Efficacy of Anti-Obesity Medications for Weight Reduction in Older Adults: A Systematic Review

Abstract

Objective

To examine weight reduction and adverse events associated with use of anti-obesity medications (AOMs) in older adults aged ≥65 years.

Methods

Seven databases were searched for studies evaluating weight reduction of FDA-approved AOMs. Studies had to include adults ≥65 years with obesity (body mass index (BMI) ≥30kg/m² or BMI ≥27kg/m² with 1 weight-related condition), with independent analysis of weight reduction for adults age ≥65. Two coauthors extracted and evaluated studies for risk of bias using standardized forms.

Results

Six experimental studies (five secondary analyses of randomized clinical trial data and one single-arm trial) and two observational studies met inclusion criteria. Seven medications were studied. Sample size of older adults ranged from 13 to 6,728. Experimental studies included predominantly patients with concurrent pre-diabetes or cardiovascular disease. All studies found statistically significant weight reduction between intervention and placebo group or compared to baseline weight. Few studies reported on adverse events.

Conclusion

Limited evidence suggests weight reduction of AOMs in older adults, with the best current evidence for the use of semaglutide in older adults with obesity and cardiovascular disease. Larger, more inclusive studies of older adults are needed to guide clinical care and determine the tolerability of AOMs for older adults.

Introduction

While medications to treat obesity have been used for decades, the treatment of obesity has been transformed with the introduction of new, highly effective anti-obesity medications. Average weight reduction achieved with semaglutide was 14.9% and ranged from 15.0 to 20.9% for escalating doses of tirzepatide in clinical trials (1, 2), and these medications provide concomitant cardiovascular, endocrine, renal, and pulmonary benefits(3, 4, 5, 6, 7). In part owing to the mounting evidence for improvement in health outcomes, there is growing enthusiasm to treat obesity with pharmacotherapies, including among the 42% of older adults with obesity in the U.S.(8).

However, the data on the efficacy and safety of anti-obesity medications in older adults are limited. Despite the high prevalence of obesity among older adults, only about 1 in 10 participants of the pivotal obesity trials used for the U.S. Food and Drug Administration (FDA) approval of semaglutide and tirzepatide were ≥65 years of age(1, 2, 9, 10, 11). Trials of older anti-obesity medications also enrolled few older adults. The mean ages of trial participants for phentermine-topiramate, orlistat, and liraglutide ranged from early 40’s to mid 50’s(12, 13, 14, 15, 16, 17, 18), and the trials testing naltrexone-bupropion excluded adults over 65(19, 20). Consequently, limited data are available to guide medical therapy of obesity in older adults.

Obesity in older adults may be different than obesity in younger adults – and this means that evidence derived from obesity studies conducted largely among younger adults may not be directly applicable to older adults. Aging-related changes in body composition result in an increase in body fat and a decline in muscle mass(21). Resting metabolic rate decreases with age without a decrease in calorie intake, resulting in a positive energy balance(21). Older adults have a longer exposure time to inflammation and mechanical stress caused by excess fat mass(22), and as a result, they have more obesity-related conditions(23). In addition, older patients may respond to medications differently than younger adults due to body composition, reduced hepatic and renal clearance, and potential for medication interactions(24). Therefore, efficacy of anti-obesity medications may differ between younger and older adults.

In addition, the frequency and severity of side effects of anti-obesity medications may differ in older compared with younger adults. For example, in clinical trials, liraglutide, semaglutide and tirzepatide were associated with 20% or higher incidence of gastrointestinal side effects, including nausea, vomiting, diarrhea, and constipation, which could lead to dehydration(1, 2, 17). The aging-related changes in the autonomic nervous system, concomitant medications, and arterial stiffness put older adults at risk for orthostatic hypotension(25). Decreased oral intake from gastrointestinal side effects of semaglutide and tirzepatide may exacerbate this, leading to orthostatic hypotension and falls. Dehydration could also result in acute kidney injury, especially in older adults who have elevated risk of chronic kidney disease(26). Studies evaluating the effects of these medications specifically in older adults are needed to provide better understanding of their safety and tolerability profile. Further, weight reduction can result in both bone and muscle mass loss(27, 28), which could be problematic since aging reduces muscle mass and bone mineral density(21, 29).

