The overall level of knowledge towards blood donation was found to be 56.5 % which is higher than community based study conducted in the city of Mekelle (49 %) [11]. However, it is lower than another study conducted among health science students in Addis Ababa (83 %) [12]. The difference in socio-economic status with residents of Mekelle and in educational status with the health science students might explain the discrepancy with the above findings. About 53.8 % of participants knew that people can donate every 3 months, this is higher than other studies conducted in Benin, (21.5 %), Chennai (51.2 %) and Mekelle (43.6 %) [11, 15, 16]. The majority of them reported voluntary blood donation as the best source of blood donation. This is consistent with finding from Chennai [16], however it is higher than studies conducted among health workers (72.2 %), and physicians in Benin (80.7 %) [15–17]. Less than half (47 %) of adults knew the risk of transmission of disease through blood transfusion which is lower than that of the study conducted in Benin (95.7 %) [15] and, higher than that of the study conducted in India [18].
Age, sex, educational status and source of information were found to be independent predictors of knowledge of blood donation. Male study participants were more knowledgeable towards blood donation which is in line with findings from Karachi and North India [19, 20]. In Ethiopian context, males are more accessible to information and spent most of their time out of their home than females. Having at least primary education was significantly and positively associated with the knowledge of blood donation, which is supported by a study conducted in Sikim, India [13]. This could be because more educated people might be in a better position to access the media and availability of awareness creation at primary and secondary school and higher educational institutions. Participants who used radio as a source of information were less knowledgeable than those who used newspaper and/or internet. This might be partly explained by the limitations of radio programs on addressing the intended goals. On the other hand the low level of knowledge found in a radio listeners might be due to their limited access, techno illiteracy and their limited educational preparedness to understand, analyze and absorb the content of the message.
The composite measure of attitude based on mean score indicates that 52.2 % of the respondents had favorable attitudes towards blood donation. This is better than a study conducted in Karachi (42 %), however it is lower than similar studies conducted in India (87.3 %), Mekelle (61 %) and Addis Ababa (68 %) [11, 12, 19, 21]. More than three-fourth of respondents had intention to donate blood in the future which is lower than other studies conducted in India (90 %) and Addis Ababa (100 %) [12, 18]. Studies conducted in India and Addis Ababa carried out among health science students showed that knowledge of blood donation was higher as a result of their profession.
The present study found that educational status, average monthly income, source of information, and knowledge were significantly associated with favorable attitude towards blood donation. Participants who had higher monthly income were more likely to have favorable attitude than lower income groups. This might be because those who have higher income may access better information sources. Radio listener had lower chance of having favorable attitude compared to newspaper reader and internet users. This might be partly explained by the limitations of radio programs on addressing the intended goals. On the other hand, the low level of knowledge found in a radio listeners might explain the unfavorable attitude of them as knowledge is an independent predictor of attitude.
In addition, about 16.1 % of respondents reported at least one history of blood donation which is lower than similar studies conducted in Benin [15], Northern Nigeria [22], South India [21], and in Addis Ababa [12]. However, it is higher than other studies carried out in India (13.9 %) [18] and Mekelle (12 %) [11]. The difference in the practice of blood donation could be due to variation in the setup of study settings since some of the studies were conducted among health professionals and some others were at school level. Nearly two-third of blood donation were practiced as voluntary blood donation. This is supported by the study conducted in South India (64.1 %) [21]. About 4.8 % of blood donors were regular donors, which is very low compared to similar study conducted in Benin (13.9 %) and facility based study from Addis Ababa (42.2 %) [12, 15]. The major reasons reported from those who didn’t practice blood donation were inability to think of it (33.8 %), lack of opportunity to donate blood (32.7 %), and lack of time (11.1 %). Global researchers also concluded that people are not donating blood because nobody approached them for donation, lack of information, unfit to donate, a need to donate for a friend or relative in future, fear of needle and knowing their viral status, the donated blood may be sold, non-remuneration, ignorance and their religion [12, 15, 17, 18, 21, 23].
Age, educational status, knowledge and attitude were significantly associated with the practice of blood donation. Older age groups were positively associated with the practice of blood donation. This is similar with findings from Karachi and Iran [21, 24]. This might be attributed to increased personal experience from donating blood. Having certificate and above educational level was associated with increased practice of blood donation which is comparable with the study carried out in Addis Ababa [12]. This might support the fact that education can positively influence the knowledge and attitudes of blood donation. In addition, being knowledgeable and having favorable attitude were found to be independent predictors of blood donation. Knowledge of blood donation is a pre-requisite in obtaining access to and providing voluntary blood donation timely and effectively. It is also an important tool in avoiding fear and building positive attitude of blood donors. This is supported by a similar study conducted in Mekelle [11].
The fact that this study was conducted at the community level could be mentioned as the strength of the study. The nature of cross-sectional study which is not possible to establish cause-effect relationship between the explanatory variables and outcome variables, and the possibility of social desirability bias were among the limitations of this study.