Risky behaviours of Jamaican adolescents with sickle cell disease

Abstract

Objectives

To describe the risky behaviours of Jamaican teens with sickle cell disease (SCD) and compare them to a national sample of Jamaican youth.

Methods

One hundred twenty two SCD adolescents, 15–19 years old, completed the standardized questionnaire used in the Jamaican Youth Risk and Resiliency Behaviour Survey (JYRRBS), which was a nationally representative survey of 1317 Jamaican youths. Information was obtained on socio-demographics, smoking, alcohol use, and sexual activity. Secondary data from the JYRRBS were extracted to measure the difference in risky behaviours between the groups.

Results

Almost 50% of SCD and 58% of national teens reported having had sexual intercourse. More SCD teens used alcohol (77.7% vs. 60.7%; P value = 0.001). Risky behaviours tended to coexist and living with a parent (odds ratio: 0.62, P value <0.01) and currently attending school (odds ratio: 0.43, P value <0.001) lowered the likelihood of having had sex.

Discussion

SCD teens engage in many risky behaviours and health care professionals should screen and counsel them at each visit.

Keywords:

• Adolescents
• Chronic illness
• Jamaica
• Risk behaviours
• Sickle cell disease

Introduction

The adolescent period has many challenges¹ and is the time most often that persons develop risky behaviours such as smoking and alcohol use that can affect their future health.^2,3 The presence of a chronic illness further magnifies the many physical, social, and psychological challenges of puberty.⁴

Sickle cell disease (SCD) is a common hereditary disorder and its effects during adolescence may be numerous. Physical growth and sexual maturation may be delayed.^5–7 School absenteeism is a significant problem⁸ and this may magnify their lower levels of social interactions.⁹ The many challenges SCD teens face may lead them to engage in risky behaviours in an effort to improve peer acceptance and foster a feeling of independence.^10,11 Even though teens with chronic illnesses have frequent contact with health care providers, common behavioural issues are only infrequently discussed.¹²

SCD affects 1 in 150 live births in Jamaica where it is the commonest genetic disorder. This study aims to improve our understanding of risky behaviours in Jamaican adolescents with SCD. The specific objectives of the study are to determine:

1.	the prevalence of risky sexual behaviours including early sexual initiation, multiple sex partners, and infrequent use of contraceptives in SCD teens and compare it to a national sample of teens
2.	the prevalence and age of initiation of alcohol and tobacco use in SCD teens and compare it to a national sample of teens
3.	what demographic (age, gender) and social (attending school, living with a parent, number of durable goods available) factors, if any, may be associated with risky behaviours.

Materials and methods

Study participants

National sample

The community-based Jamaica Youth Risk and Resiliency Behaviour Survey (JYRRBS) was conducted in 2006,¹³ where information from a nationally representative sample of 1317 Jamaican youth (598 males and 719 females; 15–19 years old) was gathered (see Table 1). Participants were enrolled via a multi-stage sampling design using Jamaican enumeration districts as the primary sampling units. The overall refusal rate was 0.2%. The objective of the survey was to derive reliable estimates on adolescent health and risk behaviours using a standardized interviewer administered questionnaire to collect data on reproductive health and substance use and abuse. Data quality was assured by training and certification of interviewers and by field supervisors performing quality checks by retesting 10% of the sample.

Table 1. Socio-demographic characteristics of SCD adolescents compared with national sample

SCD adolescents

A sample of 122 (male: 50, female: 72; SS 94, SC 24, others 4) consecutive adolescent patients (15–19 years of age) with SCD (see Table 1) presenting for routine, health maintenance care during the period between July and December 2008, were recruited. Informed consent was obtained from those ≥16 years, or informed consent of parents with assent from child if <16 years. Those teens who were ill at the time of their visit or where obtaining an informed consent/assent was not possible were not recruited to the study. No eligible participants declined to take part in the study. There was no monetary or other compensation for participants.

Measures and procedures

The study was granted ethical approval by UWI/UHWI Ethics Committee.

Enrolled persons completed, in a private area at the Sickle Cell Unit in Kingston and with a caretaker if the participant was <16 years of age, an interviewer–administered questionnaire which had been previously validated in the national sample and pretested in the SCD teens. Socio-demographic information included an indicator of socio-economic status used routinely in the Jamaica Survey of Living Conditions called an inventory of durable goods (such as stove, refrigerator, television, and car).¹⁴ Those with more ‘durable goods’ are proposed to have a higher socio-economic status. ‘Durable goods’ are used as a proxy because previous surveys found that asking people to report income only was not reliable.

The questionnaire included modules on smoking and alcohol use, as well as sexual activity. The instruments in this cross-sectional study were single-item questions used to estimate risky behaviour such as ‘Have you ever smoked a cigarette even just a puff?’, ‘Have you ever had sexual intercourse?’, and ‘In your whole life, how many people have you had sexual intercourse with?’ While we did not conduct single item reliability and validity, the overall instrument was assured validity through pretesting among a similar age group of youth as well as a panel of adolescent health expert. Reliability was attained through interviewer training and certification as well as quality checks on 10% of the sample by a field supervisor.