Three prior systematic reviews have been published on the treatment of obesity in older adults(30, 31, 32). Only one review(30), published in 2019, included studies of anti-obesity medications. This review found two studies on anti-obesity medications which are currently approved – one single arm study of thirteen patients taking orlistat(33) and one case series of nine patients taking liraglutide(34). The review found no RCTs examining the effects of anti-obesity medications in older adults. Given the recent approval of semaglutide in 2021 and tirzepatide in 2023 for treatment of obesity, a contemporary review is needed to understand the current state of the evidence.

Therefore, our primary objective was to perform a systematic review of the literature to determine the weight reduction effect of anti-obesity medications in older adults. Our secondary objective was to evaluate the adverse events or tolerance of anti-obesity medications in older adults.

Methods

We utilized the Cochrane Handbook for Systematic Reviews of Interventions to develop the methods for this study(35). This systematic review was reported in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 and Synthesis Without Meta-Analysis (Table S1)(36, 37). This systematic review is registered on PROSPERO (CRD42024498966), and the protocol and materials can be accessed at https://www.crd.york.ac.uk/prospero/. Protocol amendments are found on Table S5.

Data sources and literature search

An exhaustive search of the literature was conducted by a medical research librarian (AAG) in Cochrane Library, Google Scholar, Ovid Embase, Ovid Medline, Scopus, and Web of Science Core Collection databases to find relevant articles published from the inception of each database to June 6, 2024. Databases were searched using a combination of keywords and controlled vocabulary for FDA-approved anti-obesity medications used for treatment of overweight or obesity that included older adults. The search was not limited by language, publication type, or year (full search strategies available in Table S2). The search was peer-reviewed by a second medical librarian using the Peer Review of Electronic Search Strategies (PRESS)(38). Forward and backward citation chasing was performed using CitationChaser to identify additional relevant studies not retrieved by the database search(39). We also asked experts to identify articles that should be considered for inclusion in the review.

Study selection, data extraction, and synthesis

Studies were included if they were observational or experimental studies that evaluated weight reduction of anti-obesity medications, included adults ≥65 years eligible for anti-obesity medications, reported analyses by age ≥65, and were in the English language. Eligibility for anti-obesity medications was defined as either body mass index (BMI) ≥30 kg/m² or BMI ≥27 kg/m² to <30 kg/m² with one weight related comorbidity(40). For Asian populations, eligibility for anti-obesity medications was defined as a BMI ≥25 kg/m²(41). Anti-obesity medications included all currently approved by the FDA for chronic and short term treatment: orlistat, phentermine-topiramate, naltrexone-bupropion, liraglutide, subcutaneous semaglutide, tirzepatide, phentermine, benzphentamine, diethylpropion, and phendimetrazine(42, 43). We excluded studies focused only on monogenic obesity, syndromic obesity (including trisomy 21, Bardet-Biedl syndrome, Prader-Willi syndrome, and Fragile-X), hypothalamic obesity, or treatment of diabetes (given that dosages of anti-obesity medications and lifestyle counseling were different than in obesity treatment trials).

Search results from all databases were imported into an Endnote 21 library. Duplicates were removed using the Yale Reference Deduplicator(44). The deduplicated results were then imported into Covidence, a systematic review software for screening and data extraction. Two authors independently applied inclusion and exclusion criteria using Covidence systematic review software. After initial screening of titles and abstracts, studies were then evaluated for inclusion based on full text and supplemental data review. Any disagreements in title and abstract screening or full text review were resolved by consensus. Citations of included studies were additionally searched. Data were extracted from studies independently by two authors in Qualtrics. Extracted data included publication characteristics, type of study, sample size of total population and population ≥65, demographic and clinical characteristics of the sample, and weight reduction effect of anti-obesity medications for older adults (measured as percent weight reduction or absolute weight reduction). Data on adverse events or intolerance to treatment for older adults taking anti-obesity medications were also extracted, when available.

