Abstract
Objective Obesity is an important comorbidity in axial spondyloarthritis (axSpA); however, the prevalence of obesity in axSpA compared with the general population and associated socioeconomic factors remain unknown.
Methods This repeated cross-sectional study compared BMI (kg/m2) groups of patients with axSpA to the Swiss population at 3 timepoints (2007, 2012, and 2017). BMI categories were compared by different age, sex, and education categories using the chi-square goodness of fit test. Unpaired, 1-sided t tests were used to compare the BMI in patients with axSpA between the different timepoints.
Results Compared to the general population, patients with axSpA had a higher proportion of overweight and obesity: 18.9% of all patients with axSpA were obese, compared to 11.3% of the Swiss population in 2017. Comparison of BMI groups within sex, age, and education groups consistently showed a trend toward higher rates of overweight and obesity in axSpA. Further, patients with axSpA, especially females, showed a trend of increasing BMI over the studied 10 years. At every time point, overweight and obese patients were significantly more likely to be male, were older, and had higher disease activity than patients with normal weight. Obesity was associated with a deprived socioeconomic status as indicated by a higher proportion of patients with manual labor jobs and lower levels of education.
Conclusion The prevalence of obesity was significantly higher among patients with axSpA compared to the Swiss population, with socially disadvantaged individuals being the most affected. There is an urgent need to initiate prevention strategies for obesity in patients with axSpA.
Axial spondyloarthritis (axSpA) is a chronic inflammatory rheumatic disease that affects mainly the axial skeleton, in addition to the peripheral entheses (enthesitis) and joints (arthritis).1 Besides smoking,2,3 obesity represents the most important potentially modifiable lifestyle factor in axSpA. Obesity does not only contribute to the elevated cardiovascular-related mortality in axSpA,4-7 but is also associated with diminished response to treatment8 and higher disease activity.9
To date, there is only 1 study that has examined the difference between the prevalence of obesity in patients with axSpA and the general population.10 It was found that obesity has a higher prevalence in axSpA compared to the general population; however, compared with other cohorts, the sample of patients with axSpA in this study was rather small (N = 461), and further characterization of the BMI (calculated as weight in kilograms divided by height in meters squared) difference between patients and the general population (eg, different age groups, sex) was lacking.
Research into socioeconomic factors associated with obesity in axSpA is also lacking, although this knowledge is important for the development of targeted public health interventions.
The aim of our study was, therefore, to examine the prevalence of BMI categories among patients with axSpA compared to the general population in Switzerland at 3 points in time, stratifying by age group and sex, and to describe trends in BMI over time. Additionally, this study aimed to explore socioeconomic factors that may be associated with obesity in axSpA.
METHODS
Study design. This is a repeated cross-sectional study comparing the distribution of different BMI groups between patients with axSpA and the general Swiss population in 2007, 2012, and 2017.
Data source and patients. Patients with axSpA—diagnosed by the treating rheumatologist—were included from the Swiss Clinical Quality Management in Rheumatic Diseases (SCQM) cohort.11 Patients in the SCQM cohort are examined at enrollment and at least annually thereafter; visits follow the recommendations of the Assessment of SpondyloArthritis international Society (ASAS). SCQM data were requested with a research protocol that had to be approved by a scientific committee and by the SCQM foundation board.
For the comparison with the normal population, data from a representative collective of the Swiss population (N = 22,134) were used, which were evaluated by the Federal Statistical Office (FSO) by means of routine surveys from 2007, 2012, and 2017. The aggregated data are freely available.12
Inclusion criteria. Patients with axSpA from the SCQM with available body weight and height data, which are part of the physical examination, were retrospectively selected at each specific study time point if clinical visits were available in 2007, 2012, and 2017 to allow comparison with the FSO surveys.
Ethical consideration. The study was approved by the Ethics Committee of the Canton of Zurich (KEK-ZH-Nr. 2014-0439). All patients in the SCQM provide written informed consent before inclusion into SCQM.
