Evidence selection for a prescription drug's benefit-harm assessment: challenges and recommendations

GLP-1 receptor agonists for weight reduction in people living with obesity but without diabetes: a living benefit-harm modelling study

Background

The benefit of Glucagon-like Peptide-1 (GLP-1) receptor agonists (RAs) in weight reduction against potential harms remains unclear. This study aimed at evaluating the benefit-harm balance of initiating GLP-1 RAs versus placebo for weight loss in people living with overweight and obesity but without diabetes.

Methods

We performed benefit-harm balance modelling, which will be updated as new evidence emerges. We searched for randomised controlled trials (RCTs) in PubMed, controlled trials registry, drug approval and regulatory documents, and outcome preference weights as of April 10, 2024. We synthesize data using pairwise meta-analysis to estimate the effect of GLP-1 RAs to inform the benefit-harm balance modelling. We predicted the absolute effects of the positive and negative outcomes over 1 and 2 years of treatment using exponential models. We applied preference weights to the outcomes, ranging from 0 for least concerning to 1.0 for most concerning. We then calculated whether the benefit of achieving 5% and 10% weight loss outweighed the harms on a common scale. The analyses accounted for the statistical uncertainties of treatment effects, preference weights, and outcome risks.

Findings

We included 8 RCTs involving 8847 participants. The pooled average age was 46.7 years, with the majority being women (74%) and people living with obesity (96%). Of 1000 persons treated with GLP-1 RAs for 2 years, 375 (95% confidence interval 352 to 399) achieved a 10% weight loss, and 318 (296 to 339) achieved a 5% weight loss compared to those treated with placebo. Several harm outcomes were more frequent in the GLP-1 RA group, including 41 abdominal pain events per 1000 persons over 2 years (19 to 69), cholelithiasis (8, 1 to 21), constipation (118, 78 to 164), diarrhoea (100, 42 to 173), alopecia (57, 10 to 176), hypoglycaemia (17, 1 to 68), injection site reactions (4, -3 to 19), and vomiting (110, 80 to 145) among others. Achieving a 10% weight loss with GLP-1 RA therapy outweighed the cumulative harms, with a net benefit probability of 0.97 at year 1 and 0.91 at year 2. The absolute net benefit was equivalent to 104 (100 to 112) per 1000 persons achieving a 10% weight loss over 2 years without experiencing any worrisome harm. A 5% weight loss did not show a net benefit, with probabilities of 0.13 and 0.01 at year 1 and year 2, respectively. However, these benefits were sensitive to preference weights, suggesting that even a 5% weight loss could be net beneficial for individuals with less concern about harm outcomes. The net benefit for a 10% weight loss was highest for semaglutide, followed by liraglutide and tirzepatide, with 2-year probabilities of 0.96, 0.72, and 0.60, respectively.

Interpretation

The benefit of GLP-1 RAs exceeded the harms for weight loss in the first 2 years of treatment, yet the net benefit was dependent on individual' treatment goals (10% or 5% weight loss) and willingness to accept harms in pursuit of weight loss. This implies that treatment decisions have to be personalized to individuals to optimize benefits and reduce harms and overuse of treatments. Due to varying evidence, especially regarding harm outcomes across studies, it is necessary to continuously update and monitor the benefit-harm balance of GLP-1 RAs.

Funding

SNSF and LOOP Zurich.

Global variation of risk thresholds for initiating statins for primary prevention of cardiovascular disease: a benefit-harm balance modelling study

Background

We previously showed that the 10-year cardiovascular disease (CVD) risk threshold to initiate statins for primary prevention depends on the baseline CVD risk, age, sex, and the incidence of statin-related harm outcome and competing risk for non-CVD death. As these factors appear to vary across countries, we aimed in this study to determine country-specific thresholds and provide guidelines a quantitative benefit-harm assessment method for local adaptation.

Methods

For each of the 186 countries included, we replicated the benefit-harm balance analysis using an exponential model to determine the thresholds to initiate statin use for populations aged 40 to 75 years, with no history of CVD. The analyses took data inputs from a priori studies, including statin effect estimates (network meta-analysis), patient preferences (survey), and baseline incidence of harm outcomes and competing risk for non-CVD (global burden of disease study). We estimated the risk thresholds above which the benefits of statins were more likely to outweigh the harms using a stochastic approach to account for statistical uncertainty of the input parameters.

