We estimate 93 million US adults with overweight and obesity would be potentially eligible for tirzepatide for weight loss based on SURMOUNT-1 eligibility criteria, and of these, 85.9, 78.0, 65.9, and 53.0 million people would show ≥ 5%, ≥ 10%, ≥ 15%, or ≥ 20% weight reductions, respectively. Over half (58.8%) of those initially with obesity would no longer have obesity after treatment with 15 mg tirzepatide. Importantly, we project up to 2 million CV events could be prevented over a 10-years of treatment (among those without prior cardiovascular disease), corresponding to a 2.39% absolute and 23.57% relative risk reduction. Moreover, a cardiovascular event could be expected to be prevented for every 42 eligible persons treated with tirzepatide 15 mg for 10 years. We believe this results to have significant implications on the US population-wide impact of tirzepatide 15 mg therapy for persons with overweight or obesity as well as for ongoing CVD outcomes trials involving tirzepatide.

We also project > 20% weight losses will be observed each in more than 25 million females and males with corresponding reductions in obesity prevalences. Given the US population distribution, the greatest number of persons with these benefits will occur in White, followed by Black, Hispanic, and Asian persons. The greatest number of preventable CVD events are predicted to occur in males, given their greater baseline CVD risk, compared to females, as well as in White persons given their greater population size and baseline risk among our US sample, compared to other race/ethnic groups.

GLP1-RA as well as newer GIP-GLP1-RA are of great interest to treat diabetes and obesity. Since the US Food and Drug Administration (FDA) approval of liraglutide 3.0 mg for obesity [14], subcutaneous semaglutide 2.4 mg was later approved for obesity [15] and most recently tirzepatide also received approval for obesity treatment [16]. It has been previously reported that GLP1-RA therapy reduces weight more in those without diabetes (but with overweight or obesity) than among those with diabetes [17]. Daily liraglutide (3.0 mg) for persons without DM, but with obesity, showed in the SCALE study an average weight loss of 8%, with 63%, 33%, and 14% losing at ≥ 5%, > 10%, and > 15% of body weight, respectively [18]. Using similar BMI cutpoints for tirzepatide 15 mg in the SURMOUNT-1 trial, there was an overall 22.1 kg or 20.9% weight loss, with 90.9% and 83.5% showing weight reductions of at least 5% and 10%, respectively [11]. Most recently, oral GLP1-RA therapies, such as semaglutide 50 mg in the OASIS 1 trial, showed significant weight reductions averaging 15.1% [19], similar to that achieved in the earlier STEP 1 trial [9] with injectable semaglutide. Finally, early studies of the oral non-peptide GLP-1 receptor agonist orforglipron [20], as well as newer triple hormone agonists such as retatrutide [21], are showing benefits of significant weight loss efficacy.

Both GLP1-RA and GIP-GLP-1 RA therapies have also demonstrated improvements in cardiometabolic risk factors, making them promising targets for cardiovascular risk reduction. Liraglutide 3.0 mg daily in the SCALE study showed systolic and diastolic blood pressure reductions of 4.2 and 2.6 mmHg, respectively, and reductions in LDL-cholesterol and triglycerides of 3.0% and 13.3%, respectively, with increases in HDL-cholesterol of 2.3% [18]. In SURMOUNT-1 involving tirzepatide, there were overall (pooled treatment groups) reductions of 15% in body weight, SBP of 7.2 mmHg, total cholesterol of 4.8%, and LDL-cholesterol of 5.8%, increases in HDL-cholesterol of 8.0%, and reductions in triglycerides of 24.8% [11]. Most recently, the SELECT trial involving subcutaneous semaglutide 2.4 mg in overweight or obese persons with pre-existing CVD demonstrated a 20% relative risk reduction in cardiovascular outcomes [22]. Of note, subcutaneous semaglutide 2.4 mg most recently got the additional FDA indication for cardiovascular risk reduction in persons with CVD and overweight/obesity [23]. The SURMOUNT-MMO trial involving tirzepatide [12] will further inform on the role of these newer obesity therapies in reducing CVD outcomes in adults with overweight and obesity with and without prior CVD, and most importantly, confirming in the clinical trial setting whether our projected risk reductions in the present paper hold.

Of interest is the degree of weight loss required to have CVD benefits. Our study projects a 24% relative risk reduction for CVD events based on tirzepatide’s overall > 20% weight reduction (with 84% of patients showing > 10% weight reductions). This compares favorably a post hoc analysis of the Look AHEAD trial data, where the subgroup of subjects among both intervention and control groups combined who lost at least 10% of their body weight in the first year of the study had a 21% lower risk of the primary composite CVD outcome over 10 years of follow-up (realizing, however, the original randomized study did not meet the primary outcome) [24].

