Understanding the influence of parent-child conversations about dangers of drug use on cannabis use among high school students in Barbados

Abstract

BACKGROUND:

Parent-child conversations are very important in influencing adolescent/youth behaviours. Objective of this study was to assess association between parent-child conversations on dangers of drug use and cannabis use among secondary school students in Barbados.

MATERIALS AND METHOD:

Data from the Barbados National School Survey (NSS) 2013 were analysed. Data were collected from a nationally representative sample of high school students drawn from 2^nd to 6^th forms across public and private secondary schools in Barbados in 2013. Descriptive analysis was done for sociodemographic characteristics and parent-child conversations. Multivariate logistic regression was done to assess the association between parent-child conversations on dangers of drug use and cannabis use among the subjects before and after adjusting for covariates.

RESULTS:

Parent-child conversations on dangers of drug use were significantly associated with past year cannabis use both before and after adjusting for covariates (crude odds ratio [COR] = 1.20, 95%CI: 1.04-1.38; adjusted odds ratio [AOR] adjusted odds ratio = 1.35, 95%CI: 1.15-1.58) but with students’ past month cannabis use only in the unadjusted logit model. Being 15-16 years, male, enrolled in 4^th form, and belonging to widowed family structure was associated with reduced odds of past year and past month cannabis use.

CONCLUSION:

Significant positive association between parent-child conversations on the dangers of drug use with cannabis use is counter-intuitive and suggests that it is not just conversations but the type, content, and quality of conversations that matter in reducing students’ inclination to use cannabis.

Introduction

Parent-child conversations on dangers of drug use have been considered as one of the factors that affect children’s behaviour towards substance use[1,2] King and Vidourek[2] opined that when parents talk to their children about illicit substance use, it helps them to reduce risky behaviours relating to substance use. Parent-child conversations that combine parental rules on drugs, information on consequences of drugs and non-judgmental feedback from parents have been found to lower substance use among adolescents.[3] But perhaps it is not just having a conversation between parent and child on dangers of substance use. Both the content and quality of parent-child conversation are crucial when it comes to matters of preventing and protecting children against dangers of drug use. According to a study by Kafka and London,[4] greater frequency and quality of parent-child communication were found to be negatively linked to substance use among adolescents.[4] Three years later, another study by Cohen et al.[5] indicated that the amount of time parents spend with their children as well as the frequency of parent-child conversations were both associated with reduced risks for tobacco onset and alcohol use in the past month.[5]

On the other hand, negative/judgmental interactions between parents and children might increase the risks of adolescent substance use by minimizing the chances of bonding between parents and children.[3] Research also suggests that generally adolescents and young people have a tendency to perceive parental criticism as a social threat and that such perception may be associated with negative emotions including defiance.[6] Consequently, children may turn to substance use including alcohol and cannabis to help them manage painful and difficult situations.[7]

Children from married-parent families are more likely to have conversations with parents on dangers of drugs than those from single-parent families as single parents tend to have less time to discuss dangers of drugs with their children.[8]

Cannabis is one of the most frequently used illicit substances among Caribbean adolescents and they tend to initiate illicit drugs, including cannabis at an early age.[9,10,11] In Barbados, prevalence of cannabis use is high among secondary school students.[9,12] Young people who use cannabis are more likely to engage in multiple substance use including alcohol and cigarette.[13,14,15,16] Additionally, young people who initiate substance use, including cannabis use, at an early age have increased risk of dependence on other drugs, and tend to continue to abuse substances through adulthood.[11,17]

A number of factors have been implicated in adolescent substance use. Significant gender differences in cannabis use were observed with males reporting greater use than females.[12,13] Students from public schools are more likely to use substances than those from private schools,[18] with students in 2^nd form more likely to use substances compared to students in 6^th form among US adolescents[19] and in Jamaica.[15,20] In the Caribbean and, specifically, in Barbados, there is, however, lack of evidence on parent-child conversations as a key factor that influences cannabis use among adolescents and youth in Barbados.

Aims

The aim of this study was to assess the association between parent-child conversations on the dangers of drug use and cannabis use among secondary school students in Barbados.

Methods and Material

Study design and settings

The study used a descriptive cross-sectional survey of a nationally representative weighted sample of 8538 secondary school students in the second, fourth, fifth, and sixth grades/forms in the selected schools in Barbados in November of 2013.

