The study was conducted in Dembecha district which is one of the 15 districts in the West Gojjam Zone, Amhara Regional State in Northwest Ethiopia. It is located 350 km from Addis Ababa in the Northwest direction, 203 km from Bahir Dar in the South direction, and 38 km from Finote Selam in the East direction. There were 6 health centers and 27 health posts (the lowest administrative primary health care unit that provides basic health services to the community in Ethiopia), which provide basic health services to the people of the district by focusing on maternal and child health. The climatic condition was 83% woyna dega (temperate zone with an altitude of 1,500-2500 m above sea level) 11% dega (vegetated cool zone with an altitude of over 2,600 m above sea level) and 6% Qola (hot zone with an altitude of below 1,500 m above sea level). The population of Dembecha district mostly generates income through agriculture. The total population of the Dembecha district was 175,520 in 2022. There were 40,818 households in the district. In 2022, the estimated number of children between 12 and 59 months old in the district was 18,309 (24).

A community-based cross-sectional study design was employed from April 19 to May 25, 2022. The study population was all randomly selected children aged 12–59 months in Dembecha district, during the study period. All children aged 12–59 months in the randomly selected kebeles (the smallest administrative unit in Ethiopia) during data the collection period were included in the study. Children with visual or hearing impairment were excluded from the study since they could not perform the specific developmental domains during the assessment.

The sample size was calculated using a single population proportion formula by considering 29.4% of the developmental delay taken from a previous study in Gurage Zone, Southwest Ethiopia (25) and 95% confidence interval, 5% marginal error, a design effect of 2, and 10% non-response rate which resulted in 702.

A multistage sampling technique was used to select the study participants. First, one urban kebele and seven rural kebeles were selected using a simple random sampling method. Then, the samples were distributed proportionally based on size. Finally, participants in each kebeles were selected using a systematic sampling technique after calculating the sampling interval for each kebeles The sample frame was the list of households with children 12–59 months of age. A list of households with children was obtained from a family folder, found in the health post. In households with two children less than 5 years of age, one was selected randomly by lottery. A revisit of three times was made in a case where eligible respondents were not available by the time of data collection and those who were not available after the revisit were considered as non-respondent.

The dependent variable was the child’s developmental status categorized as normal or delayed.

The independent variables were:

Developmental delay: It was defined as a child whose ASQ-3 score fell below the cut-off points for their age in any ASQ-3 domains. The five critical domains in a child’s development are communication, gross motor, fine motor, problem-solving, and personal social (9, 26) (Table 1).

Stunting: Children with a height-for-age at Z-score below −2 SD of the median value of the World Health Organization (WHO) international standard reference (27).

Wasting: Children with a weight-for-height Z-score below −2 SD of the median value of the WHO international standard reference (27).

Underweight: Children with weight-for-age Z-score below −2 SD of the median value of the WHO international standard reference (27).

Minimum dietary diversity: The percentage of children 12–59 months who received foods from 4 or more food groups during the previous day. The seven food groups used for tabulation of this indicator are grains, roots, and tubers; legumes and nuts; dairy products (milk, yogurt, and cheese); flesh foods (meat, fish, poultry, and liver/organ meats); eggs; vitamin A-rich fruits and vegetables; and other fruits and vegetables (28).

Complementary feeding: The process of introducing solid, semi-solid, or soft foods in addition to breast milk at 6 months of age (28).

Minimum meal frequency: Percentage of children 12–59 months of age who consumed solid, semi-solid, or soft foods at least the minimum number of times during the previous day. The minimum number of times was 4 times of meals per day for 12–59 months of children (28).

A structured paper-based questionnaire was used to collect data on socio-demographic, maternal-related factors, child health-related factors, nutritional variables, and food access at the household level. It was constructed by adapting and modifying the Ethiopia Mini Demographic and Health Survey (29) 2019 and previous research on a similar topic (25, 30). First, the English version questionnaire was prepared. Then it was translated to the native language Amharic and was back translated to the English language.

