Abstract
Objective:
To examine the use of pharmacotherapy in obesity treatment in the United States from 2011 to 2016 using a large, nationally representative sample.
Patients and Methods:
Data were obtained over six years, 2011 to 2016, from the National Ambulatory Medical Care Survey. There were three types of visits identified: patients with obesity and an anti-obesity drug mention, patients with obesity and no anti-obesity drug mention, and patients without obesity and with anti-obesity drug mention. Chi-square tests were used to compare characteristics across each type of visit. To predict the odds of an anti-obesity medication mention for patients with obesity, a logistic regression analysis was conducted.
Results:
Of the overall weighted 196,872,870 office-based physician visits made by patients with obesity from 2011 to 2016, one percent mentioned an anti-obesity drug. Additionally, there were 760,470 office-based physician visits by patients without obesity but with an anti-obesity medication mention. An anti-obesity drug mention was more likely for those ages 51 or older and those residing in the South (AOR: 5.31 95% CI: 1.19-23.59).
Conclusion:
There was a slight increase in anti-obesity medication mentions, from 0.26% in 2011 to 0.28% in 2016, but only one percent of office-based visits for patients with obesity received a prescription for an anti-obesity medication. Physicians tended to prescribe anti-obesity medications to those with obesity ages 51 or older and residing in the South. Anti-obesity medication as treatment for obesity is significantly underutilized.
Introduction
The prevalence of obesity, as reported by the Centers for Disease Control and Prevention, stands at 42.2% of adults in the US.1 Projections using multinomial regression analysis estimate that by 2030, nearly one in two adults will have obesity and nearly one in four adults will have severe obesity.2 The prescribing trends of anti-obesity medications in the United States from 1999 to 2010 have previously been described.3 Xia and colleagues utilized the National Ambulatory Medical Care Survey (NAMCS) and the National Hospital Ambulatory Medical Care Survey (NHAMCS) to find that only one in fifty (~2%) adult patients with obesity were prescribed an anti-obesity medication. The results support the notion that anti-obesity medications are underutilized compared to those prescription trends for other metabolic and chronic disease states. For example, it is estimated that 15 times more anti-diabetes medications are dispensed.4 This discrepancy likely stems from a variety of factors: past beliefs that weight status is determined by willpower, inadequate health insurance coverage for pharmacotherapeutics, and both patient and clinician concerns regarding efficacy and safety.5
Since the time of this original analysis by Xia and colleagues, there have been many developments in the field of obesity medicine. In 2013, the American Medical Association (AMA) House of Delegates passed resolution H-440.842, which recognized obesity as a disease and called on physicians to engage with medical interventions beyond lifestyle modification.6 Subsequently, the Endocrine Society released updated clinical practice guidelines, which recommended the use of approved weight loss medications to promote long-term weight reduction.7 Our understanding of the biochemical and neuroendocrine pathways regulating body weight has also evolved. It is known that hormonal adaptations persist for at least one year following calorie-restricted weight loss thereby influencing weight regain.4 These physiologic responses make it difficult for individuals to maintain weight loss through lifestyle changes alone, which prompts the consideration of adjunctive pharmacotherapy.8
Though still limited, from 2009 to 2017, there has been a 64% increase in coverage provided through state employee programs for the use of pharmacotherapy (from 12 to 23 states).9 Yet, data indicate that coverage of anti-obesity medications through the Affordable Care Act (ACA) is quite limited.10 Weight loss of five percent of body weight is considered clinically significant towards the reduction of cardiovascular risk factors.11,12 Clinical trials of currently approved medications demonstrate a weight loss between 4.0 and 10.9% from baseline, while newer agents such as once weekly subcutaneous injections of semaglutide demonstrate promising results with mean weight loss of 14.9% from baseline.5,13 Lastly, clinical trials continue to demonstrate that for most anti-obesity medications, adverse events are often minimal and well tolerated.14,15
Time and again, research has demonstrated that obesity is a multifactorial disease with considerable clinical and phenotypic heterogeneity that requires a multimodal treatment approach.16 The purpose of this study is to re-evaluate prescribing trends of anti-obesity medications among adults in the US using the most recent years of the National Ambulatory Medical Care Survey (NAMCS). The primary objective of this study was to determine the anti-obesity drug prescribing patterns of US physicians from 2011 to 2016. We also examined the use of pharmacotherapy in obesity treatment and the odds of a patient being prescribed anti-obesity medication.