Characteristics of studies, results according to review aims, and risk of bias are described by type of study. Weight reduction effect estimates for studies were summarized in tables.

Risk of Bias rating

Risk of bias for experimental studies was rated using the Cochrane risk of bias tool for clinical trials(45). This tool measures bias in five domains: 1) randomization process, 2) deviation from the intended intervention, 3) missing outcome data, 4) measurement of the outcome, and 5) selection of reported results. Each domain is evaluated as either low risk of bias, some concerns, or high risk of bias. According to the tool, studies with a single category as high risk of bias are rated overall as high risk of bias. Risk of bias for observational studies was evaluated using the Newcastle-Ottawa Scale(46), modified to the objective of this review (Table S4). Studies were rated in three domains: selection, comparability, and outcome. Studies could obtain a total of eight stars for their rating. Two authors independently rated the risk of bias, and differences were resolved via consensus. For studies that were a secondary analysis of an RCT, the manuscript describing the original trial was referenced to inform response to risk of bias questions.

Results

Publication selection

Database searches resulted in 7,646 citations (Figure 1). After removing duplicates, 6,250 citations underwent title and abstract screening. Of these, 238 citations met the criteria for full text review. Subsequently, seven studies met the inclusion criteria for the study. Nine publications were identified by experts for review. We excluded 232 studies from the search results and 48 studies in citation chasing and expert recommendation results due to results not stratified by age, wrong patient population, weight reduction not an outcome, wrong study design, wrong medication, duplicate study data, non-English language, or the study had been retracted (Table S3). Of the seven publications, four were full manuscripts and three were conference abstracts (see Table 1) describing eight separate studies. Given that individual publications described multiple studies, the results are described by individual study here.

Figure 1:

PRISMA flow chart

Adapted from: Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021;372: n71. doi: 10.1136/bmj.n71. For more information, visit: http://www.prisma-statement.org/

Table 1:

Characteristics of Included Studies

Authors	Year	Type of Publication	Type of studies described	Name of trial described, if applicable	Study population
Varli et al.	2010	Manuscript	Experimental		Aged 66–83, BMI > 35 kg/m2, no diabetes
Gadde et al.	2011	Manuscript	Experimental	CONQUER	18–70 years old, BMI of 27–45 kg/m2 with two or more weight-related conditions
Valladales-Restrepo et al.	2023	Manuscript	Observational		18 or older, BMI ≥30 kg/m2, and initiated either liraglutide or orlistat
Ryan et al.	2024	Manuscript	Experimental	Weight reduction outcomes of SELECT	45 or older, BMI >27 kg/m2 who had cardiovascular disease (myocardial infarction, stroke, symptomatic peripheral arterial disease), no diabetes
Dubin et al.	2018	Abstract	Observational		over 65 and treated with anti-obesity medication three months
Wilding et al.	2018	Abstract	Experimental	SCALE obesity and pre-diabetes	18 or older, BMI ≥30 kg/m2 or a BMI ≥27 kg/m2 with one or more weight-related conditions, no diabetes
Wilding et al.	2018	Abstract	Experimental	SCALE diabetes	18 or older, BMI ≥27 kg/m2 who had type 2 diabetes
Kushner et al.	2022	Abstract	Experimental	Body composition sub-study of SURMOUNT-1	18 or older, BMI ≥30 kg/m2 or a BMI ≥27 kg/m2 with one or more weight-related conditions, no diabetes, participated in body composition sub-study

Study Characteristics

Among eight included studies, six were experimental studies (five secondary analyses of randomized clinical trials [RCTs] and one single arm trial) and two were observational studies. Five studies were randomized, blinded, placebo-controlled trials(13, 47, 48, 49). One study was a non-randomized, open label trial of orlistat(33). Both observational studies were retrospective cohort studies using electronic health record data(50, 51). Study characteristics are described in Table 1.

