The study was conducted in the North Shewa zone, Oromia region, Ethiopia. The zone has a total area of 10,322.48 km² and a population density of 138.66 people per km², with 13 rural districts and two town administrations. Furthermore, the Zone has a total population of approximately 1,639,586 people, of whom 717,552 are men and reside in 521,506 households. Fiche town – the zone's capital – is 112 kilometers north of Addis Ababa –Ethiopia's capital. Additionally, the zone has 64 health centers and five public hospitals that provide health care services to the community. A community-based cross-sectional study design with a concurrent mixed-method approach was conducted from April 1, 2021 to May 30, 2021.

A source population consisted of all adults over the age of 18 who were permanent residents (individuals who live at least 6 months) of a selected town in the North Shewa zone. A study population was randomly selected adult residents for a quantitative study. A qualitative study was conducted among purposely selected residents, health care providers, and public health experts.

The sample size was determined using a single population proportion formula in Epi Info STAT CALC. version 7.2 based on the assumptions of a 95% confidence level, 5% margin of error (d), and a 50% proportion (p) of knowledge of NCDs. Because no similar study had previously been conducted in Ethiopia, 50% was taken. After applying a design effect of 2 and a 10% non-response rate, the final sample size obtained was 846. Additionally, a total of five sessions of Focus Group Discussions (FGDs) were conducted among residents in the selected districts. Each FGD was composed of 10 residents. Additionally, 10 In-Depth Interview (IDI) was conducted among health care providers with different professionals and public health experts. Regarding a sampling technique, a multistage sampling technique was used. Towns in the zone were purposely selected since the prevalence of NCDs is high among urban populations. Thirty percent of the towns in the zone were randomly selected by lottery method, namely Fiche, Kuyu, Debretsige, Mukaturi, and Gundomeskel town. Consequently, 30% of their kebeles (small administration units) were included in the study. The sample size was proportionally allocated to each town. Finally, a computer-generated simple random sampling technique was used to select households from the kebeles using house numbers. In a household with more than one adult individual, the lottery method was used to select one of them. For the qualitative part, the purposive sampling technique was employed considering maximum variability with the assumption of obtaining relevant data to supplement the research objectives. The intention was to explore detailed information on NCDs knowledge and healthy lifestyle in society.

The data collection tool was developed after reviewing previously done studies and it has four sections including; socio-demographic, exposure to NCDs information, knowledge of NCDs, and practices of healthy lifestyles (15, 29–32). The knowledge measuring tool was reliable in this study (Cronbach-alpha = 0.72). Ten experienced BSc nurses were recruited for data collection, and onsite supervision was provided by five BSc in public health supervisors to facilitate data collection procedures on a daily basis. Data were collected through a face-to-face interview by using a pre-tested questionnaire. Additionally, FGDs and IDIs were used to get participants' experiences related to knowledge of NCDs. By using an open-ended guiding list of questions, five sessions of FGDs were undertaken to explore the knowledge of NCDs and the healthy life practice of the community. The FGDs were moderated by an experienced health professional and note taker. During the discussion, notes were taken and their voices were recorded using a tape recorder. Similarly, 10 sessions of IDIs were conducted among health care providers and public health experts.

The dependent variable of the study was the level of adequate knowledge of NCDs. Knowledge was assessed by 32 questions related to respondents' knowledge about NCDs and their risk factors. Correct answers were given a score of 1 and incorrect answers were given 0. The total possible score ranged from 0 to 32. A cut-off level ≥60%, of the individual percentage scores, was selected as an indicator of adequate knowledge (15). As well, in the current study the tool has an acceptable reliability (Cronbach-alpha = 0.81). The explanatory variables were socio-demographic (age, sex, religion, marital status, educational status, occupational status, income, and ethnicity), exposure information about NCDs (getting information from media, getting information from health professionals, getting information from family members, having family members with NCDs and having friends with NCDs) and behavioral factors (physical activity, alcohol and tobacco use).

The collected data were checked for completeness manually and entered, cleaned, and checked by Epi data manager version 4.0.2 and then exported to SPSS version 23 for analysis. Descriptive analysis of different variables was done based on their nature. Bivariable analysis was done using binary logistic regression for all independent variables and a p-value < 0.25 was used to consider candidate variables for the multivariable analysis. Finally, a multivariable binary logistic regression analysis was done to control for possible confounders and to identify factors significantly associated with an adequate level of knowledge of NCDs at a p-value of < 0.05 with a 95% confidence interval adjusted odds ratio. Model fitness was checked using the Hosmer and Lemeshow goodness of fit model and it was fitted (p-value = 0.12). as well, the variance inflation factor was set at five to detect multicollinearity among explanatory variables, and no multicollinearity was found. Besides, before analyzing data, all FGDs and IDIs were transcribed in Afan Oromo text by replaying the recorded voice from tape and the notes taken during discussion. Then, the Afan Oromo version text was translated into English language. Different ideas in the text were merged in their thematic areas and thematic framework analysis was done manually. The results were presented in narration by triangulating with quantitative findings.

Ethical approval was obtained from the ethical review committee of Salale University. A permission letter was obtained from the respective district and kebele administration before data collection. Written informed consent was obtained from each study participant.

