News Release

Around three quarters of the total obesity costs to the US healthcare system among people living with obesity are accounted for by the 20% highest-cost cases

Prevalence of type 2 diabetes approximately doubled across eight-year study period, regardless of class of obesity

Reports and Proceedings

European Association for the Study of Obesity

Obesity is known to approximately double healthcare costs in the US compared to those with a healthy weight.  A series of five studies presented at this year’s European Congress of Obesity (ECO2022), in Maastricht, Netherlands and online from 4-7 May, now shows that around three quarters of the total obesity costs to the healthcare system among people living with obesity in the US are accounted for by the 20% highest-cost cases.

The studies were conducted by Dr Marc Evans, University Hospital, Llandough, Penarth, Cardiff, UK, and Dr Jonathan Pearson-Stuttard, Head of Health Analytics at the data and analytics company Lane Clark & Peacock LLP, London, UK, and colleagues.

The five studies covered an 8-year period and included 28,583 people living with obesity in the US. The authors conclude that: “Healthcare costs and rates of hospitalisation are greater for individuals in higher compared with lower obesity classes. Our results highlight the relationship between body mass index (BMI) and increasing use of healthcare resources and suggest that obesity progression may contribute significantly to the economic burden of the disease.”

Adults (18 years and over) were identified in the IQVIA Ambulatory Electronic Medical Records database and linked to the IQVIA PharMetrics Plus administrative claims database, both commonly used databases for large scale research purposes. Individuals with a BMI measurement of 30–70 kg/m² during a baseline period (1 January 2007–31 March 2012), and with continuous enrolment in the database for at least one year before their baseline year and 8 years follow-up (up to 2020) were included in the analysis; those who were pregnant or had cancer at the start of the study were excluded. The index date was the date the person had their BMI measured.

Three cohorts were formed based on obesity class: (class I: BMI 30–<35 kg/m²; class II: BMI 35–<40 kg/m²; class III: BMI 40–<70 kg/m²). Obesity-related complications (ORCs) and total per-person direct healthcare costs (inpatient, outpatient and pharmacy costs; measured in 2019 US dollars) were evaluated in the baseline year and in year 8. High-cost cases were defined as the 20% of cases with the highest total costs in year 8; the remaining cases were designated in the low-cost category.

For the group with class I obesity, the 20% of individuals who had highest healthcare costs accounted for 79% of all healthcare costs in this group; for those with class II obesity this figure was 77% and for the group with class III obesity 74%. The authors say: “We found that at least three quarters of the total direct healthcare costs in people with obesity in US clinical practice came from 20% of individuals. People in the high-cost category of obesity had substantially more obesity related-complications than people in the low-cost category, suggesting a clear association between obesity-related complications and economic burden.”

A second analysis of the same study population estimated how many ORCs (obstructive sleep apnoea, heart failure, urinary incontinence, osteoarthritis of the knee, type 2 diabetes, prediabetes, asthma, psoriasis, gastro-oesophageal reflux disease, hypertension, dyslipidaemia, musculoskeletal pain, atherosclerotic cardiovascular disease, and chronic kidney disease/kidney failure) were present in people living with obesity at the start of the study.

The authors found that among 28,583 individuals with obesity, 12,686 (44%) had no ORCs, 7,242 (25%) had one ORC, 4,180 (15%) had two ORCs and 4,475 (16%) had three or more ORCs. The more ORCs an individual had, the higher their healthcare costs at the start of study; and average costs increased for all groups across the 8 years of the study, indicating a worsening of ORCs or development of additional ones in all categories.

In each year, costs increased with the number of ORCs; mean annual per-person costs were highest for individuals with 3 or more ORCs (year 0, USD14,290; year 8, USD20,078) and lowest for those with no ORCs at index (year 0, USD1626; year 8, USD7015). For patients with 1 ORC or 2 ORCs, costs were USD4649 and USD7089 at year 0, and USD9296 and USD 11,738 at year 8, respectively.

A third analysis showed a general trend for increasing cumulative per-patient costs with increasing obesity class for most ORCs, including established cardiovascular disease (CVD; USD126,834; 142,817; and 150,579 for class I, II and III, respectively), heart failure (USD180,140; 188,507; and 243,539) and chronic kidney disease (USD227,702; 284,414; and 298,194).


Costs also increased across the levels of obesity severity for subgroups of individuals with two or more ORCs (USD101,708; 110,709; and 111,633) or three or more ORCs (USD127,646; USD133,378; and USD135,521); and also for individuals with no ORCs (USD39,951; 44,156 and 47,623). For some ORCs (osteoarthritis, atherosclerotic CVD and type 2 diabetes), there were no consistent differences between obesity classes.


A fourth analysis looked at the prevalence of certain ORCs across the obesity classes finding that in all three classes of obesity, type 2 diabetes was approximately twice as common at the end of the 8-year study period as it was at the start. In the group with class I obesity, the proportion of people with type 2 diabetes increased from 7% to 13%; in the group with class II obesity, type 2 diabetes prevalence increased from 11% to 23%; and in the group with class III, type 2 diabetes prevalence increased from 16% to 31%.

From year one to year 8, prevalence of chronic kidney disease increased 3.3 fold (with class I obesity), 6.7 fold (class II obesity) and 5.5 fold (class III obesity), while obstructive sleep apnoea increased 2.6 times (class I obesity), by 80% (class II obesity), and by 60% (class III obesity).

The fifth and final analysis studied differences in hospitalisations and hospital costs between the obesity classes and how these costs progressed. Mean healthcare costs increased across the 8-year study period for all three classes of obesity, but more so for those with class III obesity (36%) and class II obesity (41%) compared with class I obesity (24%). Year 8 costs were 27% higher for obesity class II (US11,809) compared with class I (USD9,291) and were 34% higher for class III (USD12,472) compared with class I (USD9,291).

The proportion of people hospitalised each year was slightly higher as severity of obesity increased, however, the number of hospitalisations among those individuals hospitalised was the same in all three classes obesity: 1.3 hospitalisations per year, and yearly unadjusted hospitalisation costs per person were generally similar for the three obesity classes (see figure in poster).

“These findings provide clear evidence that people living with obesity face a broad range of comorbidities which tend to increase over time and with obesity severity with a substantial impact on healthcare resource usage and cost implications for healthcare systems,” explains Dr Pearson-Stuttard.

Dr Evans adds: “The implications are that effective weight management to prevent obesity or its progression is likely therefore both to reduce morbidity and reduce cost pressures on healthcare systems.”

Dr Marc Evans, University Hospital, Llandough, Penarth, Cardiff, UK. Contact via e-mail. E)

Dr Jonathan Pearson-Stuttard, Head of Health Analytics at the data and analytics company Lane Clark & Peacock LLP, London, UK. Contact via e-mail. E)

Alternative contact in ECO Press Centre Tony Kirby T) +44 7834 385827 E)

This press release is based on five studies at the European Congress of Obesity (ECO 2022). The material has been peer reviewed by the congress selection committee. There is no full paper at this stage.



Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.