Methodology. Patients were grouped into 4 BMI categories according to the World Health Organization (WHO) criteria13: underweight (BMI < 18.5), normal weight (BMI ≥ 18.5 to < 25), overweight (BMI ≥ 25 to < 30), and obese (BMI ≥ 30). Each time point was assessed independently of the previous one. Additional patient characteristics available in the SCQM include: sex, fulfillment of the modified New York (mNY) classification criteria (positive/negative), age, HLA-B27 status (positive/negative), presence of inflammatory back pain at diagnosis (positive/negative), family history of axSpA (positive/negative), history of dactylitis (positive/negative), history of uveitis (positive/negative), history of psoriasis (positive/negative), history of heel enthesitis (positive/negative), smoking status (ever smoker/nonsmoker/current smoker), Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), physician and patient global assessment, 12-Item Short Form Health Survey (SF-12; mental and physical component scores [MCS and PCS, respectively] completed separately), levels of C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), Bath Ankylosing Spondylitis Functional Index (BASFI), Bath Ankylosing Spondylitis Metrology Index (BASMI), and education status (mandatory, secondary, or tertiary). Patients working in jobs associated with transportation, manufacturing, or agriculture were classified as blue-collar workers, whereas patients working in less physically demanding jobs (office worker, homemaker, teacher, etc.) were classified as white-collar workers.
From the general population, the FSO report provides aggregated data on BMI categories for different age groups and sex, as well as across different educational levels.
Statistical analysis. Patient characteristics were compared using Fisher exact test for categorical variables and the Kruskal-Wallis test for continuous variables. BMI categories between the general Swiss population and patients with axSpA were compared quantitatively for each age, sex, and education category. The comparison was performed with the chi-square goodness of fit test. Wilcoxon rank-sum tests were used to compare BMI—as a continuous variable—in patients with axSpA between the different timepoints. Data analysis was performed with R (version 4.1.1).
RESULTS
Comparing BMI between patients with axSpA and the Swiss population. Figure 1 shows the difference between the proportion of overweight and obese individuals in patients with axSpA and the Swiss population across the 3 timepoints. Patients with axSpA had a higher proportion of obesity at all 3 timepoints, with 18.9% of all axSpA patients being obese compared to 11.3% of the Swiss population in 2017. The Supplementary Figure (available with the online version of this article) shows the difference in all BMI categories between patients with axSpA and the Swiss population in 2017. BMI (as a continuous variable) between the different timepoints in patients with axSpA shows a small increasing trend (Supplementary Table S1, available with the online version of this article): in 2007 the mean BMI of obese patients was 33.0, and in 2017 it was 34.0. Figure 2A shows the difference in the distribution of BMI categories between the sexes in axSpA and the Swiss population in 2017. Both in patients with axSpA and in the Swiss population, male patients were more often overweight and obese than female patients, and the axSpA BMI composition differed significantly in both groups from that of the general Swiss population, with a greater proportion of overweight and obese patients. Regarding the change in BMI between timepoints in axSpA, the above-mentioned increasing trend between 2007 and 2017 is observed in female but not in male patients (Supplementary Table S1).
Distribution of overweight and obesity between the Swiss population and patients with axSpA in the SCQM registry in the years 2007, 2012, and 2017. axSpA: axial spondyloarthritis; SCQM: Swiss Clinical Quality Management in Rheumatic Diseases.
Distribution of different BMI categories between the Swiss population and patients with axSpA in the SCQM registry between (A) female and male individuals, (B) different age groups, and (C) education categories. P values were calculated with chi-square goodness of fit test and percentage values are rounded. axSpA: axial spondyloarthritis; SCQM: Swiss Clinical Quality Management in Rheumatic Diseases.
The distribution of weight categories by age between patients with axSpA and the Swiss population is shown in Figure 2B. Overweight and obesity increase with age in both groups with a peak in the 45- to 54-year-old age group in axSpA and in the 65- to 74-year-old age group in the general Swiss population. In patients with axSpA, again, the proportion of obese and overweight patients is higher compared to the Swiss population regardless of the age category, although not significantly so in the 2 oldest groups.
Socioeconomic factors. Education status and job type, as well as the proportion of patients who have ever smoked, are shown in Table 1 for patients with axSpA in 2017. Obese patients were characterized by a lower proportion of higher education (percentage in normal weight and obese groups: 33.2% and 16%, respectively) and higher proportion of blue-collar jobs (percentage in normal weight and obese groups: 29.2% and 42%, respectively). Based on these results, which also apply to earlier timepoints (Supplementary Table S2, available with the online version of this article), socioeconomic status seems to be lower in obese patients compared to patients with axSpA in the normal weight group. Over time, the proportion of obese patients who have experienced higher education appears to be slightly increasing and the proportion of blue-collar workers is slightly decreasing. The proportions of BMI categories for 2017 differed between secondary and tertiary education for axSpA and Swiss population groups, but not for individuals with mandatory education (Figure 2C). Individuals with mandatory education had the highest proportions of obesity and overweight in patients with axSpA as well as in the general population.