Study participants and sampling

The target population consisted of high school students in Barbados at the time of the study. This sample frame was made up of all students in forms 2, 4, 5, and 6 at 22 public schools and five private schools. The students were aged 11-17 years. This yielded a sample frame of 13,000 students. Sampling involved a two-stage random sampling procedure whereby in the first-stage secondary schools were randomly drawn from a list of all the secondary schools. In the second stage, students were randomly selected from the sampled secondary schools. Finally, a nationally representative weighted sample made up of 8538 students was composed.

Data collection tool and technique

The “Survey of Secondary School Students in Barbados Standardized Questionnaire 2013” used in this study was jointly developed by the Organization of American States/Inter-American Drug Control Commission (OAS/CICAD) and national substance abuse control bodies in the Caribbean region, including the National Council on Substance Abuse, Barbados. The National Council on Substance Abuse (NCSA) of Barbados pre-tested/validated the instrument to suit their specific context. This was done by pretesting the questionnaire among 2^nd, 4^th, 5^th, and 6^th form students in one public school. The objective of this exercise was to ensure clarity of the question items and the instructions. Pretesting the instrument also allowed NCSA research department to estimate the length of time that would be needed by the students to complete the questionnaire during the official data collection. Overall, students reported only minor issues with the instrument, and these were taken into account when the questionnaire was finalized.

The questionnaire contained information on sociodemographic and school-related characteristics and substance use, including alcohol, tobacco, cannabis, and other substances such as tranquilizers (non-medical) and stimulants.

The questionnaire was self-administered, and it took students approximately one hour to complete in classroom settings during the data collection in November 2013.

Data were entered into Microsoft Excel (Microsoft Corp.: Redmond, Washington). Researchers took steps to ensure accurate and consistent data entry using double data entry method, followed by data cleaning during which incomplete and incorrect entries and typographical errors were identified and rectified.

Explanatory variable and covariates

The main explanatory variable was parent-child conversations on dangers of drug use. Information on this was elicited with the question: “Have you had any serious conversations with any of your parents/guardians about the dangers of drug use?” The response options were 1 = Yes, 2 = No (reference was No). The covariates were age, gender, grade/form, type of school, and parental marital status. Data on age (at last birthday) were collected on a continuous scale as students were required to state how old they were at their last birthday. This was transformed into a categorical variable, as follows: 1 = 11- 14 years, 2 = 15 – 16 years and 3 = 17+ years (reference option was 17+ years). Gender was coded as 1 = male and 2 = female (reference was female). Grade at school (the form/class) was coded as 1 = 2^nd grade, 2 = 4^th grade, 3 = 5^th grade, and 4 = 6^th grade (reference was 6^th grade), while type of school had two categories: 1 = public and 2 = private (reference was private school). Parental marital status had seven categories, which were coded as 1 = single (never married), 2 = married, 3 = divorced, 4 = separated, 5 = widowed, 6 = common law/living together, and 7 = other (reference was other).

Dependent variables

The main response variable was cannabis use, which was further divided into two, namely past year and past month cannabis use. To assess past year cannabis use, students were asked whether they had smoked cannabis at least once over the past 12 months. The response options were: 1 = Yes, 2 = No (reference was No). Past month cannabis use was assessed by asking the students to indicate whether they had smoked cannabis at least once over the past 30 days. The response options were 1 = Yes, 2 = No (reference was No).

Ethical considerations

The national survey was granted approval by the Ministry of Education, Barbados. At the school level, individual school principals gave approval for the study to be conducted in their schools. Parental consent was sought and obtained from the parents of the students while the students themselves also gave their consent. The students were duly informed about the objectives and nature of the study, their right to participate, to refuse participation or withdraw at any stage without any consequences. They were also told participation was voluntary. They were instructed to keep their responses anonymous. All information collected from the students was kept confidential and secure and used only for the purposes of the research.

Statistical analysis

Descriptive analyses were done for sociodemographic characteristics and parent-child conversations. In the bivariate analysis, Chi-square test was used to compute sub-group differences in cannabis use. Significance level was set at P ≤ 0.05. Logistic regression modelling (with odds ratios and 95% confidence intervals) was used to assess for association of parent-child conversations on dangers of drug use and past year and past month cannabis use. All descriptive and inferential statistics were performed using Statistical Package for the Social Sciences (SPSS) version 20.0 (IBM Corporation, Armonk, New York).