Results

The 5^th and 95^th percentiles of the 10-year risk thresholds above which the benefits of statins outweigh the harms across 186 countries ranged between 14 and 20% in men and 19–24% in women, depending on age (i.e., 90% of the country-specific thresholds were in the ranges stated). The median risk thresholds varied from 14 to 18.5% in men and 19 to 22% in women. The between-country variability of the thresholds was slightly attenuated when further adjusted for age resulting, for example, in a 5^th and 95^th percentiles of 14–16% for ages 40–44 years and 17–21% for ages 70–74 years in men. Some countries, especially the islands of the Western Pacific Region, had higher thresholds to achieve net benefit of statins at 25–36% 10-year CVD risks.

Conclusions

This extensive benefit-harm analysis modeling shows that a single CVD risk threshold, irrespective of age, sex and country, is not appropriate to initiate statin use globally. Instead, countries need to carefully determine thresholds, considering the national or subnational contexts, to optimize benefits of statins while minimizing related harms and economic burden.

Informing Patient-Centered Care Through Stakeholder Engagement and Highly Stratified Quantitative Benefit-Harm Assessments

OBJECTIVES

In a previous project aimed at informing patient-centered care for people with multiple chronic conditions, we performed highly stratified quantitative benefit-harm assessments for 2 top priority questions. In this current work, our goal was to describe the process and approaches we developed and to qualitatively glean important elements from it that address patient-centered care.

METHODS

We engaged patients, caregivers, clinicians, and guideline developers as stakeholder representatives throughout the process of the quantitative benefit-harm assessment and investigated whether the benefit-harm balance differed based on patient preferences and characteristics (stratification). We refined strategies to select the most applicable, valid, and precise evidence.

RESULTS

Two processes were important when assessing the balance of benefits and harms of interventions: (1) engaging stakeholders and (2) stratification by patient preferences and characteristics. Engaging patients and caregivers through focus groups, preference surveys, and as co-investigators provided value in prioritizing research questions, identifying relevant clinical outcomes, and clarifying the relative importance of these outcomes. Our strategies to select evidence for stratified benefit-harm assessments considered consistency across outcomes and subgroups. By quantitatively estimating the range in the benefit-harm balance resulting from true variation in preferences, we clarified whether the benefit-harm balance is preference sensitive.

CONCLUSIONS

Our approaches for engaging patients and caregivers at all phases of the stratified quantitative benefit-harm assessments were feasible and revealed how sensitive the benefit-harm balance is to patient characteristics and individual preferences. Accordingly, this sensitivity can suggest to guideline developers when to tailor recommendations for specific patient subgroups or when to explicitly leave decision making to individual patients and their providers.

Net benefit of statins for primary prevention of cardiovascular disease in people 75 years or older: a benefit-harm balance modeling study

BACKGROUND

We determined the risk thresholds above which statin use would be more likely to provide a net benefit for people over the age of 75 years without history of cardiovascular disease (CVD).

METHODS

An exponential model was used to estimate the differences in expected benefit and harms in people treated with statins over a 10-year horizon versus not treated. The analysis was repeated 100,000 times to consider the statistical uncertainty and produce a distribution of the benefit-harm balance index from which we determined the 10-year CVD risk threshold where benefits outweighed the harms. We considered treatment estimates from trials and observational studies, baseline risks, patient preferences, and competing risks of non-CVD death, and statistical uncertainty.

RESULTS

Based on average preferences, statins were more likely to provide a net benefit at a 10-year CVD risk of 24% and 25% for men aged 75-79 years and 80-84 years, respectively, and 21% for women in both age groups. However, these thresholds varied significantly depending on differences in individual patient preferences for the statin-related outcomes, with interquartile ranges of 21-33% and 23-36% for men aged 75-79 years and 80-84 years, respectively, as well as 20-32% and 21-32% for women aged 75-79 years and 80-84 years, respectively.

CONCLUSIONS

Statins would more likely provide a net benefit for primary prevention in older people taking the average preferences if their CVD risk is well above 20%. However, the thresholds could be much higher or lower depending on preferences of individual patients, which suggests more emphasis should be placed on individual-based decision-making, instead of recommending statins for everyone based on a single or a small number of thresholds.

Balance of benefits and harms of different blood pressure targets in people with multiple chronic conditions: a quantitative benefit-harm assessment

OBJECTIVE

Recent studies suggest that a systolic blood pressure (SBP) target of 120 mm Hg is appropriate for people with hypertension, but this is debated particularly in people with multiple chronic conditions (MCC). We aimed to quantitatively determine whether benefits of a lower SBP target justify increased risks of harm in people with MCC, considering patient-valued outcomes and their relative importance.