Methods
Data Source
Data were obtained over six years, 2011 through 2016, from the National Ambulatory Medical Care Survey (NAMCS). At the time of this study, 2011 to 2016 were the most recent years for which data were available. NAMCS is the leading source of nationally representative data on care delivered in physician’s offices. NAMCS is based on a random sample of all US non-federally employed physicians who are primarily engaged in office-based patient care.17 The patient visit is the unit of observation, and for each visit, a patient record form is completed by a physician or staff member. According to the National Center for Health Statistics (NCHS) guidelines, survey years with the same patient record form (survey instrument) can be combined.18 To obtain an adequate sample to assess trends in anti-obesity medication mention, the NAMCS public use data were used to create a pooled analysis of years. In 2015, the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) replaced ICD-9-CM (Ninth Revision) as the diagnosis coding scheme for the US health care system.19 ICD 9 codes were used for the years 2011 to 2015. ICD 10 codes were used for the year 2016. All NAMCS protocols were approved by the NCHS institutional review board. Our study was exempt from institutional review board approval because we used de-identified, publicly available data.
Study Population
Physician office visits were selected if they involved an adult patient aged 18 years or older. For the first analysis, visits were selected if they involved a patient with obesity. To obtain a more reliable estimate of obesity, visits were identified as (one) including an (ICD-9) code of 278.00 (obesity) or 278.01 (morbid obesity) or an (ICD-10) code of E66.9 (obesity, general), E66.01 (obesity, extreme or morbid), and other obesity related codes (E66.0, E66.09, and E66.1) as one of the three listed diagnoses for the visit, or (two) having a diagnosis of chronic obesity as shown of the NAMCS patient record form, or (three) having a BMI calculated for the patient visit of greater than or equal to 30. Two subset populations of visits were created: (1) visit of those with a clear diagnosis of obesity using ICD-9 and ICD-10 codes for obesity or the chronic obesity condition diagnosis on the patient visit form and (2) visits of the overall population of obesity using ICD-9 and ICD-10 codes for obesity, the chronic obesity condition diagnosis on the patient visit form, or those with BMI greater than or equal to 30).
For the second analysis, physician office visits were selected if they involved a patient with at least one anti-obesity drug mention on the survey. Drug codes specific to NAMCS were used to search for orlistat (Xenical), phentermine, diethylpropion, benzphetamine, phendimetrazine, liraglutide, lorcaserin, naltrexone-bupropion (Contrave), phentermine-topiramate (Qsymia). All were approved by the Food and Drug Administration (FDA) for the treatment of obesity, and all were available in the US market during the time-period of this study. Lorcaserin was withdrawn from the US market in 2020, but it was included in this analysis.
Each patient record included demographic data for race, age, and gender. Patient comorbidities (arthritis, cerebrovascular disease/history of stroke, asthma, congestive heart failure, chronic obstructive pulmonary disease, depression, diabetes, hyperlipidemia, and hypertension) were included based on the chronic comorbidity checklist on the patient record form. Information on the patient payer type, provider type (whether the physician was the patient’s primary care physician), and the provider’s region were also collected.
Statistical Analyses
To account for the complex survey design, strata and cluster were included and patient visit weights were applied to all analyses to achieve nationally representative estimates.20 All estimates presented are weighted. Sampling weights were adjusted to account for 6 years of survey data. All analyses were performed using SAS 9.4 “proc survey” suite commands, to account for the complex survey design of the database.
Anti-obesity medication prescription trends (overall, short-term medication mention: treatment less 12 weeks, and long-term medication mention: treatment greater than 12 weeks) for three time periods were examined: time one (2011-2012), time two (2013-2014) and time three (2015-2016). In addition, the total number of physician office visits by adult patients with obesity from 2011 to 2016 were summarized. Characteristics of visits by patients with an obesity diagnosis using ICD-9 and ICD-10 codes for obesity or the chronic obesity condition diagnosis on the patient visit form and visits having an anti-obesity medication mention were described. There were three types of visits identified: (one) patients with obesity and an anti-obesity drug mention, (two) patients with obesity and no anti-obesity drug mention, and (three) patients without obesity and with anti-obesity drug mention. Chi-square tests were used to compare characteristics across each type of visit. To predict the odds of any anti-obesity drug mention for patients with obesity, logistic regression analyses were conducted. One logistic regression analysis was conducted for patients with a clear diagnosis of obesity using ICD-9 and ICD-10 codes for obesity or the chronic obesity condition diagnosis on the patient visit form and another logistic regression analysis was conducted for all patients with obesity using ICD-9 and ICD-10 codes for obesity, the chronic obesity condition diagnosis on the patient visit form, or a BMI > 30. National estimates were computed, frequencies, and logistic regression were performed using the appropriate statistical procedures for the weighted NAMCS database.17
Results
There were an estimated 767 million total adult patient visits to physician offices between 2011 and 2016 (six years). Three time periods were examined: time one (2011–2012) produced a total of 256 million (33.4%) physician office visits, time two (2013–2014) produced a total of 253 million (33.0%) physician office visits, and time three (2015-2016) produced a total of 258 million physician office visits (33.6%).