Statistical data analysis

All data were entered in Epidata and analyzed using Stata™ 10.0 (StataCorp, College Station, TX, USA) software.

Group-specific estimates of means, medians, and proportions along with estimates of variability – standard deviation and inter-quartile range – were provided for the SCD and the national teenagers. Mean or median values were provided for continuous variables and analysis also provided prevalence estimates for the risky behaviours under study (sexual activity, cigarette smoking, alcohol use, and marijuana use) as well as socio-demographic (age, gender, number of ‘durable goods’ as a proxy for socio-economic status, living with a parent, attending school) variables. The means, medians, and proportions for the two groups were compared, respectively, using the two-sample t test, the Wilcoxon rank-sum test, and the χ² test. Logistic regression analysis was used to determine whether the SCD teenagers were more or less likely, compared to the national teens, to practice certain risky behaviours. The odds ratios for these comparisons were adjusted for socio-demographic variables and other risky behaviours. P value of <0.05 was considered statistically significant.

Results

Sample characteristics

There were no significant between-group differences in the gender distribution, those currently in school, employment status, and whether they lived with a parent (Table 1). The SCD teens were older by 0.4 years on average (17.0 ± 1.4 vs. 16.6 ± 1.3 years; P value < 0.001) and were at higher level of schooling (P value = 0.001). A large percentage (69.7% SCD; 65.4% national) of teens were students and 9.6% SCD and 8.0% national teens were employed.

Most of the teens (82.0% SCD; 79.7% national) lived with one or both parents; and the SCD teens had a greater number of durable goods than the national teens (P value = 0.04). 99.2% Of SCD teens and 91.9% national teens were never married, 8.1% of national teens were in a common law or visiting relationship while only 0.8% of SCD teens were in a current relationship.

Sexual activity

Almost half (49.2%) of SCD and 58.1% of national teens reported having had sexual intercourse (Table 2). Significantly more males than females had engaged in sexual activity in each group (SCD teens: 62.0% males vs. 40.3% females, P value = 0.02; national teens: 68.3% males vs. 49.6% females, P value < 0.001). The SCD teens were significantly older at age of coitarche; however, further analysis showed this difference to exist only with the girls (mean age at coitarche: SCD girls: 16.3 ± 1.1 years, national girls: 15.3 ± 1.6 years; P value = 0.001). A significantly higher proportion of national teens were currently in a sexual relationship as compared to the SCD teens (8.3% vs. 0.8%; P value = 0.003). 10.2% SCD teens and 12.9% of national teens were forced for their last sex act; however, it was a significantly higher percentage of female teens (21.2%) than males (5.3%) that were forced. The national males had significantly higher number of sex partners than SCD males (5.3 ± 3.4 vs. 3.9 ± 2.7; P value = 0.03), and males in general had significantly more partners than females for each group. Very few teens (2.0% SCD; 4.6% national) had ever experienced a sexually transmitted disease, and there were no gender or group differences detected. A significantly larger proportion of national girls reported having had a pregnancy than SCD girls (25.7% vs. 1.6%; P value < 0.001); even though significantly fewer SCD teens used any contraceptive at their last sexual encounter than national teens (62.2% vs. 88.9%; P value < 0.001).

Table 2. Risky behaviours: sexual activity

SCD teens had better access to most sources of sex information, and largest sources of information were school/class, mother, friends, and books for both groups.

Substance abuse

Alcohol was the most commonly reported substance of abuse. Almost three-quarter (77.7%) of SCD teens (Table 3) and almost two-thirds (60.4%) of national teens had consumed alcohol in the past. A significantly greater proportion of both male (88.0% vs. 66.6%, P value = 0.002) and female (70.4% vs. 55.4%, P value = 0.015) SCD teens had indulged in alcohol use than their national counterparts. However, the SCD teens started drinking alcohol at an older age.

Table 3. Risky behaviours: smoking, alcohol, and other drug use

There was no difference in smoking cigarettes between the two groups: 28.7% of SCD teens had smoked, and 58.8% of them started by age 15 years. 8.6% had smoked in the last 30 days, whereas 22.5% of national teens reported smoking in last 30 days (P value = 0.055).

Marijuana smoking was not different either by group or by gender: 17.4% of teens with SCD and 18.2% of national teens had smoked marijuana. Most teens smoked because they wanted to try it out or because their friends were smoking. However, a significantly larger proportion of SCD teens stated that they smoked because they had heard that marijuana ‘makes you feel good’ (33.3% SCD teens vs. 13.8% national teens, P value = 0.027).

There was no cocaine/crack use reported by either group of teens.

Predictors of risky behaviours

The SCD teens had lower odds (odds ratio: 0.43; P value < 0.001) than the national teens of having had sexual intercourse (Table 4). Those who used alcohol (odds ratio: 2.42; P value < 0.001) and had smoked marijuana (odds ratio: 3.73; P value <0.001) were more likely to engage in sex, whereas those currently in school (odds ratio: 0.43; P value < 0.001), female teens (odds ratio: 0.46; P value < 0.001), those living with parents (odds ratio: 0.62; P value <0.005), and those with greater number of ‘durable goods’ (odds ratio: 0.66; P value < 0.01), were all less likely to have engaged in sex.

