Systematic reviews, based on the science of research synthesis, are a valuable and useful source of information about the effectiveness of interventions and treatments in health promotion, medical care, and other areas of public health. These reviews inform many clinicians, practitioners, and policymakers prior to making decisions concerning resource investments about what interventions are most appropriate and most likely to be effective for particular audiences or populations. For the worksite health promotion practitioner, these reviews can be one of the most useful sources of information when considering what type of program to implement at the worksite.
What Makes for a "Systematic Review"?
Basically, a systematic review is a way to find out what is known or not known in the scientific literature about what does and does not work. In addition, it may provide a sense of the strength of the effectiveness of specific interventions. This is because different types of study designs produce different levels of evidence. For example, whereas a longitudinal, randomized, controlled intervention trial (RCT) may infer that a particular treatment caused a better outcome to occur, a cross-sectional, correlational study cannot do so. In other words, there is a hierarchy of evidence based upon the relative strength of the studies considered in the review. The Centre for Evidence-Based Medicine at Oxford University has outlined a hierarchy organized from the strongest to the weakest:
* systematic review of randomized controlled trials
* randomized controlled trials
* observational studies
* case series
* expert opinion
Unfortunately, not all research fits nicely into this kind of hierarchy. Considering health promotion types of research, many research questions may never be studied using RCT designs, where study participants exposed to an intervention are compared with participants who are not exposed to that intervention. For example, to study the long-term effects of physical inactivity, subjects cannot be randomly assigned to a physically active and a sedentary group from which physical activity is withheld. Furthermore, worksite health promotion research is often not amenable to RCTs, because companies are worried both about employees receiving different interventions and about contamination of study groups when workers who interact with each other on a daily basis are randomized into different interventions.
The "How-to" of a Systematic Review
The key to a well-conducted, unbiased review is to follow an organized, explicit, and reproducible process that searches for all relevant studies on a particular topic, considers the methodologies of those studies and their results, and interprets the findings of the review. More specifically, the critical steps in this process include the following:
* formulating a clinical question
* finding relevant studies
* selecting and assessing those studies
* summarizing and synthesizing relevant study results
* interpreting the review findings
* updating the review
Systematic Reviews on Obesity
The World Health Organization has referred to obesity as a "global epidemic" because of the prevalence of this serious public health concern and the related health problems that include atherosclerosis, diabetes, heart disease, and respiratory problems. In addition, obesity impacts quality of life, functional status, and productivity. Over the course of the past several decades, the accumulated research has attempted to identify successful approaches to addressing obesity, from advice on how to lose weight using behavior change counseling, to pharmacologic and surgical interventions. However, because of the number of studies in weight management and the variety of the types of studies used, it is difficult for practitioners to sort out which approaches work best, and which approaches may be best suited for the worksite setting.
Recently, the U.S. Preventive Services Task Force conducted a systematic review of obesity interventions for adults and published their findings (1). They considered the clinical question, "How effective and safe are screening methods and treatments for obesity?" The study team searched for studies between January 1994 and February 2003, and included RCTs, systematic reviews of RCTs, cohort studies, and nonrandomized controlled trials. Pharmacotherapy studies needed to have at least 6 months of follow-up; all other studies needed at least 12 months of follow-up. Their findings indicated that counseling or behavioral interventions and pharmacotherapy could result in moderate weight loss (approximately 3 to 5 kg) over at least 6 to 12 months. However, pharmacotherapy did have adverse effects. Surgery resulted in considerable weight loss in very obese people, but there was a risk of death.
Whereas this is very useful information for practitioners to receive, the results may be of limited use to worksite health promotion professionals, because pharmacotherapy and surgery may not be directly applicable to the worksite setting. Even more meaningful than this review may be a review of obesity-related interventions that are more community oriented. Such a review has been conducted by the U.S. Centers for Disease Control and Prevention; the results will become available during the latter part of 2004. Because the results are not yet published at the time of this writing, please check the Web site of the Task Force on Community Preventive Services (http://www.thecommunityguide.org). Topics such as physical activity, nutrition, tobacco, alcohol, pregnancy, and diabetes have already been completed, and the Web site is an excellent resource for worksite health/fitness professionals.
HealthPartners' Systematic Review of Weight Management Interventions
In 2004, staff at the HealthPartners Center for Health Promotion, in collaboration with Kaiser Permanente's Care Management Institute, completed a systematic review of longer-term weight management interventions, organized according to type of intervention (2). The clinical question asked for this review was, "What lifestyle strategies and/or treatment components contribute to successful weight management (weight loss and weight maintenance) defined as 5% to 7% of starting (baseline) body weight, 5% loss maintained for 1 year or longer?"
A literature search was conducted for weight loss/maintenance intervention studies published after January 1, 1997 that met the following criteria:
* randomized clinical trials ≥12 months duration
* observational studies ≥5 years duration
* meta-analysis of studies meeting study criteria
The initial search identified 1,522 studies, from which 190 met criteria for review. Ultimately, 81 studies met the criteria and were included in the evidence tables. Studies were divided into eight types of interventions:
* diet as the primary intervention
* diet plus specific guidelines for physical activity
* exercise alone
* meal replacements
* very low-calorie diets
* antiobesity medications
* behavioral interventions above and beyond basic behavioral strategies
* bariatric surgery
The main purpose of the review was to assess weight loss and maintenance from the interventions at 3, 6, 12, 18, 24, 36, 48, and 60 months. Data from 62 studies with 22,121 subjects enrolled in the study arms and with 15,775 subjects completing the study (71%) were pooled to determine the amounts of weight lost and maintained. The figure presents the results in a manner that allows for immediate comparison of the various types of interventions (data on bariatric surgery were not included in summary figure because only four studies met inclusion criteria, and of these only one was an intervention study). It is clear that the combination of diet and exercise interventions and meal replacement programs are good choices for worksite health promotion programs. Furthermore, the results show that longer-term outcomes for lifestyle interventions include approximately 5 to 7 kg of weight loss-a result associated with significant health benefits.
Worksite health promotion practitioners can use the results of systematic reviews to consider the effectiveness of specific interventions to address obesity and to consider how well these may fit into the context of the worksite setting. Systematic reviews are a great resource to consider when designing programs and services.