There was a total of 2.1 million physician office visits with an anti-obesity drug mention over the six-year time period (0.27%). Figure 1 depicts a slight increase in the use of anti-obesity prescriptions between the three time periods: 687,846 total visits with any anti-obesity drug mention during 2011 to 2012, 707,591 total visits with any anti-obesity drug mention during 2013 to 2014, and 712,679 total visits with any anti-obesity drug mention during 2015 to 2016. This represents a trend in percentage of total adult patient visits in NAMCS from 0.26% to 0.28% over the three time periods. Of the visits with any anti-obesity drug mention, there was an upward trend in the use of long-term medications such as orlistat, lorcaserin, naltrexone-bupropion, liraglutide, and phentermine-topiramate (1.03%, 11.84%, and 15.33%)and a downward trend in the use of short-term medications such as phentermine, diethylpropion, benzphetamine, and phendimetrazine (98.97%, 88.16%, 84.66%) (Figure 1).
Figure 1.
Trend in number of anti-obesity medication prescriptions by two-year period from 2011 to 2016
Of the estimated 767 million total physician office visits made during the years 2011-2016 by adult patients, an estimated 197 million (25.7%) were made by adult patients with obesity (Table 1). There was an increase in total physician office visits made by adult patients with obesity between the three time periods: 59.8 million (23.3%) total physician office visits during 2011-2012, 67.1 million (26.5%) total physician office visits during 2013-2014, 70 million (27.1%) total physician office visits during 2015-2016. The majority of the obesity patient visits were identified by body mass index (BMI). However, in addition to BMI, ICD-9 and ICD-10 codes for obesity and the chronic obesity condition diagnosis on the patient visit form identified additional patients with obesity. Thus, multiple variables were used to define obesity to improve the reliability of the estimate.
Table 1.
Total number of visits by adult patients aged 18 years and older with obesity from 2011 to 2016
| Definition of patient with obesity | Total number of adult patients visits with obesity |
Percentage of total adult patient visits (767,213,118) |
|---|---|---|
| ICD-9 code = 2780, 27800, 27801 and ICD-10 code=E660, E661, E662, E668, E669 | 13,318,702 | 1·7 |
| BMI greater than or equal to 30 | 174,680,550 | 23·6 |
| Chronic obesity condition | 66,007,793 | 8·6 |
|
Combined:
ICD-9 code/ICD-10 code or BMI greater than or equal to 30 or Chronic obesity condition |
196,872,870 | 25·7 |
|
Combined:
ICD-9 code/ICD-10 code or Chronic obesity condition |
66,140,592 | 8·6 |
Table 2 shows the characteristics of physician office visits by patients with obesity and visits having an anti-obesity medication mention. To estimate the reason for visit, ICD-9 and ICD-10 codes for obesity and the chronic obesity condition diagnosis on the patient visit form were used. Three types of visits were described: patients with obesity and an anti-obesity drug mention (1.3 million visits in total), patients with obesity and no anti-obesity drug mention (64 million visits total) , and patients without obesity with anti-obesity drug mention (760,470 visits total). Patients without obesity accounted for 36% of the total number of visits associated with anti-obesity drug mentions. Compared to visits by patients with obesity and no anti-obesity drug mention, visits with an anti-obesity drug mention were largely for female patients (P<.0001) and patients between the ages of 18-50 years (P<.0001). Among those with obesity, anti-obesity drug mentions were more likely in the South region with 80.23% of visits by patients with obesity having a drug mention (P<.0001). There was also a higher percentage of anti-obesity drug mentions among those with private insurance (P<.0001).
Table 2.