DESIGN

Highly stratified quantitative benefit-harm assessment based on various input data identified as the most valid and applicable from a systematic review of evidence and based on weights from a patient preference survey.

SETTING

Outpatient care.

PARTICIPANTS

Hypertensive patients, grouped by age, gender, prior history of stroke, chronic heart failure, chronic kidney disease and type 2 diabetes mellitus.

INTERVENTIONS

SBP target of 120 versus 140 mm Hg for patients without history of stroke.

PRIMARY AND SECONDARY OUTCOME MEASURES

Probability that the benefits of a SBP target of 120 mm Hg outweigh the harms compared with 140 mm Hg over 5 years (primary) with thresholds >0.6 (120 mm Hg better), <0.4 (140 mm Hg better) and 0.4 to 0.6 (unclear), number of prevented clinical events (secondary), calculated with the Gail/National Cancer Institute approach.

RESULTS

Considering individual patient preferences had a substantial impact on the benefit-harm balance. With average preferences, 120 mm Hg was the better target compared with 140 mm Hg for many subgroups of patients without prior stroke, especially in patients over 75. For women below 65 with chronic kidney disease and without diabetes and prior stroke, 140 mm Hg was better. The analyses did not include mild adverse effects, and apply only to patients who tolerate antihypertensive treatment.

CONCLUSIONS

For most patients, a lower SBP target was beneficial, but this depended also on individual preferences, implying individual decision-making is important. Our modelling allows for individualised treatment targets based on patient preferences, age, gender and co-morbidities.

Defining certainty of net benefit: a GRADE concept paper

Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology is used to assess and report certainty of evidence and strength of recommendations. This GRADE concept article is not GRADE guidance but introduces certainty of net benefit, defined as the certainty that the balance between desirable and undesirable health effects is favourable. Determining certainty of net benefit requires considering certainty of effect estimates, the expected importance of outcomes and variability in importance, and the interaction of these concepts. Certainty of net harm is the certainty that the net effect is unfavourable. Guideline panels using or testing this approach might limit strong recommendations to actions with a high certainty of net benefit or against actions with a moderate or high certainty of net harm. Recommendations may differ in direction or strength from that suggested by the certainty of net benefit or harm when influenced by cost, equity, acceptability or feasibility.

Finding the Balance Between Benefits and Harms When Using Statins for Primary Prevention of Cardiovascular Disease: A Modeling Study

BACKGROUND

Many guidelines use expected risk for cardiovascular disease (CVD) during the next 10 years as a basis for recommendations on use of statins for primary prevention of CVD. However, how harms were considered and weighed against benefits is often unclear.

OBJECTIVE

To identify the expected risk above which statins provide net benefit.

DESIGN

Quantitative benefit-harm balance modeling study.

DATA SOURCES

Network meta-analysis of primary prevention trials, a preference survey, and selected observational studies.

TARGET POPULATION

Persons aged 40 to 75 years with no history of CVD.

TIME HORIZON

10 years.

PERSPECTIVE

Clinicians and guideline developers.

INTERVENTION

Low- or moderate-dose statin versus no statin.

OUTCOME MEASURES

The 10-year risk for CVD at which statins provide at least a 60% probability of net benefit, with baseline risk, frequencies of and preferences for statin benefits and harms, and competing risk for non-CVD death taken into account.

RESULTS OF BASE-CASE ANALYSIS

Younger men had net benefit at a lower 10-year risk for CVD than older men (14% for ages 40 to 44 years vs. 21% for ages 70 to 75 years). In women, the risk required for net benefit was higher (17% for ages 40 to 44 years vs. 22% for ages 70 to 75 years). Atorvastatin and rosuvastatin provided net benefit at lower 10-year risks than simvastatin and pravastatin.

RESULTS OF SENSITIVITY ANALYSIS

Most alternative assumptions led to similar findings.

LIMITATION

Age-specific data for some harms were not available.

CONCLUSION

Statins provide net benefits at higher 10-year risks for CVD than are reflected in most current guidelines. In addition, the level of risk at which net benefit occurs varies considerably by age, sex, and statin type.

PRIMARY FUNDING SOURCE

Swiss Government Excellence Scholarship Office, Béatrice Ederer-Weber Foundation, and North-South Cooperation at the University of Zurich.