Results

Out of the weighted sample of 8,538 students, 3,292 students (38.6%) were aged 15-16 years. Majority of the students were females, 4801 (56.2%), while 3724 students (43.6%) stated they had a conversation with their parents on the dangers of drug use [Table 1].

Table 1.

Socio-demographic characteristics of secondary school students in Barbados, 2013 (weighted n=8538)

Variable		Frequency		Percentage
Parent-child conversations about dangers of Drugs
Yes		3724		43.6
No		4429		51.9
Age:
11-14 years		3288		38.5
15-16 years		3292		38.6
17+years		1455		17
Gender:
Male		3413		40
Female		4801		56.2
Grade:
2nd		2005		23.5
4th		2517		29.5
5th		2548		29.8
6th		1468		17.2
Type of School:
Public		8492		99.5
Private		45		0.5
Parental marital status:
Single		2559		30
Married		2657		31
Divorced		544		6.4
Separated		1290		15.1
Widow		121		1.4
Common-law		598		7
Other		330		3.9

n, number, %, percentage; Note, some data may be missing, and percentages may not add up to 100 due to rounding

Bivariate analysis indicated significant associations of past year cannabis use with parent-child conversations on dangers of drugs (P < 0.05), age (P < 0.01), gender (P = 0.05), grade (P < 0.01), and marital status. Type of school had no significant association with cannabis use over the past year. There were also significant associations of past month cannabis use with age (P < 0.01), gender (P < 0.01), grade (P < 0.01), and marital status (P < 0.01). However, there was no significant association of past month cannabis use with parent-child conversations on dangers of drugs [Table 2].

Table 2.

Association between parent-child conversations on dangers of drugs, other factors, and cannabis use in Barbados, 2013 (weighted n=8,538)

Variable		Past Year (n)		X2 (P)		Past Month (n)		X2 (P)
Parent-child conversation on dangers of drugs
Yes		563 (40.4)		6.502 (0.011)		401 (30.30)		7.20 (0.007)
No		801 (44.9)				526 (35.1)
Age:
11-14 years		237 (21.8)		316 (0.001)		182 (19.8)		148.5 (0.001)
15-16 years		733 (54.2)				513 (40.7)
17+ years		371 (55.3)				237 (40.9)
Gender
Male		636 (44.8)		3.82 (0.050)		497 (39.4)		40.85 (0.001)
Female		727 (41.3)				440 (28.0)
Grade
2nd		62 (13.5)		270.9 (0.001)		58 (10.5)		162.8 (0.001)
4th		405 (45.7)				280 (37.8)
5th		591 (50.6)				451 (40.4)
6th		287 (53.9)				149 (34.2)
Type of school
Public		1355 (42.7)		0.000 (0.99)		931 (32.9)		0.28 (0.59)
Private		9 (42.9)				7 (38.9)
Parental marital status:
Single		427 (39.7)		63.28 (0.001)		325 (33.8)		77.63 (0.001)
Married		382 (42.4)				230 (31.9)
Divorced		152 (63.6)				106 (46.5)
Separated		220 (40)				157 (29.4)
Widow		21 (77.8)				21 (77.8)
Common-law		90 (40.7)				39 (19.0)
Other		66 (47.5)				60 (48.4)

n, number; %, Percentage; X², Pearson’s Chi-square; Note Data may be missing, and some percentages may not add up to 100 due to rounding

Table 3 shows the multivariate logistic regression analysis of the association between parent-child conversations on dangers of drug use and the students’ cannabis use. Students who had conversations with their parents on dangers of drug use were ironically 1.20 times as likely to report cannabis use in the past year (unadjusted model) compared to those who said they did not discuss dangers of drug use with their parents. This association remained positive and significant in the adjusted logit model (AOR = 1.35, 95% CI = 1.15 – 1.58).

Table 3.

Associations between parent-child conversation on dangers of drugs, other factors, and cannabis use in Barbados, 2013 (weighted n=8538)