Characteristics of visits by patients with a clear diagnosis of obesity and visits mentioning anti-obesity medication: 2011 to 2016
| Characteristic | Patients with obesity % | Patients without obesity with anti-obesity drug mention (%) (C) |
P-value, column A versus column B |
P-value, column A versus column C |
|
|---|---|---|---|---|---|
| Anti-obesity drug mention (A) |
No anti-obesity drug mention (B) |
||||
| Total visits | 1,347,646 | 64,792,946 | 760,470 | ||
| Gender | <·0001 | ·50 | |||
| Female | 85·21 | 62·52 | 82·48 | ||
| Male | 14·79 | 37·48 | 17·52 | ||
| Race | ·91 | ·51 | |||
| White | 66·58 | 67·51 | 71·59 | ||
| Non-White | 32·42 | 32·49 | 28·41a | ||
| Age | <·0001 | ·66 | |||
| 18-50 | 74·08 | 37·95 | 75·98a | ||
| 51 or older | 25·92 | 62·05 | 24·02 | ||
| Provider Region | <·0001 | ||||
| Northeast | 8·91 | 26·81 | 0·00a | ||
| Midwest or West | 10·86 | 42·28 | 32·11a | ||
| South | 80·23 | 30·91 | 67·89a | ||
| Health Insurance | <·0001 | <·0001 | |||
| Private | 53·15 | 48·11 | 80·89 | ||
| Public or otherb | 25·84 | 42·43 | 11·23a | ||
| Self-pay | 16·77 | 3·72 | 4·38a | ||
| Missing | 4·24 | 5·74 | 3·50a | ||
| Provider | ·01 | ·68 | |||
| Patient’s primary care physician | 62·20 | 51·37 | 58·39 | ||
| Not | 28·30 | 45·25 | 34·07 | ||
| Missing | 9·50 | 3·38 | 7·54a | ||
| Arthritis | ·001 | <·001 | |||
| No | 90·81 | 78·88 | 78·29 | ||
| Yes | 9·19 | 21·12 | 21·71a | ||
| Respiratory disease c | ·23 | ·75 | |||
| No | 9·78 | 15·90 | 8·52 | ||
| Yes | 90·22 | 84·10 | 91·48a | ||
| Cardiovascular disease d | <·0001 | ·80 | |||
| No | 24·63 | 58·13 | 25·96 | ||
| Yes | 75·37 | 41·87 | 74·04a | ||
| Depression | ·38 | ·40 | |||
| No | 85·17 | 82·13 | 81·15 | ||
| Yes | 14·83 | 17·87 | 18·85a | ||
| Diabetes | <·0001 | ·07 | |||
| No | 93·16 | 69·15 | 85·25 | ||
| Yes | 6·83 | 30·85 | 14·75a | ||
| Hyperlipidemia | ·002 | ·14 | |||
| No | 78·76 | 60·93 | 84·66 | ||
| Yes | 21·24 | 39·07 | 15·34a | ||
| Total comorbidities | <·0001 | <·0001 | |||
| One comorbidity or less | 45·15 | 16·20 | 67·96 | ||
| More than one comorbidity | 54·85 | 83·80 | 32·04a | ||
Less than 30 unweighted records
Other includes workers compensation, no charge, and charity
Includes asthma and chronic obstructive pulmonary disease
Includes hypertension, cerebrovascular disease, congestive heart failure, and ischemic heart failure
For visits involving comorbidities, there was a statistically significant lower percentage of visits by patients with obesity and an anti-obesity drug mention for arthritis, diabetes, and hyperlipidemia. Of the patients with obesity and an anti-obesity drug mention, those with respiratory disease, cardiovascular disease, and at least one comorbidity had a statistically significant higher percentage of visits than those without comorbidities. There were 487,532 patients without a clear diagnosis of obesity with an anti-obesity drug mention. When comparing those patients to patients with obesity and an anti-obesity drug mention, statistically significant differences were identified for health insurance, arthritis, and visits with more than one comorbidity. However, these particular results have a raw cell count of less than 30 and do not meet the conditions to consider these estimates reliable.17
For visits with a clear diagnosis of obesity using ICD-9 and ICD-10 codes for obesity or the chronic obesity condition diagnosis on the patient visit form, 2% had an anti-obesity drug mention. For visits of all patients with obesity using ICD-9 and ICD-10 codes for obesity, the chronic obesity condition diagnosis on the patient visit form, or a BMI > 30, 1% had an anti-obesity drug mention. Table 3 shows the logistic regression results for predicting the odds of an anti-obesity drug mention for patients with a clear diagnosis of obesity. After adjusting for gender, race, patient age, provider region, insurance type, patients primary care physician, and total comorbidities, physician office visits made by those between the ages of 18-50 years (AOR: 0.22; 95% CI: 0.07-0.71) were 78% less likely to have an anti-obesity medication mention when compared to visits made by those ages 51 or older. An anti-obesity drug mention was less likely for those between the ages of 18-50 years. Contrarily, in the South, visits made by those with obesity were 5.31 times more likely to have an anti-obesity drug mention when compared to visits made in the Midwest or West (AOR: 5.31 95% CI: 1.19-23.59).