Participants

Inclusion criteria for participation in experimental studies and observational studies are described in Table 1. All five experimental studies and one observational study had a BMI requirement which varied from ≥27 kg/m² to >35 kg/m²(13, 33, 47, 48, 49, 50). Three experimental studies had inclusion criteria for weight-related conditions. One study required participants to have type 2 diabetes(47). One study required participants to have 2 or more weight-related comorbidities, which resulted in 68% of the trial population having diabetes(13). One study required participants to have cardiovascular disease, including prior myocardial infarction, prior stroke, or peripheral artery disease(48). One study preferentially enrolled participants with pre-diabetes, resulting in 61.2% of the trial population having pre-diabetes(47). Four experimental studies excluded adults who had diabetes(33, 47, 48, 49). Only two studies limited enrollment to adults ≥65(33, 51). Observational studies additionally limited inclusion based on receipt of select anti-obesity medications(50, 51).

Baseline characteristics of participants ≥65 years of age were described in two studies. In the single arm study of orlistat, mean age was 71.2 years (SD 4.9 years), mean BMI was 39.9 kg/m² (SD 4.6 kg/m²)(33). Older adults included in the study by Dubin, et al. had a mean age of 71.1 years (SD 4.2), were all White, were 80.7% female, and had a mean BMI of 36.1 kg/m² (SD 5.3).

Interventions

The most commonly studied medication was liraglutide (4 studies). No studies evaluated naltrexone-bupropion, benzphentamine, diethylpropion, or phendimetrazine. Medications included in studies can be seen in Figure 2. Five experimental studies used up to the maximum doses of liraglutide, phentermine-topiramate, orlistat, semaglutide, and tirzepatide(13, 33, 47, 48, 49). Five experimental studies compared the study drug to placebo(13, 47, 48, 49), while one study had no comparison group(33). Medications and comparators of each experimental study can be seen in Table 2. Doses of medications studied in the observational studies were not described, and there were no comparator groups. Medications studied in each observational study can be seen in Table 3.

Figure 2:

Representation of medications in studies of anti-obesity medications inclusive of and reporting on outcomes in older adults

Table 2:

Characteristics of and weight reduction efficacy in experimental studies analyzing anti-obesity medications in older adults

Reference	Study Name	Secondary analysis	Total sample size	Sample aged ≥65	% of population ≥65	Intervention medication and dose(s)	Comparison group	Duration	Weight change in intervention group	Weight change in comparison group	Estimated treatment difference
Manuscripts
Varli et al. 2010		No	13	13	100.0%	Orlistat 120 mg tid	N/A	24 weeks	−8.4% (SD=1.2 kg)	N/A	−8.4% (SD=1.2 kg)
Gadde et al. 2011	CONQUER	Yes	2487	219	8.8%	Phentermine 7.5 mg – topiramate 46 mg Phentermine 15 mg – topiramate 92 mg	placebo	56 weeks	−7.8% (5.7%–9.9%) −9.4% (8.0%–10.9%)	−3.2% (1.6%–4.7%)	−4.6% −6.2%
Ryan et al. 2024	Weight-reduction outcome in SELECT	Yes	17604	6728	38.2%	Semaglutide 2.4 mg	Placebo	59 months	NA	NA	Age 65 to <75 −8.1% (−8.53 to −7.66) Age ≥75 −7.49% (−8.37 to −6.62)
Abstracts
Wilding et al. 2018	SCALE Obesity and Pre-diabetes	Yes	3652	201	5.5%	Liraglutide 3.0 mg	placebo	56 weeks	–8.4%	–4.2%	−4.2%
Wilding et al. 2018	SCALE diabetes	Yes	621	119	19.2%	Liraglutide 3.0 mg	placebo	56 weeks	–7.2%	–2.5%	−4.7%
Kushner et al. 2022	Body Composition sub-study in SURMOUNT-1	Yes	255	20	7.8%	Tirzepatide 5 mg, 10 mg, 15 mg	Placebo	72 weeks	–22%	–3.8%	–18.2%