Variable		Past Year				Past Month
		COR (95% CI)		AOR (95%CI)		COR (95% CI)		AOR (95% CI)
Parent-child conversation on dangers of drugs		1.20 (1.04-1.38)		1.35 (1.15-1.58)		1.24 (1.05-1.45)		1.18 (0.99-1.40)
Yes
No		1		1		1		1
Age:
11-14 years		1.48 (0.89-2.43)		1.33 (0.72-2.46)		0.31 (0.14-0.70)		0.27 (0.11-0.64)
15-16 years		0.34 (0.21-0.56)		0.32 (0.18-0.57)		0.11 (0.05-0.25)		0.15 (0.06-0.38)
17+ years		1		1		1		1
Gender
Male		0.86 (0.75-1.00)		0.81 (0.69-0.95)		0.60 (0.51-0.76)		0.56 (0.47-0.67)
Female		1		1		1		1
Grade
2nd		7.53 (5.64-10.05)		1.39 (0.89-2.77)		4.45 (3.17-6.23)		1.92 (1.16-3.19)
4th		1.39 (1.12-1.73)		0.50 (0.36-0.71)		0.85 (0.66-1.09)		0.50 (0.34-0.73)
5th		1.15 (0.93-1.41		0.83 (0.63-1.10)		0.76 (0.60-0.96)		0.63 (0.46-0.86)
6th		1		1		1		1
Type of school
Public		1.09 (0.46-2.58)		0.68 (0.26-1.80)		1.29 (0.50-3.31)		1.06 (0.37-3.00)
Private		1		1		1		1
Parental marital status:
Single		1.37 (0.96-1.95)		1.66 (1.13-2.44)		1.89 (1.28-2.73)		2.14 (1.44-3.19)
Married		1.22 (0.85-1.75)		1.26 (0.85-1.86)		1.99 (1.35-2.93)		1.98 (1.31-2.47)
Divorced		0.51 (0.33-0.78)		0.45 (0.28-0.72)		1.07 (0.69-1.66)		1.13 (0.71-1.80)
Separated		1.35 (0.93-1.97)		1.45 (0.97-2.18)		2.24 (1.50-3.35)		2.32 (1.52-3.53)
Widow		0.26 (0.10-0.68)		0.32 (0.11-0.86)		0.27 (0.10-0.71)		0.33 (0.12-0.90)
Common-law		1.31 (0.85-2.00)		1.12 (0.70-1.80)		3.96 (2.41-6.50)		3.31 (1.95-5.62)
Other		1		1		1		1

COR, crude odds ratio, AOR, adjusted odds ratio, CI, confidence interval, 1, reference

Although the association between having parent-student conversations on dangers of drug use and past month cannabis use was significant in the unadjusted logit model, when the covariates were controlled for (adjusted logit model), the association became statistically non-significant (COR: 1.24, 95%CI: 1.05-1.45; AOR: 1.18, 95%CI: 0.99-1.40).

Regarding some of the covariates, in the adjusted logit model, being 15–16 years of age, being a male, enrolled in 4^th form and belonging to divorced or widowed family setting were all significantly associated with reduced odds of past year cannabis use (AOR = 0.32, 95%CI = 0.18– 0.57; AOR = 0.81, 95%CI = 0.69 – 0.95; AOR = 0.50, 95%CI = 0.36 – 0.71; AOR = 0.45, 95%CI = 0.28– 0.72, and AOR = 0.32, 95%CI = 0.11 – 0.86), respectively. However, students belonging to single-parent family structure were 1.66 times more likely to report cannabis use in the past year compared to their mates from other family structures.

Students who were aged 11-14 years of age, 15-16 years of age, males, attending 4^th or 5^th form and belonging to widowed family structure were 73%, 85%, 44%, 50%, 37%, and 67% less likely to use cannabis in the past month compared to their counterparts (reference groups). On the other hand, students belonging to single parent, married, and separated family structures all had increased odds of cannabis use in the past month (AOR = 2.14, 95%CI = 1.44 – 3.19; AOR = 1.98, 95%CI = 1.31 – 2.47; AOR = 2.32, 95%CI = 1.52 – 3.53).

Discussion

In this study, students who had conversations with their parents on the dangers of drug use were more likely to report past year cannabis use compared to students who opined that they did not have conversations with parents on dangers of drug use in both the unadjusted and adjusted regression models. However, although a significant and positive association was found between parent-child conversations on dangers of drug use and past month cannabis use in the unadjusted logit model, the association weakened and became statistically non-significant in the adjusted logit model. The reason for this finding is not clear. Perhaps the type of conversations/interactions between parent and child may be instrumental in influencing the choices and substance use behaviour of adolescents, including cannabis use. Adolescents are at a critical developmental stage of learning to be assertive and as such they are usually very sensitive to critical and negative parental rules and regulation including rules on substance use. Adolescents/young people may therefore develop a tendency to resist perceived parental criticism and judgement.[6] This, in turn, may lead to negative emotions and possibly generate defiance as well as contributing to their making choices including risky choices/behaviour to further defy perceived negative parental criticism. These negative emotions and defiance may in turn increase their risk of substance use. This finding is consistent with Chaplin et al.[3] in 2014 who found that negative/judgmental interactions between parents and children were associated with a stressful emotional exchange which tended to increase adolescent illicit substance use risks by minimizing the chances of bonding between parents and children.[3]

