This study used a quantitative observational cross-sectional analytical design. This is related to the study being aimed at comparing hookah users and non-users in terms of family functioning and family satisfaction.

The intention for this study was to obtain a heterogeneous sample from low and middle-high socio-economic areas. The economic criterion used for stratification in this study was based on the location of the area. That said, data were collected across townships, the Cape Flats and the northern suburbs of Cape Town, South Africa. These areas were randomly selected based on the previous geographical and socio-economic dispensations which took place during apartheid in the country. These areas clearly highlight the status of race as well as socio-economic status (SES). The township area used in this study is bordered by Jan Smuts Drive to the west and two separate highways, one to the south and another to the east. Additionally, data were collected from the area located east of the city centre on the Cape Flats and south of the highway. The area predominantly serves as a residential area but also comprises a railway station, stadium as well as industrial and commercial zones. Furthermore, the suburb in which data were collected forms a part of the greater Cape Town metropolitan area and is surrounded by farms producing wine and wheat.

Study participants were recruited by means of systematic and convenience sampling. Initially, a systematic ‘door-to-door’ sampling or recruitment process was used, which was adapted from the process shown by Flynn et al.²¹ Every fifth house from the starting point was approached in the recruitment process. However, once data collection started, the systematic sampling approach deemed challenging due to people not being at home or unwilling to participate. The researchers then began the process of convenience sampling by randomly selecting participants in the same street and/or community. This is related to researchers accessing participants based on their availability. The reason for these sampling techniques as deemed appropriate was the time period set aside for data collection, as data were to be collected within 2–3 months. Additionally, individuals who agreed to be a part of the study represented an entire family. Therefore, one participant represented an entire household. Participants were included in the study if they were the parent of at least one child aged between 0 and 17 years. This includes both parents. Furthermore, individuals were excluded from the study if they were secondary caregivers to a child, such as a grandparents or other family member, or had children aged 18 years and older. In low socio-economic environments, fieldworkers involved with these communities were used. These fieldworkers were informed to approach and invite a participant from every fifth family from their starting point in the identified areas. Students were used as fieldworkers in middle-high socio-economic areas, with the same sampling strategy applied by selecting every fifth family from their starting point. This sampling strategy was used to ensure that fieldworkers were not biased in their selection of participants as well as frequent debriefing between fieldworkers and the primary investigator to minimise bias.

The questionnaire used was compiled to address family functioning and family satisfaction among hookah users and non-users. Demographical information such as gender, age, education level and family structure as well as a section that focused on hookah use in the family was included in the questionnaire. Furthermore, two scales, Satisfaction with Family Life Scale and Family Functioning (the Family Assessment Device) were used to understand family functioning and family satisfaction of the participants.

The raw data were double-entered into the Statistical Program for Social Science (SPSS) and coded and cleaned in order to check for errors. Data analysis included descriptive statistics and inferential statistics. The inferential statistics used were independent t-tests to test significant differences concerning family functioning and family satisfaction among hookah users and non-users.

On receiving ethics clearance and the commencement of data collection, researchers sought informed consent from all participants. This consisted of participants being presented with information pertaining to the details of the study and assuring that all information obtained from them would be confidential and they have the right to withdraw from the study at any time without any consequence. Participation was completely voluntary. Furthermore, no harm came to the participants during the data collection process.

Ethical approval for the study was obtained from the Humanities and Social Science Research Ethics Committee of the University of the Western Cape (Reference number: HS/16/3/48).

The profile of the final sample predominantly comprised the following: female participants: 54%, English-speaking participants: 50%, average age of participants: 33 years, education level of participants: Grade 11 and employed participants: 54.7%. Based on the original issue of 1400 questionnaires, the return rate for the current study was 85.21%. Therefore, the final sample comprised 1193 participants, each participant representing a family of which 72% were from low SES areas and 28% from middle-high SES areas. The difference in sample size for the low and middle-high SES areas was mainly caused by lack of participation in the middle-high SES area.

Overall, the current study found that 34% of participants indicated smoking of hookah pipe in which the age of onset was 16.5 years (X¯ = 16.48). Majority of the participants (37%) who responded to the question related to hookah pipe use were from low SES areas as opposed to middle-high SES areas (26%).

In the current study, it was found that various substances were used in the family. These substances include marijuana, also known as dagga (21%); mandrax (3%); heroine (2%); cocaine (4%); alcohol (49%); tik (7%) and glue (2%). The following aspects were also found in the family: (1) knowledge of the harmful effects of hookah pipe (65%); (2) hookah is predominantly combined with alcohol (30%); (3) hookah pipe is mostly used at parties (72%); (4) hookah use in the family (60%) and (5) acceptability of use in the family (24%). A substantial proportion of participants (17%) indicated that children were present when hookah pipe was used in the family. This study also found that children as young as 2–6 years (3%) use hookah pipe. As previously indicated, one participant represented the entire household. That said, when focusing on hookah use in the family, 62% of the participants stated that hookah pipe was not used as a means of socialising in the home whereas 28% indicated that hookah pipe was in fact used as a means of socialising n the family (see Table 1).

