School Drug Testing: A Critical Review of the Literature

This paper explores the question of whether school drug testing is an effective solution to tackle adolescent substance abuse problems. Research studies in major academic databases and Internet websites are reviewed. Several observations are highlighted from the review: (1) there are few research studies in this area, particularly in different Chinese contexts; (2) the quality of the existing studies was generally low; and (3) research findings supporting the effectiveness of school drug testing were mixed. Methodological issues underlying quantitative and qualitative evaluation studies of the effectiveness of school drug testing are also discussed.


INTRODUCTION
A survey of the websites of several international organizations (e.g., Office on Drugs and Crime of the United Nations, International Narcotics Control Board, National Institute of Drug Abuse in the United States, and European Monitoring Center for Drugs and Drug Addiction) shows that illicit drug use is a thorny global problem to be resolved. Probably because of the influence of the popular culture and youth subculture, substance abuse among young people has also become an acute global problem [1,2]. With reference to the findings reported in some of the major databases on adolescent development, such as Monitoring the Future, Youth Risk Behavior Surveillance (YRBS), and National Household Survey on Drug Abuse (NHSDA), adolescent substance abuse is a concern for policy makers and health professionals [3]. From the results of the 2008 National Survey on Drug Use and Health, it was found that 9.3% of youths aged 12-17 were current illicit drug users [4].
To tackle the gradual worsening of adolescent drug abuse, school drug testing has been adopted in some Western countries in order to cope with the problem. In 1995, the U.S. Supreme Court approved drug testing for student athletes in public high schools. In 2002, the U.S. Supreme Court broadened the ruling to include all students taking part in competitions against students at other schools in after-school activities [5]. Ever since its inception, there has been much debate on the necessity and value of student drug testing, particularly its effectiveness. Roche et al.[6] reviewed the theories, assumptions, and limitations of the underlying rationales for school drug testing. They also reviewed some of the studies in the field and concluded that the quality of the studies was generally low. Although the study of Roche et al.[6] is a pioneering attempt to review some of the studies in the field, there are three limitations. First, the studies under review were not exhaustive, as some of the studies reported in academic journals and the Internet were not included. Second, although the quality of the studies under review was discussed in the paper, the details (e.g., problems of the design, methodology and data analyses, biased conclusions, etc.) were not included. Third, findings that support the effectiveness of school drug testing and those that oppose it were not separately reported. Against this background, the present study attempted to review the literature on the effectiveness of school drug testing. Findings from the literature that support it and those that criticize it are individually presented. Besides, quality of the studies is evaluated in detail. Finally, methodological issues intrinsic to quantitative and qualitative studies of the effectiveness of school drug testing are also discussed.

Search Strategy
The primary aim of this systematic review was to explore the effectiveness of the contentious issue, school drug testing. Searches were undertaken within the major academic databases: PsycINFO, Social Work Abstracts, Medline, CINAHL, and Sociological Abstracts, using multiple keywords: random drug test or drug testing or school drug testing or drug detection. In addition, empirical studies reported in the websites on the Internet were also reviewed using the above terms. The studies under review in the current study are outlined in Appendix 1. Some of the authoritative websites on school drug testing are presented in Table 1.  (Table 2), there are also some that do not (Table 3). Fourth, as shown in Table 4, quality of the existing studies was generally not high; therefore, doubt is cast on their conclusions on the effectiveness of drug test. Added to this, there were few well-designed quantitative studies and wellcrafted qualitative evaluation studies in the field.