Another plausible explanation for the finding of this current study is that adolescents and young people are usually not worried about the future and, may not be in a position to clearly foresee the future outcomes of a scenario the same way their parents would.[21] Thus, if parent-child conversations on dangers of drug use are centred solely on the consequences of substance use, such conversations may not be very effective in achieving a behavioural change in adolescents.

Moreover, simply having a conversation with a child on the dangers of drug use may not be sufficient to stop a child from engaging in risky behaviour. Both the content and quality of parent-child conversation are imperative protective factors against adolescent substance use and in substance use prevention.[4,5]

Being 15-16 years of age was associated with a reduced risk of past year (chronic) cannabis use in this study in both (unadjusted and adjusted models). A significant inverse relationship between students’ age (being 15-16 years) and past month (current) cannabis use was also observed. This finding is similar to the findings of National Institute on Drug Abuse,[22] which suggest that adolescents (aged 12 years or older) had low prevalence rates for cannabis use for (past year and past month) compared to older students (18-22 years).[22] This may be explained by the notion that students in this age group are still under the influence of their guardians/parents. Such influence may provide a buffer for students against drug use, including cannabis use.[20] The notion that students’ being of an age that is still under parental guidance may be a protective factor against substance use is also consistent with earlier studies which indicated parental emotional support and parental supervision (including forms of parental relationship such as communication and support) for adolescents were associated with a lower probability of substance use.[23,24]

Being in 4^th or 5^th form was associated with higher odds of past year cannabis use in the unadjusted model. But after controlling for covariates, the association became significant and negative for 4^th formers. Researchers such as Branstetter et al.[23] (2011) and Becker and Hu[25] (2008) have pointed to the effectiveness of controlling for other factors and its potential in modifying strength of statistical significance. It is interesting that after controlling for covariates the direction and strength of relationship changed from positive to negative. However, being in 4^th and 5^th form was associated with reduced risk of cannabis use in the past month in both unadjusted and adjusted models. This finding may also be explained by the fact that 4^th and 5^th forms are senior classes in the Barbadian secondary education system and as such students in these forms are expected to be focusing on serious study and preparation for their final examinations. It is possible, therefore, that the desire to perform well in their final examinations and the associated demand for serious study may serve as a protective factor against risky behaviour by taking their mind and time away from such behaviours.

Being from single-parent family structure was associated with increased odds of cannabis use in both past year and past month. Perhaps this finding may be explained by the notion that single parenthood is associated with a number of challenges which may include less time at home as single parents may be juggling different jobs in order to provide for their families. As such, single parents may lack not only the time for having supportive and quality conversations on the dangers of drug use but also emotional support for their adolescent children/wards. The finding of this study agrees with the findings of Mak, et al.[8] (2010) who opined that children from married-parent families are more likely to have conversations with parents on dangers of drugs than those from single-parent families as single parents tend to have less time to discuss dangers of drugs with their children.[8]

Limitations and recommendations

One of the limitations of this study is that it did not allow researchers to respond to specific questions regarding parent-child conversations such as the issue that students who had conversations with parents on dangers of drug use were more likely to use cannabis. The researchers recommend that substance use prevention programmes should be geared towards teaching parents to communicate with their wards in a way that is regular, consistent, unambiguous, and non-judgemental while navigating the subject of substance use.

Conclusions

Parent-child conversations on the dangers of drug use were significantly associated with drug use, but from a counter-intuitive perspective in which students who had conversations with their parents on dangers of drug use were more likely to use cannabis. This finding suggests therefore that it is not just conversations on the dangers of drug use but the type, content, and quality of conversations that matter in reducing students’ inclination to use cannabis. It is possible that negative criticisms and adversarial conversations between parents and children may have an unintended effect of increasing children’s likelihood to use cannabis. Therefore, substance use prevention/control programmes should focus on teaching/encouraging parents to use open, frequent, empathetic, non-critical, and non-judgmental communication while discussing substance use with their children/wards.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.