There are, however, few data on the population-wide impact of these therapies in populations with overweight or obesity, while their eligibility and impact on CVD events have been previously studied among persons with diabetes. Arnold et al. showed in the US Diabetes Collaborative Registry [25] among 182,525 patients with diabetes derived from 313 multispecialty practices, 48% of patients to fit LEADER eligibility criteria, and estimating 247 MIs and 329 CV deaths per year of treatment could be prevented. We also estimated liraglutide eligibility and potential preventable events among US NHANES adults with diabetes, estimating 15.4% (4.2 million) of those with diabetes to fit LEADER eligibility criteria and from LEADER trial CVD outcome risk reductions observed, 21,209 primary composite CVD events could be prevented annually [26]. Moreover, in a recent analysis of NHANES [27], 51.1% of US adults were estimated to meet the FDA eligibility criteria for tirzepatide with Black (56.6%) and Hispanic adults (55.0%) most often qualifying. Most recently, we also examined the US population impact of subcutaneous semaglutide 2.4 mg, identifying 93 million US adults with overweight or obesity to be eligible based on STEP 1 trial eligibility criteria, with up to a 1.8% absolute (17.8% relative) CVD risk reduction projected that translates to an estimated 1.5 million preventable CVD events over 10 years [28]. The current study provides further data supporting the potential impact of tirzepatide-eligible US adults on obesity prevalence and CVD risk reduction.

Our study has several strengths, limitations, and assumptions. Our study of the NHANES cohort of US adults utilized sample weighting which allows for us to derive US population estimates of SURMOUNT-1 eligible US adults, as well as to estimate how many CVD events could be prevented from its use in the ethnically diverse US population. While NHANES does rely on self-reported measures such as CVD history and cigarette smoking, previous reports have confirmed the reliability of such self-report information [29]. Importantly, the well-known cardiovascular risk factors, namely, weight (for calculation of body mass index), blood pressure, and lipids, were all measured in NHANES participants.

In addition, our simulated impact of the SURMOUNT-1 trial results on our US population sample is not without limitations, most notably assuming the impact of weight reduction and cardiovascular risk factor changes will be similar in our sample as compared to the SURMOUNT-1 sample, as well as the projection of risk factor changes to CVD event reduction can be estimated by the CVD risk algorithms we utilized. Importantly, while NHANES is largely representative of the US population and forms the basis for reported US prevalence data on cardiovascular disease and its risk factors [30], it does not include institutionalized individuals and being a volunteer sample will also lack homeless and other underrepresented persons.

There are also differences between our participants and the SURMOUNT-1 clinical trial participants given the latter was an international trial (e.g., in sex/ethnicity distribution and the prevalence of certain comorbidities) that may affect actual CVD risk, and therefore, the observed effects on weight and cardiovascular risk factors seen in SURMOUNT-1 may not necessarily be translatable to our NHANES sample. For instance, with our sample being slightly older and with a higher proportion men as well as higher average cholesterol and systolic blood pressure levels would result in higher baseline CVD risk (although higher HDL-C levels would result in lower risk), our projections of CVD risk and preventable CVD events may be different than would be the case if our sample was more comparable to that of the SURMOUNT-1 trial.

We are also assuming treatment effects applied to our cohort are the same as in the SURMOUNT-1 trial which was of course a select clinical trial sample and not necessarily fully translatable to real-world populations such as NHANES, thus requiring validation from CVD outcomes trials such as SURMOUNT-MMO [12]. However, such clinical trials, rather forming the basis for treatment indications and guidelines, often have select eligibility criteria along with intensive efforts to ensure adherence to study medication, thus are also not fully translatable to real-world clinic populations.

Also, our estimates are based on the effects seen from the 15-mg dosage in SURMOUNT-1 and suggests the potential impact if all eligible patients were to be on this dosage of the drug; however, it is realized some patients may not achieve this dosage due to side effects or other reasons, and thus, our projections would be less than what we estimated. Our estimates also assume long-term safety and adherence to the therapy, and such information both on long-term safety and adherence are limited. Moreover, since sex and ethnic-specific effects of SURMOUNT-1 are not available, we utilized the overall trial effects for sex and ethnic-specific analyses, and if there were differential effects of tirzepatide by sex or ethnicity, our sex and ethnic-specific projections may be less precisely estimated.

Finally, we have applied the treatment effects seen from the tirzepatide group in the SURMOUNT-1 trial and we cannot be certain whether the preventable CVD events we estimate would be exclusively the result of tirzepatide. Also, since we do not have actual event data resulting from tirzepatide since the CVD outcomes trial is still ongoing, we have used the Framingham Risk Scores to estimate the number of CVD events that can be expected to occur over a defined period of time (e.g., 10 years). Our Framingham risk algorithms, like newer scores such as the Pooled Cohort equations, do not include all possible factors (e.g., family history or other “risk enhancing” factors) that could affect CVD risk. Two important advantages of the Framingham total CVD risk scores [13] we have used are that we are able to include persons beginning at age 30 as well as estimate the prediction of total CVD events (not just myocardial infarction and stroke, but also angina, heart failure and peripheral arterial disease).