3.80)	.06	1.96	(0.59–6.49)	.27
Dispensary	30 (46.15)	35 (53.85)	4.86	(2.19–10.7)	<.00	4.29	(1.07–17.20)	.04
Gravidity
Prime	76 (63.87)	43 (36.13)	Reference			–	–	–
2 to 3	145 (71.08)	59 (28.92)	0.72	(0.45–1.17)	.18	–	–	–
4+	43 (58.90)	30 (41.10)	1.23	(0.68–2.24)	.49	–	–	–
Parity
Nullipara	81 (64.80)	44 (35.20)	Reference			–	–	–
Primipara	78 (67.83)	37 (32.17)	0.87	(0.51–1.49)	.62	–	–	–
Multipara	105 (67.31)	51 (32.69)	0.89	(0.5–1.47)	.66	–	–	–
Pregnancy interval
Below 2 years	57 (51.82)	53 (48.18)	Reference			Reference
Above 2 years	132 (79.04)	35 (20.96)	0.29	(0.17–0.48)	<.00	0.17	(0.08–0.38)	<.00
1st ANC gestation
Below 12 weeks	47 (77.05)	14 (22.95)	Reference			–	–	–
Above 12 weeks	217 (64.78)	118 (35.22)	1.83	(0.97–3.45)	.06	–	–	–
Total ANC visits
1 to 3	79 (52.32)	72 (47.68)	Reference			Reference
4 to 7	99 (69.72)	43 (30.28)	0.48	(0.30–0.77)	.00	0.38	(0.16–0.93)	.03
8+	86 (83.50)	17 (16.50)	0.22	(0.12–0.40)	<.00	0.31	(0.11–0.84)	.02
Received IPT
Yes	249 (70.14)	106 (29.86)	0.25	(0.12–0.48)	<.00	0.23	(0.07–0.84)	.03
No	15 (36.59)	26 (63.41)	Reference			Reference
Received ITN
Yes	242 (67.60)	116 (32.40)	0.66	(0.33–1.30)	.23	–	–	–
No	22 (57.89)	16 (42.11)	Reference			–	–	–
Received mebendazole
Yes	238 (72.12)	92 (27.88)	0.25	(0.15–0.44)	<.00	0.31	(0.10–0.987)	.05
No	26 (39.39)	40 (60.61)	Reference			Reference
Received FEFO
Yes	235 (71.87)	92 (28.13)	0.28	(0.17–0.49)	<.00	0.28	(0.08–0.86)	.15
No	29 (42.03)	40 (57.97)	Reference			Reference
HIV status
Positive	13 (52.0)	12 (48.0)	1.931	(0.85–4.35)	.113	2.451	(0.729–8.239)	.147
Negative	251 (67.7)	120 (32.3)	Reference			Reference
Adherence
Adhere	198 (73.6)	71 (26.4)	0.388	(0.25–0.60)	<.001	0.365	(0.161–0.826)	.016
Not adhere	66 (52.0)	61 (48.0)	Reference			Reference
Knowledge
Knowledgeable	149 (81.42)	34 (18.58)	0.27	(0.17–0.42)	<.00	0.39	(0.18–0.85)	.02
Not knowledgeable	115 (53.99)	98 (46.01)	Reference			Reference
Attitude
Favorable	234 (74.05)	82 (25.95)	0.19	(0.10–0.36)	<.00	0.20	(0.06–0.74)	.02
Neutral	12 (41.38)	17 (58.62)	0.77	(0.30–1.97)	.59	0.60	(0.10–3.61)	.58
Not favorable	18 (35.29)	33 (64.71)	Reference			Reference
Dietary status
Adequate diet	209 (71.58)	83 (28.42)	0.45	(0.28–0.71)	.00	0.41	(0.18–0.90)	.03
Inadequate diet	55 (52.88)	49 (47.12)	Reference			Reference

ANC = antenatal care, AOR = adjusted odds ratio, COR = crude odds ratio, FEFO = ferrous sulfate-C-folic acid, IPT = intermittent preventive treatment for malaria, ITN = insecticide treated nets.

4. Discussion

Anemia in pregnancy is still a public health concern because it is one of the indirect causes of maternal death and has been linked to poor outcomes for both, the mother and the newborn. The purpose of this study was to determine the prevalence of anemia in the third trimester, its determinants, facility preparedness, and anemia preventive services received by pregnant women.