Socioeconomic differences between different BMI categories in patients with axSpA from 2017.
Characteristics of patients with axSpA. Baseline characteristics of patients with axSpA assessed in 2017 (N = 2175), stratified by BMI category, are shown in Table 2. Compared to patients with normal weight, overweight and obese patients were significantly more likely to be male (percentage male in normal, overweight and obese patients: 50.4%, 66% and 59.8%), were older (mean [SD] years in normal weight, overweight, and obese groups: 43.6 [13.4], 46.2 [12.5], and 49 [11.5]), and had longer symptom duration (mean [SD] years with symptoms in normal weight, overweight, and obese groups: 14.5 [11.5], 16.1 [11.3], and 16.1 [11.5]). These findings were consistent at all 3 timepoints investigated (Supplementary Table S3, available with the online version of this article).
Baseline characteristics including individual ASAS criteria of patients with axSpA from 2017 in different BMI categories.
Obese patients had higher rates of arthritis (percentage in normal weight and obese groups: 56.1% and 62.5%, respectively), heel enthesitis (percentage in normal weight and obese groups: 68.2% and 75.7%), and elevated CRP (percentage in normal weight and obese groups: 59.8% and 75.7%) compared with patients with normal weight. Further, the rate of definite radiographic changes in the sacroiliac joints was higher in obese patients (percentage positivity of mNY classification criteria in normal weight and obese groups: 74% and 83.5%; Table 2).
Table 3 and Supplementary Table S4 (available with the online version of this article) show various disease activity variables in patients from 2007 and 2017. Overall, scores—including different modalities of the disease—showed significantly higher value in obese patients compared to patients with normal weight. BASFI indicates disease-associated functional disability (mean [SD] BASFI in normal weight and obese groups: 2.0 [2.1] and 3.6 [2.6], respectively) and BASMI indicates mobility limitation (mean [SD] BASMI in normal weight and obese groups: 1.6 [1.7] and 2.7 [2.0], respectively). The BASDAI includes several modalities, all of which are associated with axSpA: fatigue, spinal pain, joint pain/swelling, areas of localized tenderness, morning stiffness duration, and morning stiffness severity. Here, too, significantly increased scores were found in obese patients (mean [SD] BASDAI in normal weight and obese groups: 3.5 [2.2] and 4.5 [2.3], respectively). Objectively measurable CRP (mean [SD] CRP in mg/L in normal weight and obese groups: 4.9 [8.7] and 7.4 [11.2], respectively) and ESR (mean [SD] ESR in mm/h in normal weight and obese groups: 11.5 [12.8] and 14.4 [13.4], respectively) were also significantly higher in obese patients compared to patients with normal weight. Of all the disease activity scores, only the SF-12 MCS, which assesses mental health, was not different between normal weight and obese groups (mean [SD] SF-12 MCS in normal weight and obese groups: 46.8 [11.1] and 45.3 [12.1], respectively).
Disease activity variables between different BMI categories in patients with axSpA in 2017.
DISCUSSION
Our study shows for the first time to our knowledge a comparison of BMI between patients with axSpA and the overall population across 3 timepoints, including comparisons within sex, age, and education groups. Patients with axSpA were obese and overweight in a higher proportion at all timepoints, and there was an association with higher disease activity in both groups, and obese patients belonging to a socially disadvantaged group.
From a pathogenetic perspective, axSpA is a disease caused by genetic, environmental, and potentially unknown factors. Biomechanical loading is often highlighted as one of the most important environmental factors in axSpA.14 In the case of obesity, which is associated with higher stress on ligaments and joints, this could be an even more important factor for disease initiation and perpetuation. In psoriatic arthritis (PsA), a disease belonging to the family of spondyloarthritides, obesity is an established risk factor for developing the disease and is associated with higher disease activity.15 In line with this, a previous study found significant positive effects of a short-term weight loss treatment with a very low energy diet on disease activity—independent of medical treatment—in patients with PsA.16
Despite the biomechanical stress hypothesis, the direction of causality between obesity and axSpA is unclear. It can be argued that patients with high disease activity are less physically active and therefore gain weight. Additionally, the frequent use of tumor necrosis factor inhibitor therapy in axSpA can result in an increase in body weight and BMI as a possible side effect.17 Obesity is also an established risk factor for osteoarthritis18 and mechanically induced enthesopathy,19,20 which can also accompany axSpA and may mimic inflammation-related symptoms. However, given the overall negative effects of obesity, including increased cardiovascular risk, action on obesity in axSpA is warranted.