Furthermore, participants were asked about parents’ acceptance pertaining to smoking the hookah pipe. Although the majority indicated that the hookah pipe use was not accepted in the family, those who indicated its acceptance stated that parents were more consenting of their own family members using the hookah pipe in the family (24%) than non-family members (20%). This means that parents were accepting of their own family members smoking hookah pipe in the home as compared to non-family members such as friends.

The manner how the family functions in terms of cohesion, conflict, expressiveness, permissiveness and being satisfied is as follows. The results shown in Table 2 suggest that when families are cohesive (r = 0.53; r² = 0.28) and expressive (r = 0.47; r² = 0.16), family members are be more satisfied with others in the family. This suggests that cohesion in the family accounts for 28% and expressiveness for 22% towards family satisfaction. In addition, results also suggest that if conflict (r = −0.40; r² = 0.16) and permissiveness (r = −0.19; r² = 0.04) are present in the family, its members are less satisfied with others in the family. This is associated with conflict in the family accounting for 16% and permissiveness accounting for 4% towards family satisfaction. These results are related to the study being focused on family function and family satisfaction between hookah users and non-users.

The results in Table 2 show that there are significant differences between hookah users and non-users based on family functioning and being satisfied in the family. Hookah users were less satisfied in the family (X¯ = 3.44), had more permissive family functioning (without rules and discipline) (X¯ = 2.34) and had less expressiveness (X¯ = 2.74) and cohesion (X¯ = 2.94) in the family but greater conflict (X¯ = 2.28). It was observed that family relationship was a challenge in the families of hookah users. These results provide insights into the potential risk factors of hookah use in a family.

Establishing prevalence rates in the study was important. Prevalence rates indicate the extent of problem and are significant in implementing any sort of intervention or prevention strategy as it provides an indication of the individual having the problem, thus who should be the target group. Nationally, studies report rates of prevalence as between 18%¹⁴ and 60.9%.²³

Of concern in the current study was the age of onset of smoking hookah, which was younger than indicated in the previous studies conducted in South Africa⁹ but slightly higher than as shown in international research,²³ where majority of participants started smoking hookah pipe between 13 and 15 years of age. In line with the identified age of onset of hookah use was the prevalence of hookah use among children and adolescents. This finding suggests a similar prevalence of use as observed in a previous pilot study in terms of increasing age per category as well as use among young children. This then particularly becomes a public health concern, as young people who start smoking tobacco at an early age are at a risk of nicotine addiction as opposed to those who start later.⁹ Global research mainly reflects its use among adolescents¹ and not younger children, which could be a phenomenon pertinent to South Africa, and specifically to the Western Cape.

In this study, the family is situated as a context for use. Although parents are the ones who should be held accountable for the care and wellbeing of their children, families provide exposure and experience to child and adolescent to use hookah as indicated in the pilot study conducted by Roman et al.⁹ In particular, majority of the participants indicated that hookah pipe was not used as a means of socialising in the family; however, for a smaller percentage hookah was used in the family as a form of socialising, and was acceptable by family members. This is in line with the research which found that parents were more accepting of their children using hookah in the family, rather than going outside to use it.^24,25 This is further supported by a qualitative study conducted in Iran which focused on the factors contributing to the first hookah use by women.¹¹ In the stated study, an in-depth perspective was provided of the traditions of socialising in families in Iran by preparing and using hookah pipe. Previous research in other countries suggest that if one family member smokes hookah pipe, there is a stronger likelihood that children would follow.²⁰ This is in line with the small percentage of children aged between 2 and 6 years who were found to smoke hookah pipe. This then becomes a blueprint for permissibility and acceptability from the family to the community to the society. Although for the majority of participants the hookah pipe was not used in the family home as a means of easier communication or socialising, it should be noted that in other families it enhanced family interactions.

This research is the first study to provide information regarding the status of families in terms of family functioning and family satisfaction among hookah users and non-users. Although family functioning is pertinent in the development of children and adolescents, and also in moderating what children are exposed to,⁹ this study found families’ ability to function effectively somewhat problematic. More specifically, it found that the families of hookah users had less cohesion, expressiveness and family satisfaction compared to families of non-users, in turn indicating that the families of hookah users experience more conflict and permissiveness in their homes. This could be related to these families’ likelihood of being exposed to violence, as well-functioning families were found to moderate exposure to and perpetration of community violence.⁹ Therefore, when comparing the results of family functioning and family satisfaction between families of hookah users and non-users, it is suggested that the use of hookah pipe is a result of poor family functioning, which in turn influences family satisfaction. However, conversely this does not suggest that poor family functioning and family satisfaction cause the use of hookah pipe.