The results show that 33.3% of women had anemia which is above the acceptable level of 20% to 30% in developing countries. This result suggests that many pregnant women who are in their third trimester in this setting are at higher risk of postpartum hemorrhage, preeclampsia, and eclampsia, poor neonatal birth outcomes including low birth weight, preterm birth, and small gestation age.^[35] This highlights the need for immediate interventions in this setting. These findings are in line with a study done in Katsina Metropolis which showed that anemia prevalence among pregnant women in the third trimester attending different hospitals was 34.38%^[36]

However, there are studies conducted in various areas that reported anemia in the third trimester at a higher percent compared to this study, which includes a study conducted in Egypt which reported a prevalence of 72.1%,^[37] Burkina Faso, 54%,^[38] and Unguja, Tanzania 82.9%.^[39] The difference could be explained by the differences in population characteristics and behavioral characteristics of participants. As for the study in Egypt, it was conducted in a hospital that serves people of low socioeconomic and low education levels, and that which was conducted in Unguja, most of the participants were reported to have poor nutritional practices. All these affect the utilization of ANC services and adherence to the management provided which increases the risk of anemia.

A notable association between anemia and women’s age was found. The study’s findings indicate that a significant number of younger pregnant women were more likely to be anemic than those who were older. This finding may have been influenced by the fact that younger women may make poor health-related decisions,^[40] poor eating behaviors,^[41] and low self-esteem which may influence the behaviors of seeking medical attention^[42] and perceiving a problem as less serious than it is, and also they are less autonomous.^[43] This suggests that being youthful may make a woman less cautious, therefore an eye must be on teen pregnant women as they are more prone to anemia and other pregnancy-related complications as compared to older women. This finding is similar to the study conducted in Enugu State, Nigeria, where a significantly low hemoglobin level (8.5 g/dL) was observed among teen pregnant as a result of poor eating behaviors and less micronutrient uptake.^[44] A link between a woman’s age and anemia was also observed in a study conducted in East Africa^[6] and Uganda.^[45] However, other studies reported that older women were more anemic than younger ones, and that has been associated with monotonous and insistent pregnancies that predispose them to anemia.^[46,47]

This study shows a significant association between the number of antenatal visits and anemia in the third trimester. As the likelihood of developing anemia decreased with increased antenatal attendance, in this study the subsequent number of women attending fewer visits had anemia as compared to their counterparts. This has been well established that regular ANC attendance exposes women to more services and interventions that lower the risk of anemia in pregnancy. Therefore, efforts should be taken to ensure pregnant women are completing the 12 ANC visits as required. This finding is similar to other studies conducted in Tanzania, which reported that the low coverage of ANC visits increases the possibility of not detecting and treating potential anemia during pregnancy.^[41] In Ethiopia, a lower prevalence of anemia was observed among women attending antenatal (17.8%), thus ANC influenced reducing anemia.^[48] Also in Ghana, antenatal visits had a positive impact on pregnancy, including anemia reduction.^[49]

Study results indicate a considerable association between marital status and anemia, marriage being a protective factor against anemia, as women who were married were less likely to have anemia as compared to those who were not married. This finding implies that married women may have a spouse’s support in economics which allows a woman to access medical care, diverse food, and support in preventive services uptake and adherence, all of these decrease the risk of anemia. The finding is comparable to studies conducted in other African countries, whereby married pregnant women were at lower risk of developing anemia as compared to those not married as a result of economic stability that influences dietary habits and access to health care.^[46,50–52]

The study revealed a noteworthy association between the pregnancy interval and anemia, with a considerable percentage of anemic women having had their previous pregnancy within 2 years compared to those who had it for 2 or more years. This can be explained by the fact that a woman with a short pregnancy interval may have insufficient iron storage, putting her at risk of becoming anemic in future pregnancies. This highlights the importance of family planning interventions for postpartum women.

This result is similar to the polynomial analysis conducted in 21 countries which shows that in northern Nigeria, women with a short birth interval were 2.37 times more likely to have anemia than the others.^[53] Similar results were also reported in Ethiopia.^[54] However, in the study conducted in Brazil and central Sudan, the pregnancy interval showed no association with anemia.^[55,56] More robust studies in this area are needed to confirm the association.

The study demonstrates a significant association between women’s knowledge and anemia, with low-knowledge groups having greater rates of anemia than knowledgeable groups. This suggests that knowledgeable women can recognize the risks associated with anemia and may be prompted to take preventative measures. Therefore, interventions focusing on raising awareness about anemia prevention could enhance women’s health-seeking behaviors. This study aligns with findings from a similar study conducted in Ethiopia among pregnant women attending an antenatal clinic, which also showed a significant association between anemia and women’s knowledge of the condition.^[57,58]

Women’s attitudes in this study have a significant effect on their anemic condition. Negative attitude was associated with a higher prevalence of anemia. This finding suggests that the cultural practices and beliefs of women in this community which shape their attitude are hindering them from healthcare services utilization, adherence to treatment, or consuming iron-rich food. This finding is similar to other studies findings conducted in other parts of Tanzania,^[59,60]. However other studies reported a higher rate of anemia despite positive attitude.^[61,62] The difference could be due to differences in the assessment and measurement of attitude in these studies.

