Actionable evidence: How to tackle obesity in primary care?

Scenario: A 45 year old male with a BMI of 32 presents to his primary care physician. What intervention should be suggested to him to reduce his BMI? Should only behavioral modification be adequate or should orlistat be added to it?

Evidence: A recent study titled “Effectiveness of Primary Care–Relevant Treatments forObesity in Adults: A Systematic Evidence Review for the U.S. PreventiveServices Task Force" published in the Annals of Internal Medicine looks into the evidence.

The study tried to answer 4 key questions –

1. Is there direct evidence that primary care screening programs for adult obesity or overweight improve health outcomes or result in short-term (12 to 18 mo) or sustained (>18 mo) weight loss or improved physiologic measures? a) How well is weight loss maintained after an intervention is completed?
2. Do primary care–relevant interventions (behaviorally based interventions and/or pharmacotherapy) in obese or overweight adults lead to improved health outcomes? a) What are common elements of efficacious interventions? b) Are there differences in efficacy between patient subgroups (i.e., age 65 y or older, sex, race/ethnicity, degree of obesity, baseline cardiovascular risk)?
3. Do primary care–relevant interventions in obese or overweight adults lead to short-term or sustained weight loss, with or without improved physiologic measures? a) How well is weight loss maintained after an intervention is completed? b) What are common elements of efficacious interventions? c) Are there differences in efficacy between patient subgroups (i.e., age 65 y or older, sex, race/ethnicity, degree of obesity, baseline cardiovascular risk)?
4. Do primary care–relevant interventions in obese or overweight adults lead to short-term or sustained weight loss, with or without improved physiologic measures? a) How well is weight loss maintained after an intervention is completed? b) What are common elements of efficacious interventions? c) Are there differences in efficacy between patient subgroups (i.e., age 65 y or older, sex, race/ethnicity, degree of obesity, baseline cardiovascular risk)?

Outcome: The first key question remains unanswered as the authors were unable to identify any study comparing screening vs non-screening of obese individuals.

The second, third and fourth key questions were answered. The summary –

Behavioral interventions

Parameter	Outcome
Weight Loss	Weighted mean difference in mean weight change between behavioral intervention and control group is -3.01 kg 95% CI (-4.02 to -2.01) at 12 to 18 months. Behavioral interventions lasting longer continued to show weight loss. Weight los was maintained up to a year following active intervention phase.
Mortality	No difference in death rate but limited by small number of trials.
Cardiovascular disease	No difference in CVD events or CVD related deaths in 3 large good quality trials
Hospitalization	No difference, limited by low hospitalization rate.
Health related quality of life/depression	No difference in depression, small favorable change of health related quality of life with weight loss
Incidence of diabetes mellitus	Weight loss of 4 to 7 kg reduced diabetes incidence by 50% or more over 2 to 3 years.
Glucose tolerance	Mean decrease in glucose level = 0.30 mmol/L (5.4 mg/dL)
Lipids	Weighted mean difference in LDL cholesterol level between treatment and control group = -4.94 mg/dL 95%CI (-7.32 to -2.56)
Blood pressure	Weighted mean difference in mean change in blood pressure between treatment and control group = -2.48 mmHg 95%CI (-3.25 to -1.71)
Waist circumference	-2.7 cm 95%CI( -4.1 to 1.4)
Adverse effects	None

Orlistat (Note- Almost all the trials of orlistat had simultaneous behavioral interventions)

Parameter	Outcome
Weight Loss	Weighted mean difference between behavioral intervention and control group is -2.98 kg 95% CI (-3.92 to -2.05).
Mortality	There were 4 studies which reported this outcome. Each study reported 1 death in in the orlistat group, but no clear relationship with treatment.
Cardiovascular disease	Not reported.
Hospitalization	Not reported.
Health related quality of life/depression	No difference in depression, Orlistat had greater satisfaction with treatment, less overweight distress and improvement in the vitality subscale of SF – 36.
Incidence of diabetes mellitus	Decreased incidence by 9-10% but concerns regarding generalizability and reliability.
Glucose tolerance	Mean decrease in glucose level = 0.31 mmol/L (5.5 mg/dL). Glucose reductions are greater in orlistat than placebo possibly because they were conducted in diabetic patients.
Lipids	Weighted mean difference in LDL cholesterol level between treatment and control group = -11.37 mg/dL 95%CI (-15.75 to -7.00)
Blood pressure	Weighted mean difference in mean change in blood pressure between treatment and control group = -2.04 mmHg 95%CI (-2.97 to -1.11)
Waist circumference	-2.3 cm 95%CI( -3.6 to -0.9)
Adverse effects	Gastrointestinal symptoms of mild to moderate intensity which resolves spontaneously, liver injury, vitamin E and K deficiency, increased incidence of hypoglycemia.

The authors also examined metformin which had modest effects on weight loss -2.85 kg 95%CI (-3.52 to -2.18).

Comment

The behavioral interventions included counseling regarding the importance of diet and exercise. There the interventions were not described fuly in the study. There was variability in the methods employed for behavioral modifications in the between the studies. On the other hand, orlistat studies had high attrition rate and were mostly funded by pharmaceutical companies.

Bottom line

In the primary care setting, behavioral intervention is effective and safe and should be considered as a first line therapy for weight loss. Orlistat may be added later. There are significant adverse effects associated with orlistat. Metformin may be tried in diabetics.

Leblanc ES, O'Connor E, Whitlock EP, Patnode CD, & Kapka T (2011). Effectiveness of primary care-relevant treatments for obesity in adults: a systematic evidence review for the u.s. Preventive services task force. Annals of internal medicine, 155 (7), 434-47 PMID: 21969342