DISCUSSION
Despite the heightened public concern for school drug testing and its controversial nature, empirical studies that examine the effectiveness of drug testing in the school context are, surprisingly, limited in number. From the perspective of evidence-based practice, research studies play an indispensable role in clarifying the effectiveness of school drug testing and providing support for the policy. As Chinese people constitute roughly one-fifth of the world"s population, the absence of school drug testing research is definitely undesirable, particularly in view of the fact that mandatory drug testing is legally acceptable in mainland China. Furthermore, research on school drug testing is indispensable when voluntary school drug testing was implemented in the Tai Po district of Hong Kong on a trial basis in 2009/10 school year.
The present review shows that there are mixed findings on the effectiveness of school drug testing. It is noteworthy that while there are findings that indicate that drug testing had no positive effect, there are findings that support the effectiveness of school drug testing. This picture is also clearly shown in the study of Goldberg et al.[7], which is one of the few prospective trials in the field. As pointed out by Goldberg et al.[7], "although these findings may differ in other schools or regions of the United States, this study lends credence to some DAT deterrent effects, especially for past year use for drugs, at two time points, and for drugs and alcohol at two time points. However, because some substance abuse mediators appeared to worsen and past month substance use never changed, more research should be performed to assess the policy of drug and alcohol testing"s overall effects" (p. 428, italics added). Similarly, Knight and Levy[8], in an editorial of the Journal of Adolescent Health [8], pointed out that "although we might hope that the present study by Goldberg would help to end the national debate, this hope is unlikely to be realized on the basis of this report, which includes ample "evidence" to fuel the fire on both sides" (p. 419).
As far as the quality of the studies under review is concerned, the review shows that the quality of the existing studies was not high. In addition, it is noteworthy that the findings in the studies under review cannot give a definitive answer to the question of whether school drug testing is effective. For example, although a large sample size was used in the study of Yamaguchi et al.[9], the major limitation was its cross-sectional design because "one cannot make definitive causal interpretations regarding effects of drug testing; only a panel design in a randomized or natural experiment can do so. Perhaps schools that instituted drug testing initially had higher use, and drug testing reduced those levels to levels similar to those at other schools" (p. 164). With the aim to explore the association between student drug use and drug-testing policies in schools, Yamaguchi et al. [9] concluded that "while lack of evidence for the effectiveness of drug testing is not definitive, results suggest that drug testing in schools may not provide a panacea for reducing student drug use that some (including some on the Supreme Court) had hoped" (p. 164). However, probably because of the large sample involved, this study has commonly been taken as evidence against school drug testing.
Obviously, the sustainability of school drug testing depends principally on the amount and quality of research evidence supporting its value and effectiveness. There are two lines of evaluation research that should be done in the future. To begin with, quantitative research utilizing experimental designs should be conducted. However, there are at least six issues that should be addressed in experimental studies. First, selection bias (pregroup differences) may confound the results. Studies utilizing pre-experimental designs, such as the one conducted by Yamaguchi et al. [9], are particularly vulnerable to this threat. Second, it is noteworthy that a drug testing scheme will heighten other schools" sensitivity to drug prevention, which may minimize the treatment effect in the experimental groups. Third, it is possible that  experimental schools may step up antidrug measures in schools, which would eventually exaggerate the treatment effect of school drug testing. Fourth, political and community responses to a drug testing scheme may influence student attitudes before, during, and after the implementation process. Fifth, the choice of outcome measures and honest disclosure of substance abuse behavior will definitely affect the evaluation outcomes. Sixth, as adolescent substance abuse may have a low base rate in places like Hong Kong, it may be difficult to detect real differences between the experimental group and control group unless very large sample sizes and sensitive measures are used. Finally, researchers have to consider carefully whether "blinding" can be feasibly and meaningfully carried out in related experimental studies. The second line of research is qualitative evaluation studies. Besides those qualitative findings reported in academic journals (Tables 2 and 3), there are numerous qualitative accounts of the value and problems of school drug testing. For example, while a high school principal pointed out that "the committee worked very hard to provide a tool which would have a positive effect on our students. The extremely low number of positive tests indicates the program is worth the cost" [10, p. 1], Knight and Levy[11] warned that the view that drug testing in schools can prevent adolescent substance abuse has to be interpreted with caution because their efficacy has not yet been proven and drug tests are associated with significant technical concerns. Samples not randomly drawn  Sample errors associated with the percentage data not properly addressed Table 4 continues Shek: Student Drug Testing TheScientificWorldJOURNAL (2010) 10, 356-365  When researchers conduct qualitative evaluations of school drug testing, it is important to pay particular attention to the rigor of the studies. Shek et al.[12] pointed out that there are 12 principles that should be upheld in qualitative evaluation studies: (1) statement of the philosophical base of the study; (2) justification for the number and nature of the participants of the study; (3) detailed description of data collection procedures; (4) discussion of biases in the study; (5) description of steps taken to guard against biases or arguments that biases should and/or could not be eliminated; (6) pay attention to reliability issues; (7) considering triangulation strategies; (8) use of peer checking and member checking; (9) use of audit trails; (10) examination of alternative explanations; (11) accounting for negative evidence; and (12) examination of limitations of the study. Obviously, methodological rigor of future qualitative evaluation studies in this field can be strengthened by upholding these principles.
Adopting a balanced perspective, school drug testing schemes may not be a panacea for adolescent substance abuse. In the long run, effort should be made to integrate school drug testing with other preventive measures, such as preventive drug education and positive youth development [13,14,15,16,17], to help young people to stay away from drugs. Fundamentally, it is important to take an evidence-based approach to evaluate the strategies to tackle adolescent substance abuse, including a school drug testing scheme.