Table 3:

Characteristics of and weight reduction efficacy in observational studies analyzing anti-obesity medications in older adults

Reference	Study context	Data source	Total sample size	Sample aged ≥65	% of total	Medication	Duration	Weight reduction effect
Manuscript
Valladales-Restrepo et al. 2023	Patients enrolled in Columbian health care system	Electronic health records and pharmacy fill data	294	38	12.9%	Liraglutide, orlistat	52 weeks	Adults aged 65 and older had 3.8 (95% CI 1.2–12.4) times the odds of achieving 5% weight loss compared to adults under 40
Abstract
Dubin et al. 2018	Obesity metabolic clinic in Venice, Florida	Electronic health record	57	57	100%	Liraglutide, lorcaserin, naltrexone-bupropion, phentermine, phentermine-topiramate, phentermine-topiramate extended release	12 weeks	−11.4% (SD 1.4%)

In addition to receiving the study drug or placebo, the participants in the five RCTs also received lifestyle counseling(13, 47, 48, 49). Three RCTs provided diet and exercise counseling consisting of a 500 kcal deficit diet and 150 minutes of physical activity per week(47, 49). One RCT provided individualized dietary and physical activity counseling, with a focus on lifestyle modifications to manage cardiovascular risk(48). One RCT did not detail the lifestyle counseling provided(13).

Duration of studies ranged from 24 weeks to 59 months for experimental studies(13, 33, 47, 49) and from 12 to 16 weeks in length for observational studies(50, 51). Duration of individual studies can be seen in Table 2 for experimental studies and Table 3 for observational studies.

Analysis and Weight Reduction Efficacy

Secondary analyses of RCTs stratified results by age to determine the weight reduction effect in older adults with obesity. The primary outcome of five experimental studies(13, 33, 47, 49) was weight reduction, while the primary outcome for one experimental study was a composite outcome of death from cardiovascular causes, nonfatal myocardial infarction, or non-fatal stroke, with a prespecified secondary outcome of weight reduction(48). One study was a body composition sub-study of a larger obesity trial, which limited the available sample of older adults to 20(49). All six experimental studies presented weight reduction results as percentage of total body weight reduction(13, 33, 47, 48).

All studies showed statistically significant weight reduction among older adults taking anti-obesity medications. All secondary analyses of RCTs showed statistically significant weight reduction for adults ≥65 in treatment groups compared to placebo groups(13, 47, 49), and the single arm study of orlistat showed weight reduction compared to baseline weight(33). Weight reduction outcomes of experimental studies can be seen in Table 2 separated by publication type (manuscripts and abstracts).

The primary outcome for both observational studies was weight reduction(50, 51). One study presented percentage body weight lost from baseline(51), and the other study presented odds of achieving 5% weight reduction for adults ≥65 as compared to adults younger than 40(50). Weight reduction outcomes of observational studies can be seen in Table 3 separated by publication type.

Adverse events and treatment intolerance

Two experimental studies(47) and one observational study(51) described adverse events or intolerance to therapy specifically among older adults. Authors of the secondary analyses of the SCALE trials narratively described an increase in reported adverse events in older adults, especially gastrointestinal events, but did not quantify this increase(47). One retrospective cohort study found that intolerance to therapy was noted by 15.4% of older adults taking phentermine-topiramate extended release, 40% of older adults taking naltrexone-bupropion, 33% of older adults taking phentermine-topiramate, 25% taking liraglutide, and none for phentermine(51).

