Background/Significance

According to research, cigarette smoking is detrimental to the human body being linked to causing multiple chronic conditions including heart disease, bronchitis, and chronic obstructive pulmonary disease (USDHHS, 2014).  Ninety percent of smokers have begun smoking before the age of 19 with an estimate of more than 400,000 premature deaths annually (Centers for Disease Control and Prevention, 1994).  Since children have limited awareness of the negative effects of tobacco use, preventing tobacco use is vital to the health of the country and citizens.

For the years 1998 and 2001, 60% of smokers in Thailand reported they begun smoking between the ages of 15 and 19 (National Statistics Office of Thailand, 2004). Starting unhealthy habits early in childhood can carry over into adulthood increasing the risk for chronic conditions.  Therefore, it is important to implement programs that prevent tobacco and drug use in childhood. In the United States, a school-based intervention program which focused on life skills training (LST) showed to be an effective program in decreasing drug abuse. LST programs aimed to discourage children from initiating cigarette smoking and drug use.

Purpose of the Study/Research Question/Hypothesis

While LST programs have shown to be effective in the United States by reducing the number of students who reported cigarette smoking, intervention programs had not yet been researched in Thai students. Since Thai students had similar traits to American students regarding the age students initiated tobacco use, researchers wanted to test a school-based LST program in Thai students. This study aimed to determine if implementing a program that utilized LST strategies would reduce drug and tobacco use in Thai high school aged students grades 7-12. (Seal, 2006).  The research question is: will a school-based intervention LST program decrease the incidence of tobacco and drug use among Thai high school students grades 7-12?  A hypothesis was not directly stated, but there was one-tailed hypothesis that there would be a statistical difference between the students exposed to the normal program to the students who received the LST program.

Methods

The LST program was conducted in two different randomly selected Bangkok High Schools using 85 students in grades 7-12 (Seal, 2006). To evaluate the effectiveness of the LST program, the study design was randomized and compared pretest and posttest.  The tests consisted of students completing questionnaires before training and 6 months after the LST training. “Data were collected on knowledge about the health consequences of tobacco and drug use, attitudes toward tobacco and drug use prevention, and the frequency of tobacco and drug use in the past 2 months.  Also collected were data on life skills, and refusal, decision-making, and problem-solving skills” (Seal, 2006). Trained research assistants and researchers collected the data from two groups: control and intervention group. Students in the control group received the normal training, while the students in the intervention group received the LST training for 10 class periods. The program provided skills related to tobacco and drug use including: self-awareness skills, decision-making/problem-solving skills, stress/coping skills, and refusal skills.  The skills included in the training included demonstration, instruction, feedback, presentation, role-playing, and games.

The Statistical Package for Social Sciences was used to complete analysis of the data. Analysis of variance (ANOVA) and Chi squaredstatistics were used to test any difference in the baseline variables.  T-test were also used to account for any effects from the intervention using a statistical significance level of P<0.05.

Measures

Table 1:

Variables with operationalization and level of measurement

Variable	Operationalization	Level of measurement
Demographic Data Age Gender Grade Knowledge on health consequences Attitudes toward tobacco and drug use prevention Life Skills Refusal Decision-making Frequency of tobacco and drug use	Years Male or Female Grade Level Yes/No questions Likert 4-point scale Scenario Questions Assessment Scenario Questions Assessment None, once per week, more than once per week	Ratio Nominal Ratio Nominal Interval Interval Ordinal

Findings

Using ANOVA and Chi squared statistics, results showed that there were no significant differences between the control and intervention groups during pretest (Seal, 2006).  All of the p-values for the pretest were greater than 0.05 indicating no significant difference between groups.  Furthermore, the study found that for the posttest results displayed a positive impact on the intervention group.  The null hypothesis was rejected due to a p-value less than 0.01. Also, life skills were improved in the intervention group versus the control group with a p < 0.01, leading to a rejection of the null hypothesis. In conclusion, the results for all 3 variables, knowledge, attitudes, and life skills for tobacco and drug use prevention in the intervention group had a statistically significant positive effect.