Table 3.
Logistic regression results for predicting the odds of an anti-obesity drug mention for patients with a clear diagnosis of obesity from 2011 to 2016.
| Predictor | Adjusted Odds Ratio |
95% Confidence Interval |
|---|---|---|
| Gender, female | 1·38 | 0·33-5·74 |
| Race, white | 0·94 | 0·29-3·10 |
| Patient age, 51 or older | 0·22 | 0·07-0·71 |
| Provider region | ||
| Northeast | 1·42 | 0·21-9·54 |
| South | 5·31 | 1·19-23·59 |
| Midwest or West | 1·00 | |
| Insurance type | ||
| Private | 1·00 | |
| Public or othera | 0·19 | 0·03-1.37 |
| Self-pay | 0·73 | 0·11-4·99 |
| Patients primary-care physician | 2·85 | 0·59-13·90 |
| Total comorbidities | ||
| One comorbidity or less | 1·00 | |
| More than one comorbidity | 1·02 | 0·21-4·85 |
Other includes workers compensation, no charge, and charity
Table 4 shows the logistic regression results for predicting the odds of an anti-obesity drug mention for all patients with obesity. After adjusting for gender, race, patient age, provider region, insurance type, patients primary care physician, and total comorbidities, physician office visits made by those between the ages of 18-50 years (AOR: 0.14; 95% CI: 0.05-0.41) were 86% less likely to have an anti-obesity medication mention when compared to visits made by those ages 51 or older. An anti-obesity drug mention was less likely for those between the ages of 18-50 years. This was similar to the results of only those with a clear diagnosis of obesity. Contrarily, in the South, visits made by those with obesity were 3.69 times more likely to have an anti-obesity drug mention when compared to visits made in the Midwest or West (AOR: 3.69 95% CI: 1.17-11.63). These results were also similar to the results of only those with a clear diagnosis of obesity.
Table 4.
Logistic regression results for predicting the odds of an anti-obesity drug mention for the overall population of patients with obesity from 2011 to 2016.
| Predictor | Adjusted Odds Ratio |
95% Confidence Interval |
|---|---|---|
| Gender, female | 2·30 | 0·61-8·63 |
| Race, white | 1·04 | 0·36-2·97 |
| Patient age, 51 or older | 0.14 | 0·02-0·82 |
| Provider region | ||
| Northeast | 1·19 | 0·23-6·12 |
| South | 3·69 | 1·17-11·63 |
| Midwest or West | 1·00 | |
| Insurance type | ||
| Private | 1·00 | |
| Public or othera | 0·16 | 0·24-6·66 |
| Self-pay | 0·72 | 0·12-4·23 |
| Patients primary-care physician | 3·54 | 0·96-13·07 |
| Total comorbidities | 0·14 | 0·02-0·82 |
| One comorbidity or less | 1·00 | |
| More than one comorbidity | 2·04 | 0·58-7·12 |
Other includes workers compensation, no charge, and charity
Discussion
Data from the National Ambulatory Medical Care Survey (NAMCS) over six years from 2011-2016 demonstrated a slight increase in visits with an anti-obesity drug mention from 0.26% to 0.28% during the study period. Despite this slight increase in visits for all individuals with an anti-obesity drug mention, still only one percent of the 197 million office-based visits for patients with obesity mentioned an anti-obesity drug. These findings complement that of Elangovan et al. who found, using electronic health record-derived databases between 2010-2019, an increase in the prevalence of anti-obesity medications yet still significant underutilization of these resources.21 There was an upward trend in the use of long-term medications such as orlistat, lorcaserin, naltrexone-bupropion, liraglutide, and phentermine-topiramate, and a downward trend in the use of short-term medications such as phentermine, diethylpropion, benzphetamine, and phendimetrazine. In 2012, the United States Food and Drug Administration (FDA) approved anti-obesity medications for long term use, so it was expected that we started to see a shift away from short term medication use to long term medication use. Compared to patients with a clear diagnosis of obesity but without an anti-obesity drug mention, visits for patients with a diagnosis of obesity and an anti-obesity drug mention were more likely to be for those who were female, between the ages of 18-50, and from the South. Of visits for patients with an anti-obesity drug mention, 760,470 such visits were for patients without a clear diagnosis of obesity; these patients were less likely to be nonwhite, have arthritis, have diabetes, and have more than one comorbidity. Importantly, estimates for those with an anti-obesity drug mention but without obesity were based on less than 30 unweighted records thus, these estimates cannot be considered reliable because they do not meet the reliability standards proposed by the National Center for Health Statistics.20 For visits with a clear diagnosis of obesity, 2% had an anti-obesity drug mention. For overall visits of patients with obesity, 1% had an anti-obesity drug mention. Nonetheless, for those with a clear diagnosis of obesity, logistic regression results demonstrated that factors such as being between 18-50 years of age and living in the South were independent predictors of having an anti-obesity drug mention. For all patients with obesity, logistic regression results demonstrated the same.