Risk of Bias Ratings

Cochrane risk of bias ratings for experimental studies can be seen in Figure 3. Risk of bias rating was completed at the level of each study (i.e., not each manuscript). The SCALE trial, CONQUER trial and body composition sub-study of the SURMOUNT-1 trial were at high risk of bias due to missing outcome data and differing reasons for missing data across intervention and comparator arms(1, 13, 16, 17, 48, 49). In these trials, participants in the intervention arms more often had missing follow-up data arising from withdrawal due to adverse events, and participants in the placebo arms more often had missing follow-up data arising from withdrawal due to ineffective therapy, potentially biasing efficacy estimates towards the null. The SELECT trial had low risk of bias(3, 48). Varli et al. had high risk of bias due to missing outcome data and some concerns arising from deviations from intended intervention and selection of the reported result(33).

Figure 3:

Cochrane Risk of Bias ratings for Experimental Studies

Risk of bias assessment of included observational studies can be seen in Table 4. Valladales-Restrepo et al. received 7 out of 8 stars(50). The study had bias in the outcome, where there was an incompletely described loss to follow-up over the study timeframe. Dubin, et al., received 4 out of 8 stars, demonstrating biases in selection and comparability(51).

Table 4:

Modified Newcastle Ottawa Quality Assessment Scale for Observational Studies

	Dubin et al.	Valladales-Retstrepo et al.
Selection
Representativeness of the exposed cohort		*
Ascertainment of the exposure	*	*
Demonstration that outcome of interest was not present at the start of study	*	*
Comparability
Comparability of cohorts on the basis of the design or analysis		**
Outcome
Assessment of outcome	*	*
Was the follow-up long enough for outcomes to occur	*	*
Adequacy of follow-up cohorts
Total	4/8	6/8

All categories have one possible star except comparability of cohorts the basis of the design or analysis which has two possible starts.

According to the Cochrane risk of bias tool, five experimental studies had significant (>5%) rates of loss to follow-up. In SCALE diabetes, SCALE obesity and pre-diabetes, CONQUER, and body composition sub-study of SURMOUNT-1 outcome data was missing for 17.4%, 22.2%, 38.9%, and 37.3% of total trial participants, respectively(1, 13, 16, 17). Importantly, none of the RCT studies described loss to follow-up rates specifically among older adults. In the single arm orlistat trial, 27.8% of the trial sample was lost to follow up(33). One observational study did not describe missing data(51). The other observational study noted significant missing data in their limitations, but the magnitude of missing data was not described(50).

Discussion

This systematic review identified eight existing studies (described by four manuscripts and three abstracts) examining the weight reduction effect of FDA-approved anti-obesity medications among older adults. The largest study included over 6,700 older adults and demonstrated significant weight reduction effect of semaglutide among older participants. Little evidence is available for other anti-obesity medications as the other included experimental studies were limited by small sample sizes of older adults and high risk of bias, mainly due to participant attrition and therefore missing data. Only two observational studies examined weight reduction associated with anti-obesity medications among older adults, and they were also limited by small sample sizes and were rated as moderate quality. Overall, there is a paucity of data for the weight reduction efficacy for most anti-obesity medications in older adults and, thus more studies are needed.

It is important to consider specific subpopulations of patients included in the anti-obesity medication studies. Trials of liraglutide and phentermine-topiramate were mainly conducted among participants with diabetes or prediabetes. Clinical trials of GLP-1RAs and the GIP/GLP-1 RA have shown that adults with diabetes lose less weight than those without diabetes(1, 11, 52). Additionally, the potential benefit of treatment with anti-obesity medications may also differ between those with and without diabetes or prediabetes, making these results less generalizable to older adults without diabetes and pre-diabetes. The study of tirzepatide included in this review specifically excluded adults with diabetes; however, the sample of older adults was small (N = 20) in the sub-study of 160 adults (49). The SELECT trial of semaglutide also excluded people with diabetes, rather all participants had established cardiovascular disease(48). Therefore, these results pertain to older adults with obesity and cardiovascular disease rather than obesity alone.

