The study aimed to examine the prevalence of smoking, the kinds of substances children smoke, predictors of smoking, reasons for and factors that lead children to smoke, and regulation of smoking among children in Ghana. The findings indicate that 3.2% of children had ever smoked, and it is more prevalent among male children (5.5%) than female children (0.8%). On the one hand, the prevalence of smoking among children in this study was low compared to earlier studies in Serbia [15] and Malaysia [14]. For instance, Veres et al.‘s [15] in Serbia among high grade primary students found that 35% of students had ever smoked. Norbanee et al.‘ [14] study in Malaysia among primary school children found that 25% of children had ever smoked. On the other hand, the prevalence of smoking among children in this study was high compared to a study in Brazil [16]. Silva et al.‘s [16] study among children aged 7–17 in Brazil found that 2.4% of children had ever smoked. The differences in smoking prevalence among children could be attributed to differences in the sample population, respondents’ ages, and measurements. For instance, Silva et al.‘s [16], Norbanee et al.‘ [14], and Veres et al.‘s [15] studies were conducted among school going children, while this study was conducted among both school going and non-school-going children. Also, Veres et al.‘s [15] study was conducted among children aged 11–15, while Silva et al.‘s [16] and Norbanee et al.‘ [14] studies were conducted among children aged 7–17.

The study also found that male children reported higher smoking prevalence than female children. The multivariate analysis also showed that male children are more likely to have ever smoked. These findings are similar to those of earlier studies [1, 17, 18]. A plausible explanation is that males are more likely to engage in risky behaviour than females [19, 20].

Furthermore, smoking was more prevalent among older children than younger children. The multivariate analysis also showed that children aged 8–10 and 11–13 were less likely to have ever smoked. This finding supports earlier studies that smoking is more prevalent in older children than younger children [14, 16]. Older children may be more exposed to tobacco products and are, therefore, more likely to try smoking than younger children.

Children residing in the Coastal zone were more likely to have ever smoked than children living in the Middle belt and Northern zone. Greater Accra region is situated in the Coastal zone, and studies have found that the prevalence of smoking is highest in the region [1, 11]. Therefore, the high prevalence of smoking in the Greater Accra region may explain why children in the Coastal zone are more likely to have ever smoked than children residing in other ecological zones.

According to the study, cigarette smoking (78.3%) is far more common among children in the country than marijuana or shisha usage (18% and 3.7%, respectively). This supports earlier findings by St Claire et al. [21], which found that cigarette smoking is the most popular smoking habit among children. The results also highlighted sex differentials in smoking, with cigarettes and marijuana smoking being more prevalent among male children, while shisha smoking was more prevalent among female children. Shisha smoking was more prevalent among female children because of the perception that it is less harmful compared to cigarettes and marijuana [22]. Females are often less likely to take risks than males [19, 20].

The study also revealed that some children who used marijuana had a history of smoking cigarettes, supporting previous research that indicates an association between marijuana and cigarette use [23]. This finding suggests a connection between smoking cigarettes and switching to marijuana use.

Regarding the frequency of smoking, the majority of children (57.1%) had smoked once, 32.3% had smoked repeatedly, 5.6% were unable to recall how many times they had smoked, and 5% had smoked twice. Despite the low prevalence of smoking among children, it is worrisome that 32.3% had smoked repeatedly, and 5.6% could not recall how frequently they had smoked. Another concerning finding is that over 2 out of 10 children who had ever smoked also smoked marijuana. These findings urgently call for policy makers to strategise and strengthen their policies, programmes, and interventions to address smoking among children.

A little over four out of ten children smoked for fun and due to group culture. Children also occasionally smoked to get high, to kill boredom or to calm their nerves, or just out of curiosity. Also, children were influenced by friends; parents, family members, and neighbours who smoke; curiosity; and advertisements and videos to initiate smoking. These findings corroborate previous studies, which found that friends [14, 15, 17, 24], parents, family members, and neighbours who smoke [14, 15], curiosity [24, 25], and advertisements [3] influenced people to initiate smoking. These findings highlight the need for parents, relatives, and neighbours who smoke to be mindful of their smoking habits since it negatively impacts children’s smoking habits. Also, parents and guardians need to show interest in who their children spend time with or associate with. Policy makers should enforce the Tobacco Control Regulations 2016 (L.I. 2247), which regulates the advertisement, promotion, and sponsorship of tobacco products.

The multivariate analysis also found that children who live with their biological fathers are more vulnerable to smoking than those living with both biological parents. The lack of proper supervision by fathers may explain this phenomenon.

The qualitative findings further show that some children started smoking because they used to run cigarette deliveries for their family members, which offered them a chance to try it out. This situation strengthens the case for enforcing the Tobacco Control Regulations 2016 (L. I. 2247) and the Public Health Act 2012 (Act 851), which prohibits the sale of cigarettes to persons under 18. This finding supports an earlier study, which found that access to substance smoking products, such as e-cigarettes, influence smoking [26].

The qualitative findings point to various locations where children prefer to smoke, and many are hidden from the general public. Children’s smoking areas include deteriorating, incomplete, and abandoned structures, as well as bushes, beaches, dark places at night, bars and places far from their residences. Most children smoke in secret locations since they know smoking is considered a deviant act in Ghana. This finding supports Glenstrup et al.‘s [27] study, which found that adolescents who smoke feel discomfort when their smoking behaviour is exposed, so they keep their smoking habits hidden from the public.

The government has made some progress in passing laws, ratifying international treaties, establishing entities, raising public awareness, and putting in place controls over the use and distribution of tobacco products, which is typically what most children in the country smoke. Smoking among children is still prevalent despite the preventive measures taken and the government prohibiting the sale and use of tobacco products among children. According to the findings, this is typically caused by the lack of funding for government organisations to monitor and regulate measures that deter children from smoking. Although Ghanaian law forbids children from purchasing tobacco products, inadequate regulatory procedures allow children to access marijuana and tobacco products through unlawful channels. Although there is some compliance among specific retailers in urban areas, many tobacco product vendors in rural areas continue to sell to children out of ignorance or a lack of regard for law and order [28, 29].