A total of 823 respondents have completed the study making a response rate of 97.3%. The mean age (standard deviation) of the respondents was 31.83 (11.04) years. Of the total participants, more than half, 443 (53.8%) were females. Regarding marital status, 497 (60.4%) of them were married. Approximately, one-third, 255 (31%) of them attended above secondary education. Furthermore, the majority, 628 (76.3%) of the participants were from the Oromo ethnic group. Additionally, nearly a fourth, 189 (23%) of them were a merchant (Table 1).

On the 32 items tool measuring respondents' knowledge about NCDs, the mean score (standard deviation) of respondents was 14.6 (4.34). Of the total, 279 (33.9%) of the participants had an adequate level of knowledge (Figure 1). Regarding the causes of NCDs, more than two-thirds, 568 (69.0%) stated that NCDs are caused by supernatural power (God). In line with this, a misconception about the cause of the disease was observed among FGD participants. For instance, a 31 years old male discussant said “… Eating vegetables can lead to chronic diseases such as diabetes and hypertension due to their low nutritional value and weakness in increasing protection against these diseases…” Regarding the onset time, about 446 (54.2%) of the study participants said that NCDs onset only at an older age (Table 2). Similarly, participants of IDI explained that there is no sufficient knowledge in the community regarding NCDs. For example, a 34 years old female clinical nurse said that “Because there is a lack of public awareness about these diseases, people seek medical attention after they develop complications…” Also, a male public health expert of 29 years old working at a health center explained that the community has no awareness about NCDs. He has elucidated that “In general, there is a lack of awareness among communities about their health status. They drink as much alcohol as they desire, and no one engages in physical activity... They are unconcerned about NCDs...”

In bivariable analysis, variables such as sex, age group, educational status, marital status, income, occupation, presence of TV, getting information from health professionals, having neighbors with NCDs, and presence of a family member(s) with NCDs were significantly associated with the knowledge of NCDs. In multivariable analysis, income, getting information from health professionals, owning a TV, having a family member with NCD, and being widowed/divorced were significantly associated with adequate knowledge of NCDs. Accordingly, respondents who earn a monthly income of ≥3,578 were almost two times more likely to have adequate knowledge about NCDs as compared with those who earn a monthly income of <3,578 (AOR = 1.68, 95% CI: 1.05–2.68). Respondents who received information from health professionals were more than twice more likely to have adequate knowledge of NCDs when compared with those do not receive information from health professionals (AOR = 2.29, 95% CI:1.53–3.45). Moreover, having family members with NCDs was associated with the knowledge of residents i.e., respondents who have family members with NCDs were more than two times more likely to have adequate knowledge of NCDs when compared to their counterparts (AOR = 2.23, 95% CI:1.49–3.35). Besides, respondents who possessed television were more than twice more likely to have adequate knowledge about NCDs as compared to those who don't possess television (AOR = 2.48, 95%CI:1.32–4.65). Additionally, the marital status of the respondents determines the level of NCDs knowledge. Thus, the likelihood of having adequate knowledge among divorced and widowed participants is 0.80 less likely as compared to unmarried participants (AOR = 0.20, 95%CI:0.07–0.59) (Table 3).

Concerning information exposure, all of the study participants heard about NCDs. The majority of the participants, 619 (75.2%) of them heard about NCDs from Television (TV) (Figure 2). Comparable to this a 32 years old male participant in FGD stated that “…I usually watch TV in my spare time, and there are advertisements on TV that promote healthy behavior such as reducing salt consumption, quitting smoking, and receiving immunization against diseases such as the human papillomavirus, which causes cervical cancer.…”

Of the total, 218 (26.5%) of study participants have a family member(s) with NCDs. This view is echoed by a 24 years old female FGD participant as; “… My father has been suffering from hypertension and diabetes mellitus. Once upon a time, he was admitted to the hospital due to an extremely high blood sugar level, for which I had been caring for him. And, a nurse told me that such diseases have their roots at a young age and advised me to take preventive measures such as eating vegetables, exercising regularly, avoiding high-fat foods, and getting regular health check-ups.…”

The prevalent NCD was hypertension accounting for 187 (22.7%). This is also evidenced by IDI results. A 41 years old male public health officer stated that “…I've been serving in this community for over 8 years, and the most common NCD I've observed is hypertension, followed by diabetes mellitus…”

As NCDs are one of the major health and development challenges we face today, investigating a related theme will lead to a deep understanding of the problem and searching for an effective solution like detection, screening, treatment, and palliative care. Particularly, Ethiopia's health system is working to implement awareness-raising programs on NCDs and risk factors for different segments of the population (22). Without knowledge about NCDs and their risk factors, it is difficult to achieve a reduction in the incidence and prevalence of NCDs (36). As a result, determining the level of knowledge about NCDs emphasizes the significance of conducting regular surveillance for NCDs' risk factors and initiating prevention programs (15).

All questionnaires were translated to the local language (Afan Oromo) by two independent bilingual translators and back-translated to English to guarantee consistency. Then it was pre-tested on 5% of the total sample size in the district which was not selected for actual study to evaluate readability, understandability, completeness, and reliability, and modification was made accordingly. Data collectors and supervisors were trained on the objectives of the study, how to collect data, and ethics. Filled questionnaires were checked for completeness and consistency on daily basis. Overall activities of the study were supervised by the supervisors and regulated by the investigators. Despite these, being a cross-sectional study, made it impossible to draw causal inferences. Additionally, a self-report technique was used to assess the knowledge, which may have resulted in recall bias and social desirability bias.