With the current prevalence of obesity of over 42.2%1 and estimates that nearly 50% of adults will have obesity by 2030,2 ensuring that highly effective anti-obesity medications are being sufficiently utilized by providers is key to providing evidence-based care to patients. Following Xia and colleague’s influential paper, other studies have found markedly lower adoption of anti-obesity medications relative to anti-diabetes medications4, modest increases in post-bariatric anti-obesity pharmacotherapy,21 and a 1.3% prescription rate of anti-obesity medications in an eight-center study.22 We further emphasize the low rates of anti-obesity drug utilization and update Xia et al.’s work by providing a more updated analysis of national mention trends incorporating more-recent updates from the AMA6 and the Endocrine Society.7
Indeed, with developments such as the AMA recognizing obesity as a disease and the Endocrine Society recommending weight loss medications for long-term weight reduction, as well as a greater physiological understanding of obesity, we had hypothesized that there would be an increase in the mention of anti-obesity drugs during visits for patients with obesity. Between 2005-2010, Xia and colleagues reported an anti-obesity drug mention rate for patients with obesity of two percent,3 yet our findings employing NAMCS found a decreased mention rate of only one percent in the subsequent six years. This could in part be because our study only employed NAMCS and not both NAMCS and National Hospital Ambulatory Medical Care Survey (NHAMCS) like Xia and colleagues.
One explanation of these findings may lie in implicit and explicit bias against patients with obesity. In an analysis of 93 different stigmas, Pachankis and colleagues found that obesity was rated amongst the highest on the origin dimension of stigma, which is linked to greater perceptions of controllability and higher rates of social rejection.23 Beliefs that obesity is a behavioral concern rather than a physiological one are widely held among providers5,24 and help explain why lifestyle changes are recommended over anti-obesity drugs despite lifestyle changes being largely insufficient on their own.8 Notably, providers also hold related beliefs that individuals with obesity are more weak-willed and are less likely to be compliant with treatment.25-27 Some evidence suggests that these attitudes have worsened over time as one study found that explicit “anti-fat” attitudes increased among obesity specialists between 2001 and 2013.28 Persistence, and potential exacerbation, of stigmatizing mindsets over time may partially explain why prevalence of anti-obesity drug mentions have not increased since Xia’s analysis. Other explanations may also underline the low anti-obesity drug mention rate we found in this study. Namely, providers may be concerned about the safety and efficacy of anti-obesity medications given that the FDA has withdrawn approval for a number of such drugs, including fenfluramine, dexfenfluramine, sibutramine, and, most recently, locaserin.29 These concerns are coupled with inadequate health insurance coverage for anti-obesity drugs: Medicare does not cover these drugs and only nine states have Medicaid programs that do.10 This failure of public health insurance to cover anti-obesity drugs may work to explain our finding of private insurance predicting anti-obesity drug mentions for patients with obesity. Nonetheless, there has been a remarkable 64% increase in coverage through state employee programs over the past decade,9 and most anti-obesity medications are efficacious with negligible or well-tolerated side effects.14,15 These concerns highlight the need for greater advocacy by primary care providers and professional associations alike to increase insurance coverage for anti-obesity medication, as well as educational interventions to deracinate weight-related stigma and allay concerns regarding drug safety. Empowering patients to ask their providers about anti-obesity medications may be another avenue for positive change.
