How Transparent is Behavioral
Interventions Research for Pathological Gambling?
A Systematic Review
Iman Parhami, M.D., MPH, Arlene Fink,
Ph.D.,
Margarit Davtian, M.A., Michael D. Campos, Ph.D., Aaron
Siani, Richard J. Rosenthal, M.D.,
Timothy W. Fong, M.D.
UCLA Gambling Studies Program
Department of Psychiatry and Biobehavioral
Sciences
University of California, Los Angeles
(Click here for PDF of Poster)
ABSTRACT
BACKGROUND:
While a number of gambling interventions are noted in the scientific
literature, many experts in the field believe that this body of work does not
adhere to rigorous research standards, and, as a result, prevents the field from
making firm claims about the true effectiveness of such interventions. This
concern can affect healthcare policy makers, funders, providers, and ultimately
the well-being of clients. For that reason, gambling interventions should
consist of sophisticated reports with sufficient detail and transparency
(clarity) for proper evaluation. In an effort to improve reports, investigators
in various fields have conducted systematic reviews to explore the ‘quality of
reporting’ using standardized reporting guidelines.
OBJECTIVES: We
systematically retrieved and reviewed behavioral intervention studies for
pathological gambling, identified areas commonly and rarely transparent in these
reports, and determined whether adherence to transparent reporting changed over
time.
METHODS: We
utilized specific search strategies in PubMED, Web of Science, and PsychINFO to
identify behavioral treatment interventions for pathological gambling published
in peer-reviewed journals between January 1, 2000 to April 1, 2011. We created a
59-item transparent reporting checklist, with 23 particularly important items,
based on the TREND (Transparent Reporting of Evaluations with Non-Randomized
Designs) statement. The TREND statement, established in 2004 and endorsed by
numerous organizations, is an accepted guideline for transparency, or clarity,
in the reporting of behavioral and public health interventions. We calculated
proportions of studies that met each applicable item and conducted regression
models to study the relationship between the year of publication, whether an
item was satisfied, and the proportion of items met by each study.
RESULTS: Out
of 384 studies screened, 26 were eligible and reviewed for transparent
reporting. Each individual item on our transparent reporting checklist was
satisfied by 0% to 100% of the studies (median =77%, standard error = 4.4%) and
each particularly important item was satisfied by 8-100% of the studies (median
= 54%, s.e. = 6.9%). 9 out of 23 particularly important items were met by less
than half of the studies, including ‘reported activities to increase compliance’
(met by 46% of the studies), ‘discussed the success of and barriers to
implementing the intervention’ (38%), ‘described the baseline characteristics of
those lost to follow-up and those retained’ (24%), ‘compared between study
population at baseline and target population of interest’ (15%). In addition,
42% of the studies reported their involvement with an institutional ethical
review board and 65% reported their funding sources. No relationship was found
between transparent reporting (satisfying each item or aggregate number of items
met by each study) and the year of publication.
CONCLUSION:
Reports of behavioral interventions for pathological gambling do not appear to
meet guidelines for transparent reporting. Although the recognition of these
guidelines is in its infancy, this shortcoming prevents clinicians and
policy-makers from adequately interpreting, evaluating, and translating these
interventions into empirically supported interventions in the clinical setting.