As for the preventive services provision in healthcare facilities, this study shows that all healthcare facilities adequately provide the ITN and IPTp. This is encouraging provided that malaria prevalence in our setting is still high.^[63] However, the assessed facilities fall short in the provision of mebendazole and FEFO. This creates a serious concern because during pregnancy the demand for iron increases for the growth and development of the fetus^[64] and its deficiency poses serious problems for both the mother and the unborn child.^[65] On the other hand, failure to take mebendazole may increase the risk of worm infestation that brings poor birth outcomes such as low birth weight.^[66] Studies have shown that mothers who receive deworming medication during pregnancy reduce their child’s risk of death by 14% within the first 4 weeks of birth, and help pregnant women avoid having a low birth weight baby.^[67]

For facility preparedness, all of the facilities assessed are not well prepared for the provision of anemia preventive services. The most available and adequate commodities in all healthcare facilities are the ITN and IPT. This can be explained by the fact that Tanzania has established the National Voucher Scheme that has subsidized ITNs to pregnant women and newborns. Each expectant woman attending an antenatal clinic receives 3 US dollars literally known as the “Hati Punguzo” voucher (which covers 75% of the ITNs cost).^[68] Moreover, the ITNs are readily available throughout Tanzania because of this national program.^[69] However, the program currently lacks effective organization in the provision of FEFO and Mebendazole, leading to stock shortages of these essential commodities.^[70] The national authorities need to develop strategies to ensure the availability of these items at no cost, given their critical importance to pregnant women.

Healthcare provider training is also inadequate. About 16.7% of healthcare facility staff providing ANC services lack specific training packages. Moreover, despite having the established guidelines, only 33.3% of the facilities apply them. The findings aligned with the research conducted in Bangladesh, which revealed inadequate adherence to guidelines among healthcare providers delivering ANC services.^[71]

5. Conclusion

The prevalence of anemia among pregnant women in the third trimester in the Dodoma region is alarming and it is predicted by being at a young age. The protective factors are being married, having a college education level, having a pregnancy interval of more than 2 years, having 8 or more antenatal visits, having adequate knowledge, and positive attitude. All facilities provide all the recommended anemia preventive services but FEFO and Mebendazole have lower coverage. All facilities are not well prepared for the provision of preventive services, and the items that fall short in most of the facilities are FEFO and Mebendazole. Since anemia persists and stockouts make FEFO and mebendazole insufficient, facilities should conduct regular stock counts and forecast their commodities needs. Additionally, Tanzania policy should prioritize low-cost commodities to make services more affordable.

5.1. Limitation of the Study

This study adopts a sample size calculation formula for proportion estimation and did not consider the sample size calculation for regression, this may have led to an underestimation of the sample size enough to explore the determinants. To address this limitation, researchers employed robust data analysis techniques, specifically adjusting for confounders using logistic regression. This approach enhanced the reliability of the findings despite the constraints of a limited sample size. Furthermore, the study focused exclusively on pregnant women attending clinics, excluding those from the broader community, which limits the generalizability of the results. Additionally, most predictors discussed are more noticeable in severe anemia, but our study did not have any case of severe anemia. As a result, these findings should be interpreted with caution.

Acknowledgments

The authors would like to extend their Acknowledgments to the University of Dodoma for its valuable support during the accomplishment of this work.

Author contributions

Conceptualization: Twilumba Charity Nyagawa.

Data curation: Twilumba Charity Nyagawa.

Formal analysis: Twilumba Charity Nyagawa, Saada Ali Seif, Fabiola Vincent Moshi.

Investigation: Twilumba Charity Nyagawa.

Methodology: Twilumba Charity Nyagawa, Saada Ali Seif, Fabiola Vincent Moshi.

Resources: Twilumba Charity Nyagawa.

Supervision: Saada Ali Seif, Fabiola Vincent Moshi.

Validation: Saada Ali Seif, Fabiola Vincent Moshi.

Visualization: Twilumba Charity Nyagawa, Saada Ali Seif, Fabiola Vincent Moshi.

Writing – original draft: Twilumba Charity Nyagawa, Saada Ali Seif, Fabiola Vincent Moshi.

Writing – review & editing: Twilumba Charity Nyagawa, Saada Ali Seif, Fabiola Vincent Moshi.