There are several important limitations to this study. For one, the NAMCS permits the listing of only eight drugs, which may have biased the findings of this study towards a lower percentage of visits with mention to anti-obesity drugs. Additionally, as visits were sampled through this survey, it is possible that patients with more severe conditions may be overrepresented, although severity is difficult to estimate because the NAMCS contains little to no laboratory data for analysis. Finally, our diverse means of identifying patients with obesity may have created an overly heterogeneous population for analysis, yet our identification criteria were critical to make a more reliable estimate of obesity, especially given that ICD codes do not always identify such patients. We conducted separate analyses for the different populations of patients with obesity. In addition, this study could have benefited greatly from the inclusion of visits from hospital outpatient departments. However, 2012 to 2017 data from the National Hospital Ambulatory Medical Care Survey were on hold at the time of this study.
Conclusion
Overall, we conclude that, although there have been minor increases in anti-obesity drug mentions between 2011-2016, anti-obesity drug mention rate for patients with obesity remains low at one percent. For those whose visits do mention anti-obesity medication, they are more likely to be 51 years of age or older and reside in the South. Policy and educational interventions should be fully endorsed to address the multifactorial roots of low anti-obesity drug utilization and ultimately ensure all patients with obesity have equitable access to evidence-based care.
Financial Support
This work was supported by the Physician/Scientist Development Award (PSDA) granted by the Executive Committee on Research (ECOR) at MGH (FCS), NIH P30 DK040561 (FCS), L30 DK118710 (FCS).
List of Abbreviations
- 95% CI
95% Confidence Interval
- ACA
Affordable Care Act
- AMA
American Medical Association
- AOR
Adjusted Odds Ratio
- BMI
Body Mass Index
- FDA
United States Food and Drug Administration
- ICD-9-CM
International Classification of Diseases, Ninth Revision, Clinical Modification
- ICD-10-CM
International Classification of Diseases, Tenth Revision, Clinical Modification
- NAMCS
National Ambulatory Medical Care Survey
- NCHS
National Center for Health Statistics
- NHAMCS
National Hospital Ambulatory Medical Care Survey
Footnotes
Conflict of Interest Disclosure
The authors have no conflicts of interest to disclose. This article has not been published or submitted for publication elsewhere.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Adult Obesity Facts. Overweight and Obesity Web site. https://www.cdc.gov/obesity/data/adult.html. Updated 11 February 2021. Accessed 13 February 2021. [Google Scholar]
- 2.Ward ZJ, Bleich SN, Cradock AL, et al. Projected U.S. State-Level Prevalence of Adult Obesity and Severe Obesity. N Engl J Med. 2019;381(25):2440–2450. [DOI] [PubMed] [Google Scholar]
- 3.Xia Y, Kelton CM, Guo JJ, Bian B, Heaton PC. Treatment of obesity: Pharmacotherapy trends in the United States from 1999 to 2010. Obesity (Silver Spring). 2015;23(8):1721–1728. [DOI] [PubMed] [Google Scholar]
- 4.Thomas CE, Mauer EA, Shukla AP, Rathi S, Aronne LJ. Low adoption of weight loss medications: A comparison of prescribing patterns of antiobesity pharmacotherapies and SGLT2s. Obesity (Silver Spring). 2016;24(9):1955–1961. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bessesen DH, Van Gaal LF. Progress and challenges in anti-obesity pharmacotherapy. The Lancet Diabetes & Endocrinology. 2018;6(3):237–248. [DOI] [PubMed] [Google Scholar]
- 6.Recognition of Obesity as a Disease. American Medical Association; May 2013. Resolution: 421(A-13). [Google Scholar]
- 7.Apovian CM, Aronne LJ, Bessesen DH, et al. Pharmacological management of obesity: an endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2015;100(2):342–362. [DOI] [PubMed] [Google Scholar]
- 8.Lee M, Lauren BN, Zhan T, et al. The cost-effectiveness of pharmacotherapy and lifestyle intervention in the treatment of obesity. Obes Sci Pract. 2020;6(2):162–170. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Jannah N, Hild J, Gallagher C, Dietz W. Coverage for Obesity Prevention and Treatment Services: Analysis of Medicaid and State Employee Health Insurance Programs. Obesity (Silver Spring). 2018;26(12):1834–1840. [DOI] [PubMed] [Google Scholar]