The SELECT trial enrolled the largest number of older adults taking an anti-obesity medication to date. The secondary analysis of this trial demonstrated significant weight reduction with semaglutide in older adults, with the estimated treatment difference between semaglutide and control groups between 7.5% and 8.1%. Notably, the percentage weight reduction seen in the SELECT trial is less than seen in STEP 1, the first large, phase 3 trial of semaglutide for weight reduction, where the estimated treatment difference was −12.4% at week 68(2). Authors of the secondary analysis of SELECT speculated several reasons for this – weight reduction was emphasized less to SELECT participants compared to STEP 1 participants, the dose of semaglutide was ramped up at a slower rate in SELECT if side effects developed, and fewer participants in SELECT achieved the maximum dose of semaglutide. Additionally, participants in SELECT were older, with a mean age of 61.6 compared to a mean age of 46 in STEP 1(2, 3), and there were fewer women in SELECT, all of which could have affected weight reduction.

We found few data on the adverse event rates associated with the use of anti-obesity medications in older adults. Two experimental studies found adverse events increased with age(47). One retrospective cohort study found intolerance ranged from 15% to 40% in various anti-obesity medications(51); these reported rates were higher than treatment discontinuation due to adverse events in clinical trials of the same medications(13, 17, 19). Thus, a better understanding of the tolerability profile of anti-obesity medications in older adults is needed to guide clinical practice. Tolerability data could be very helpful to inform how medication doses can optimally be up-titrated in older adults and when dose reduction or medication cessation is needed. There also may be certain pre-existing conditions that predispose certain older adults to adverse events, and this knowledge could help guide decision-making for older adults who meet treatment criteria.

Although this review shows that several anti-obesity medications lead to significant weight reduction in older adults, more studies are needed to determine if these medications can meet important treatment goals relevant to older adults. It is currently unclear how anti-obesity medications impact the physical function of older adults. Trials conducted among adults taking liraglutide and semaglutide for obesity treatment have shown improvements in self-reported physical function and quality of life(2, 17), so it is possible older adults may reap these benefits. However, given that older adults with obesity often have physical function limitations and may have sarcopenia(21), there is potential for harm. Since physical function and quality life of trial participants were monitored in many of the trials of anti-obesity medications, there is an opportunity to perform analyses stratified by age. It is additionally crucial to measure the physical function and quality of life of older adults taking anti-obesity medications outside of clinical trials where exercise counseling may differ. Longitudinal monitoring of bone density, muscle mass, and muscle strength is also needed. Given new anti-obesity medications have additionally been shown to provide secondary cardiovascular disease prevention, improvement in sleep apnea, and improvement in blood pressure, lipids, and dysglycemia in clinical trial populations(1, 2, 3, 7, 10, 11, 17), treatment of older adults with anti-obesity medications may result in improvements in weight-related conditions and potentially help mitigate polypharmacy. Further analyses of trials that included older adults (e.g., SURMOUNT-1 which included 6% of participants aged ≥65 years(1)) and of observational data of older adults taking these medications are needed to further understand if these potential benefits are present in older adults.

There are additional limitations to the studies identified in this review. First, most analyses did not describe baseline characteristics of the older adults in the trials, making it unclear if trial populations are representative of the older adult population with obesity and whether characteristics of older adults were balanced between groups. Second, three of the included articles were conference abstracts. These studies were included due to the paucity of available data; however, there are details not described in abstracts that limit the use of these data – no baseline characteristics, few methodological details, and no listed limitations of the studies. Without this information, it is challenging to understand the rigor or assess the quality of these studies. Third, samples sizes in observational studies were also small, and only one study controlled for confounding factors. Fourth, observational studies included the effect of multiple weight reduction medications into one point estimate. The various anti-obesity medications have different mechanisms of action and different magnitudes of weight reduction(53), and it is therefore difficult to interpret a point estimate that combines them. While observational data could be an important tool to understand the weight reduction effect of these medications for older adults, larger and higher quality studies are needed.