Critical Appraisal

There is uncertainty whether the sampling methods fully represented the population.  The two Bangkok schools were chosen at random but the article did not state whether socioeconomic demographics were considered.  Only two schools in the same city participated, while schools in other Thai cities may produce different results.  There were 85 students in each group, but the article did not state if the distribution of students from each school into each group were equitable. For example, 85 students from the same school may have been in one group such as the control group, with the students from other school in the intervention group. Also, it was not disclosed how many students participated from each school.  The data may not represent the population, if more students from one school participated than the other. Also, it is not clear if the sample size was adequate enough, a power analysis was not performed.  Without the power analysis, it would be difficult to establish whether there would be a significant difference between the control and intervention groups.

In the study, the data collection obtained through questionnaires, before the intervention and 6 months afterwards, may have been sufficient for the study, but there may be some uncertainty.  In the control group, during the posttest, there was one more student that reported using tobacco and drugs once per week than in pretest (22 vs 21). In addition, there was one less student in the control group that reported no tobacco and drug use in the posttest versus the pretest (58 vs 59).  It may not be significant if one student decided to initiate tobacco and drug use, but if more students from one group started or stopped versus the other, the LST program may not be as effective.  Also, if students are not completely honest on questionnaires, due to fear or other reasons, there may be some uncertainty in the data.

Many factors were lacking from the presentation of the data.  For example, only the mean and standard deviation was given for the variables of knowledge on health, attitudes, and life skills.  There were no further details in regard to the distribution of the data.  With a low sample size, and uncertainty of normalcy of distributions, it is not clear if the data collected was a true representation of the population.

The comparison of the pretest and posttest questionnaire scores in the control group did not show any statistical difference, while there was a statistical difference in the scores of the treatment group. However, there was no statistically significance between the control group and the intervention group.  The researchers used the correct statistical analysis (two sample t-test), but through further using a paired t-test for pretest and posttest scores may illicit more useful data, justifying the study be conducted in a larger population.

The presented data in the charts complemented the narrative. The charts are easy to understand by clearly listing samples, means, and standard deviations. The p-values were also listed clearly and each table described where the p-value was from.  A discrepancy noticed is the significance level of 0.05 was noted in the article, but a 95% confidence interval was never stated in the charts or article.  There were significant findings for many of the variables which were supported by the corresponding p-values in the tables.

The comparison of the pretest and posttest questionnaire scores in the control group did not show any statistical difference, while there was a statistical difference in the scores of the treatment group. However, there was no statistically significance between the control group and the intervention group.  The researchers used the correct statistical analysis (two sample t-test), but through further using a paired t-test for pretest and posttest scores may illicit more useful data, justifying the study be conducted in a larger population.

The methods in the article were clearly described in the data analysis section.  The subjects were all accounted for in each of the test, and the appropriate testing approach was utilized.  Since no population standard deviation was noted, a t-test was used for the questionnaire data obtained from the variables knowledge on health consequences, attitudes, and life skills. All p-values were listed for each analysis.

The conclusion the LST program was effective in preventing tobacco and drug use creates some bias in the report.  In the discussion section, it was mentioned that there was no significant statistical difference between the control group and the intervention group (Seal, 2006).  Yet, in the conclusion, it was stated, “the results support the conclusion that a LST program is effective in preventing tobacco and drug use among high school students (Seal, 2006). While the LST program may have had an effect on students in the intervention group, there was no significant statistical difference groups. Therefore, there are not concrete correlation between frequency of tobacco and drug use and the variables. Implementation of the study on a larger scale may further find correspondence between control and intervention which may decrease any bias.

References

• Centers for Disease Control and Prevention. Annual smoking – attributable mortality, years of potential life lost, and economic costs – United States, 1995–99.
  
  MMWR
  
  2002;
  
  51
  
  : 300– 303.
• Seal, N. (2006). Preventing tobacco and drug use among Thai high school students through life skills training.
  
  Nursing and Health Sciences,
  
  
  8
  
  (3), 164-168. doi: 10.1111/j.1442-2018.2006.00275.x
• US Department of Health and Human Services.
  
  The Health Consequences of Smoking. A Report
  
  
  of the Surgeon General.
  
  Washington, DC: US Department of Health and Human Services/Centers for Disease Control and 1revention/ National Center for Chronic Disease Prevention and Health Promotion/ Office on Smoking and Health, 2004.