- 10.Gomez G, Stanford FC. US health policy and prescription drug coverage of FDA-approved medications for the treatment of obesity. Int J Obes (Lond). 2018;42(3):495–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jensen MD, Ryan DH, Apovian CM, et al. 2013 AHA/ACC/TOS Guideline for the Management of Overweight and Obesity in Adults. Circulation. 2014;129(25 suppl 2):S102–S138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Heymsfield SB, Wadden TA. Mechanisms, Pathophysiology, and Management of Obesity. N Engl J Med. 2017;376(3):254–266. [DOI] [PubMed] [Google Scholar]
- 13.Wilding JPH, Batterham RL, Calanna S, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. N Engl J Med. 2021. [DOI] [PubMed] [Google Scholar]
- 14.LeBlanc ES, Patnode CD, Webber EM, Redmond N, Rushkin M, O'Connor EA. Behavioral and Pharmacotherapy Weight Loss Interventions to Prevent Obesity-Related Morbidity and Mortality in Adults: Updated Evidence Report and Systematic Review for the US Preventive Services Task Force. JAMA. 2018;320(11):1172–1191. [DOI] [PubMed] [Google Scholar]
- 15.Patel DK, Stanford FC. Safety and tolerability of new-generation anti-obesity medications: a narrative review. Postgrad Med. 2018;130(2):173–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Srivastava G, Apovian CM. Current pharmacotherapy for obesity. Nat Rev Endocrinol. 2018;14(1):12–24. [DOI] [PubMed] [Google Scholar]
- 17.Centers for Disease Control and Prevention. 2016 NAMCS micro-data file documentation. https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Dataset_Documentation/NAMCS/doc2016.pdf. Accessed February 21, 2021. [Google Scholar]
- 18.National Center for Health Statistics. Ambulatory Health Care Data. https://www.cdc.gov/nchs/ahcd/index.htm. Accessed February 21, 2021. [Google Scholar]
- 19.Lau DT, Strashny A, Phan K, Blum AL, Burke-Bebee S. Evaluation of Transition From ICD-9-CM to ICD-10-CM Diagnosis Coding System in the National Ambulatory Medical Care Survey. National health statistics reports. 2018(120): 1–10. [PubMed] [Google Scholar]
- 20.National Center for Health Statistics. Survey Methods and Analytic Guidelines. https://www.cdc.gov/nchs/ahcd/survey_methods.htm. Accessed February 21, 2021. [Google Scholar]
- 21.Elangovan A, Shah R, Smith ZL. Pharmacotherapy for Obesity-Trends Using a Population Level National Database. Obes Surg. 2021;31(3): 1105–1112. [DOI] [PubMed] [Google Scholar]
- 22.Saxon DR, Iwamoto SJ, Mettenbrink CJ, et al. Antiobesity Medication Use in 2.2 Million Adults Across Eight Large Health Care Organizations: 2009-2015. Obesity (Silver Spring). 2019;27(12):1975–1981. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pachankis JE, Hatzenbuehler ML, Wang K, et al. The Burden of Stigma on Health and Well-Being: A Taxonomy of Concealment, Course, Disruptiveness, Aesthetics, Origin, and Peril Across 93 Stigmas. Pers Soc Psychol Bull. 2018;44(4):451–474. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Foster GDW, T.A.; Makris AP; Davidson D; Sanderson RS; Allison DB; Kessler A Primary Care Physicians’ Attitudes about Obesity and Its Treatment. Obesity Research. 2003; 11(10): 1168–1177. [DOI] [PubMed] [Google Scholar]
- 25.Phelan SM, Burgess DJ, Yeazel MW, Hellerstedt WL, Griffin JM, van Ryn M. Impact of weight bias and stigma on quality of care and outcomes for patients with obesity. Obes Rev. 2015;16(4):319–326. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Persky S, Eccleston CP. Medical student bias and care recommendations for an obese versus non-obese virtual patient. Int J Obes (Lond). 2011;35(5):728–735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Puhl R, Brownell KD. Bias, discrimination, and obesity. Obes Res. 2001;9(12):788–805. [DOI] [PubMed] [Google Scholar]
- 28.Tomiyama AJ, Finch LE, Belsky AC, et al. Weight bias in 2001 versus 2013: contradictory attitudes among obesity researchers and health professionals. Obesity (Silver Spring). 2015;23(1):46–53. [DOI] [PubMed] [Google Scholar]
- 29.Onakpoya IJ, Heneghan CJ, Aronson JK. Post-marketing withdrawal of anti-obesity medicinal products because of adverse drug reactions: a systematic review. BMC Med. 2016; 14(1):191. [DOI] [PMC free article] [PubMed] [Google Scholar]