Table 4. Predictors of adolescent risky behaviours using Logistic regressions

Older teens (Odds ratio: 1.21, P value = 0.005), those who had used alcohol (odds ratio: 2.48, P value < 0.001) and those who had smoked marijuana (odds ratio: 7.96, P value < 0.001) had higher odds of having smoked cigarettes.

Teenagers with SCD were more likely (odds ratio: 2.72, P value < 0.001) to have consumed alcohol compared to national teens. In addition, individuals who smoked cigarettes or marijuana, and had initiated sexual activity were almost 2.5 times more likely to have drunk alcohol.

Discussion

This study is important as it takes a snapshot look at risky behaviours in Jamaican teens with SCD, and compares prevalence and predictors of these behaviours with a national sample of teens. Health care professionals may assume that risk taking behaviours are less in adolescents with a chronic illness; however, many studies have actually documented the opposite.^15–17 Focus during health visits is usually limited to disease management¹⁸ and routine preventative tasks may be ignored.

Our study shows lower prevalence of most risky behaviours in SCD teens; however, the results of these behaviours may have more serious implications in them than in non-SCD adolescents. Almost half the SCD teens reported having engaged in sexual activity, similar to findings from another study which reported that 51.3% of teens had engaged in sexual activity.¹¹ Contraceptive use by SCD teens was poor, and it may be due to their misconception^19,20 that they have low fertility. Even though they reported fewer pregnancies than the national sample, the SCD teens need to be counseled about the greater risks of both perinatal and maternal morbidity and mortality^21–26 associated with pregnancy in the SCD mother. They also need to understand their risk of having an offspring with SCD. Although SCD teens reported greater access to sex information, we did not evaluate whether they obtained this information from their health care providers. Previous studies show that providers may not address this topic in chronic illnesses despite its importance in this age group.^15,27

Cigarette smoking may have various ill effects in persons with SCD, and has been shown to have some link to development of acute chest syndrome,^28–30 suggesting it may play a role in affecting the pathophysiology of SCD. The frequency of smoking is comparable to the only other report¹¹ of smoking in SCD teens with almost 30% having had a cigarette and 6.5% currently smoking. Even though our SCD teens appeared to use marijuana to ‘feel good’, in the other study in Jamaican young adults with SCD,³¹ marijuana use appeared to be unrelated to disease severity or presence of pains.

Our study reports high levels of alcohol use as compared to Britto et al. who reported a much lower prevalence (36.9%) among SCD teens,¹¹ and Levenson et al. who reported 31% of SCD adults to be abusing alcohol.³² Whereas this study is not able to determine what might be causing this high prevalence of alcohol use in the SCD teens, Levenson determined no difference in disease severity or emotional problems in those who used alcohol more than in those who did not.

The co-existence of multiple risky behaviours is similar to other studies^3,16,33,34 in adolescents, and within the context of presence of a chronic illness, the potential for severe complications simply rises with presence of multiple risky behaviours.

It is also well known that smoking and alcohol abuse in adulthood tends to originate during the adolescent period;³⁵ however, the higher risk in those with a chronic illness needs to be recognized and highlighted to the teen. Teens who engage in risky behaviours also tend to be poorer on disease management and treatment compliance.³⁶

It is important to remember that those with higher engagement in risky behaviours may suffer from higher levels of depression and anxiety.^13,37 As persons, including teens, with SCD are more likely to have mental health issues such as depression and anxiety,^38,39 the presence of risky behaviours may worsen this aspect of their psychological functioning.

Various social factors, such as school attendance and living with a parent, have been implicated in the presence or absence of these risky behaviours in this study. Past studies have reported that African-American, as well as Caribbean, teens with greater social support and lower delinquency are less likely to engage in risky sexual behaviours.^40–42 In a population-based reproductive health survey of Jamaican adolescents,⁴³ school enrolment was protective against females being sexually active and males having multiple partners. Family functioning, school and peer support have also shown association with adaptation to chronic childhood illness.^44–46

This study has highlighted that teens with SCD will engage in many risky behaviours, and even though this occurs less frequently in them as compared to national teens, health care professionals need to be aware of and screen for these behaviours at routine visits. Therefore, they should review⁴² with SCD teens the negative consequences of risky behaviors as well as educate patients and their families on the positive effects of parental support and continued school attendance. In Jamaica, with the comprehensive model of care that is employed at the Sickle Cell Unit,⁴⁷ and which includes a multi-disciplinary team of primary care physicians, counsellors, and nurses providing care to their clients, the setting is ideal to allow discussions with teens about risky behaviours.

Acknowledgments

We would like to acknowledge the assistance of Prof. Rainford Wilks, who provided the overall oversight of the JYRRBS, including receiving funding from the Ministry of Health, Jamaica and the U.S. Agency for International Development (USAID) for that particular survey.