The child development assessment was done using the ASQ-3. The ASQ-3 has five subscales: communication, gross motor, fine motor, problem-solving, and personal-social. The ASQ-3 was answered with 10 points for “yes,” 5 points for “sometimes,” and 0 points for “not at all.” Each child was asked 30 questions and assessed out of 300 scores; and six questions out of 60 scores for each development milestone. Each domain was classified into three (high risk for developmental delay, needs monitoring, and on schedule) for each age category based on ASQ-3. Finally, child development was categorized as developmental delay and normal development (needs monitoring and on schedule). Developmental delay refers to a high risk of developmental delay whereas normal development refers to a low-to-moderate risk (9, 25, 26) (Table 1).

The ASQ-3 had a validity of 0.83–0.88, reliability of 0.90–0.94, sensitivity of 0.38–0.91, and specificity of 0.39–0.95 (26, 31). This tool was used in a study conducted in the Wolaita Sodo and Gurage zones in South Ethiopia. Also, the ASQ-3 developmental assessment tool was used by 56.3% of health professionals who were working at public hospitals in Addis Ababa, Ethiopia, 2018 (32).

Children’s dietary diversity was assessed with 7 food groups by using the WHO infant and young child feeding indicator assessment tool with some modifications to fit the context based on the last 24-h recall method (28).

Data was collected from caregivers or mothers of the children by two public health professionals and three health extension workers who could communicate well in the local language. The face-to-face interviews were used to collect data for socio-demographic factors and child-related factors. However, most parts of the developmental questions were assessed when a child performed or failed the activity required. The supervisors had oversight over the data collection process and the data collectors and supervisors had regular meetings with the principal investigator at the end of each day of the data collection period.

Anthropometric measurements for height/length were collected as per the World Health Organization (WHO) guidelines. A portable stadiometer was used to measure the height of older children (above 2 years) and a calibrated length board was used for younger children (below 2 years). Older children were measured in a standing position, while younger children below 2 years old were measured in a recumbent position. Children’s height and weight were measured without shoes, hats, and hair ornaments. When measuring height, a child’s head, shoulders, buttocks, and heels were attached to the vertical surface of the stadiometer. The height measurement was recorded to the nearest 0.1 cm. The weight of each child was measured while barefoot and light clothing and recorded to the nearest 0.1 kg using weighing scales (33). All variables apart from anthropometric measurements and observed ASQ-3 measures were self-reported by the mother/caregiver of a child.

The quality of the data was maintained by translating the English version questionnaire into the Amharic language, then it was translated back to English to check its consistency. To assess the appropriateness of wording, clarity of the questions, and respondent reaction to the questions, a pre-test was conducted on 5% of the calculated sample size in the Dega Damot district which had similar basic socio-economic characteristics, and the necessary amendments were made. Before data collection, data collectors and supervisors were trained for 1 day by the principal investigator about the overall purpose of the study, how to approach the study participants, and the way to collect data. During the data collection time, close supervision and monitoring were carried out by supervisors and the principal investigator to ensure the quality of the data. Finally, the supervisor and investigator checked the data for accuracy, clarity, consistency, and completeness daily.

The collected data was checked for completeness. The pre-coded data was entered into Epi data version 3.1 statistical software. Then the data was exported to SPSS version 25 for analysis. Descriptive analysis such as frequency, percentage, mean and standard deviation was computed, and the results were presented using texts, crosstabs, tables, and figures. The child’s developmental status was computed for each milestone as delayed and normal development and the five milestones were combined to determine the child’s overall developmental status. Finally, child development was categorized as developmental delay and normal development. WHO Anthro version 1.0.4 software was used to convert the anthropometric measure weight, height, and age value into Z-score of the indices HAZ, WAZ, and WHZ. Children whose height-for-age, weight-for-height, and weight-for-age < −2 SD from the median of the reference population were considered stunted, wasted, and underweight, respectively. Binary logistic regression was performed to identify factors significantly associated with child development delay. Those variables with a p-value <0.25 in bivariable analysis were entered into the multivariable analysis to control for possible confounding variables and to examine the association. The strength of association was measured using the adjusted odds ratio with 95% confidence intervals. Statistical significance was declared at a p-value of <0.05. A good fitness of the model was observed from the Hosmer Lemeshow test. The absence of multicollinearity was checked by using variance inflation factors and tolerance. The variables variance inflation factor and tolerance were < 10 and > 0.1, respectively.