Despite the limitations, this review has important clinical implications for the treatment of older adults with obesity. Promising data from SELECT shows older adults with obesity and established cardiovascular disease taking semaglutide lose 7–8% of their body weight compared to placebo; it is unknown if the magnitude of weight change is the same for older adults without cardiovascular disease.(48) Our study also found trials that assessed the weight reduction effect of older medications including phentermine-topiramate and orlistat; however, the use of these medications in older adults may be limited by their side effects including dizziness, insomnia, and cognitive impairment(54) and fecal urgency(55), respectively. Given minimal observational data on older adults, it remains unclear whether weight reduction achieved in the context of a clinical trial (with access to frequent visits, monitoring, and generally better medication adherence(56)) is similar to weight reduction achieved in usual care settings. There are few data regarding the safety profile of anti-obesity medications specifically for older adults. Clinicians should carefully consider the risks versus benefits of treatment unique to older adults (for example, potential risks: gastrointestinal side effects and dehydration; potential benefits: prevention and treatment of weight-related conditions) with these medications.

This review has important implications for future research. First, older adults, particularly those who represent the population of older adults living with obesity, should be targeted for enrollment in future obesity trials. Second, existing data from many trials of anti-obesity medications have not been stratified by age to understand effects among older adults. Data from multiple RCTs could be pooled in an individual patient level meta-analysis to determine efficacy and safety of these medications. Third, more observational studies should be done to analyze the use of these medications in older adults. With the advancement of health information technology, numerous databases exist which track both pharmacy fill data and clinical data(57). These resources could be utilized to evaluate effectiveness and safety of these medications for older adults. Fourth, studies evaluating the physical function and quality of life of older adults taking anti-obesity medications are needed.

Limitations

Our review should be considered in the context of several limitations. Results of a systematic review can be subject to publication bias, where negative studies are less frequently published. Half of the included publications were abstracts, which provide limited information on the analysis. We were unable to perform a meta-analysis due to the heterogeneity of trial populations, interventions, duration, outcome analysis, and study quality. Our review included studies that measured weight reduction and reported on the secondary outcome of adverse events. There may be studies that reported only adverse events and not weight reduction for older adults, which would not have been included in this review.

Conclusions

There are few studies testing the weight reduction efficacy of anti-obesity medication in older adults, but existing data demonstrate significant weight reduction effect of several medications in this population. The largest study demonstrated significant weight reduction efficacy of semaglutide in older adults with obesity and cardiovascular disease. Other secondary analyses of RCTs were limited by low enrollment of older adults and high risk of bias. There are very few observational studies that examined weight reduction effect of anti-obesity medications among older adults. Few studies reported rates of adverse events specifically in older adults. Efforts should be made to leverage existing trial data to examine weight reduction effects stratified by age. Large, well-designed observational studies are needed to better understand the efficacy and safety of anti-obesity medications in older adults.

Study Importance.

What major reviews have been published on this subject?

Three systematic reviews have been published on the treatment of obesity in older adults, but only one included anti-obesity medications in their search. With the recent approval of semaglutide and tirzepatide for treatment of obesity, a new systematic review is needed to understand the state of the evidence.

What are the new findings in this manuscript?

Eight studies were included and demonstrated significant weight reduction with anti-obesity medications in older adults. One trial compared semaglutide to placebo in over 6,700 older adults. Other studies were limited by small sample sizes and moderate to high risk of bias, mainly due to attrition. Few studies reported rates of adverse events specifically among older adults.

How might these results change the direction of research or focus of clinical practice?

Limited evidence in this review suggests several anti-obesity medications can effectively reduce weight among older adults, but safety and adverse event data specific to older adults are lacking. More clinical trials and observational studies of older adults taking anti-obesity medications are needed.