An ethical clearance letter was obtained from the Research Ethics Committee of the College of Health Science of Debre Markos University (Reference Number: HSC/R/C/Ser/PG/Co/118/11/14). A written official letter along with the ethical clearance was submitted to Dembecha District Health Office by the principal investigator to get permission from the Health office before the data collection. In addition, written informed consent was taken from the mothers of the children. While conducting the study, children who were found to be malnourished, their mothers were informed and advised about the measures to be taken, and were linked to the nearby public health services for treatment and support. The obtained information was kept confidential by locking it with a password.

About 96.51% of children were born in a health facility. The majority of children (65.0%) were born at term. More than two-thirds (69.9%) of the children were born via spontaneous vaginal delivery. Most of the mother (95.4%) did not develop hypertension during pregnancy. Half of the mothers (51.3%) had no history of illness or disease during pregnancy. More than half of the mothers (57.9%) drank alcohol while pregnant (Table 3).

Three hundred eighty-four (55.8%) children were females. The mean age of the children was 31.0 months (SD = 12.4 months). Two-thirds (66.6%) of the children were toddlers aged 1–3 years. More than three-fourths (87.8%) of children were born with normal birth weight. In terms of feeding, 373 (54.2%) began complementary feeding at the age of 6 months. Less than half of children (41.0%) eat less than four times per day. In the previous 24 h, 57.0% of children met the minimum dietary diversity (four food groups; Table 4).

Concerning the nutritional status of children, about 27.8% (95% CI: 24.4, 31.3%), 9.2% (95% CI: 7.1, 11.6%), and 4.5% (95% CI: 3.1, 6.3%) were stunted, underweight, and wasted, respectively (Figure 1).

The mean ASQ-3 score was 155.4, with a standard deviation of 25.7, a minimum score of 80, and a maximum of 270. The mean ASQ-3 score for each domain ranged from 27.2–35.9, with a standard deviation from 5.5–8.2, and a range from 10 to 55 for communication,10 to 60 for fine motor, and 15 to 60 for the remaining of the milestones. This study found that 26.7% (95% CI: 23.5, 30.2%) of children have developmental delays. It is expressed as communication delay (17.2%), gross motor delay (20.6%), fine motor delay (17.4%), problem-solving delay (16.7%), and personal social delay (18.2%). The remaining 73.3% with 95%CI (69.8, 76.5%) were on normal development to their age (Table 5; Figure 2).

About 28 (4.1%; 95% CI: 2.7, 5.8%) children had developmental delays in all the five developmental domains.

Maternal age, marital status, weight at birth, gestational age, time of initiation of complementary feeding, birth order, place of delivery, mode of delivery, child age, stunting, underweight, parity, meal frequency, dietary diversity, and family size were associated with child development delay in bivariable logistic regression analysis and identified as candidates for multivariable logistic regression analysis (Table 6).

In multivariable logistic regression analysis; toddler age of the child, LBW, cesarean section mode of delivery, preterm delivery, initiation of complementary feeding before 6 months, stunting, inadequate meal frequency, and inadequate dietary diversity were significantly associated with developmental delay at p-value of 0.05 level of significance (Table 6).

Children born with a LBW were nearly 5 times more likely to experience developmental delay than children with a normal birth weight (AOR = 4.90; 95% CI: 2.14, 11.48). Toddlers (1–3 years) were 2.6 more likely to experience developmental delay than children who were in the age of preschool (3–5) years (AOR = 2.60; 95% CI: 1.42, 4.87). Children born by cesarean section were nearly 9 times more likely to have developmental delay (AOR = 8.60; 95% CI: 3.93, 18.65; Table 6).

Children who started complementary feeding before 6 months of age were 8.4 times more likely to have developmental delay than children who started at 6 months (AOR = 8.40; 95% CI: 3.61, 19.63). Preterm children were 2.5 times more likely to experience developmental delay compared to children who were delivered at term (AOR = 2.50; 95% CI: 1.28, 4.74). Children who had meal frequency of <4 times were 3 times more likely to develop developmental delay compared to children who had meal frequency of ≥4 times (AOR = 3.20; 95% CI: 1.74, 5.94; Table 6).

Children who had inadequate dietary diversity were 3 times more likely to develop developmental delay compared to children who received more than four food groups (AOR = 3.10; 95% CI: 1.68, 5.85). The odds of developmental delay among stunted children were nearly 3 times higher as compared to normal children (AOR = 2.90; 95% CI: 1.67, 5.22; Table 6).