In a European comparison, Switzerland has one of the lowest obesity rates at 11.3%.12 Established public health measures to combat obesity in Switzerland include: the Swiss Nutrition Policy 2017-2024,21 Swiss Food-based Dietary Guidelines,22 Switzerland Pledge,23 Ordonnance du DFI sur les huiles et graisses comestibles et leurs dérivés,24 Physical Activity Strategy for the WHO European Region 2016-2025,25 Health 2030,26 and the National Strategy for the Prevention of Noncommunicable Diseases.27 Our data show that obese patients with axSpA mostly have similar socioeconomic risk factors as the general population. However, obesity is more common in axSpA in every assessed category, warranting additional public health interventions. Primary preventive interventions aim to prevent the development of the disease, although in the case of axSpA other factors that characterize a risk population that would allow screening programs are lacking. Secondary and tertiary interventions aim to tackle obesity in patients with axSpA. The Swiss Ankylosing Spondylitis Association promotes measures such as sport and exercise, as well as certain nutrition, but primarily not with the aim of weight reduction.28 The Swiss League Against Rheumatism as well as the Swiss Society of Rheumatology have no recommendations or measures for weight reduction in axSpA. Based on our findings, there is an urgent and imperative need to initiate and expand programs.
In everyday clinical practice, the weight of every patient with axSpA should be determined, and at the same time other cardiovascular risk factors should be sought. From a therapeutic point of view, the management of obesity in patients has changed considerably in recent years, so that in addition to lifestyle changes (eg, increasing physical activity, dietary changes), drug interventions for weight reduction are increasingly available. Most recently, drug treatment with tirzepatide, a gastric inhibitory polypeptide (GIP) analog which activates both the glucagon-like peptide-1 (GLP-1) and GIP receptors, showed a weight reduction of up to 20% after 72 weeks.29 In particular, an effective drug therapy option seems important, as obese patients with axSpA can be massively restricted in their capacity to perform or increase physical exercise due to elevated disease activity. Future studies should directly assess the effect of drug-induced weight reduction in obese patients with axSpA (eg, in terms of reducing disease activity).
Our study has several advantages. The SCQM axSpA cohort is a large mononational axSpA registry,11 founded in 2005. This enabled comparisons to be made between the Swiss population and patients with axSpA over 10 years with the inclusion of a sufficient number of patients. In addition to the standardized collection of data describing patient characteristics and disease activity, the SCQM collected socioeconomic information from patients at an early stage. Further, the data collected by the FSO can be considered official, reliable data.
Our study also has some limitations. Although we included data from 2009-2017 and no more recent data, it can plausibly be hypothesized that it is still relevant because a clear and persisting trend has been observed over the years with a small increase in BMI in female patients with axSpA. Moreover, since no specific measures have been initiated to address obesity in axSpA, it can be assumed that it will not disappear of its own accord. The FSO provides data from the Swiss general population, although it cannot be ruled out that these data also include patients with axSpA; however, it can be assumed that (1) due to the assumed prevalence of axSpA, these are only a few cases and (2) that if these patients were excluded, the differences would be even greater. Another limitation of our study is the fact that axSpA recommendations on diagnosis and treatment, number of treatment options, and other possible influencing factors changed over the included timepoints. However, most of the changes were aimed at better treatment of axSpA, and yet we see a growing trend toward obesity.
Extrapolation of the results to other regions of the world is only possible to a limited extent. However, a study from another country showed also that obesity is more common in axSpA than in the general population.10
Future studies should both review our results in other cohorts and study interventions to reduce the prevalence and associated burden of obesity in axSpA. The latter includes both measures at the level of public health in the sense of prevention approaches and measures directed at individuals in everyday clinical practice. Further, the causality of the association between obesity and axSpA and factors leading to higher disease activity in obese patients with axSpA should be investigated in a broader context with a translational approach. Should obesity turn out to be primarily a risk factor for the development of axSpA, the field would be open for further studies to promote primary prevention and to find biomarkers which further enhance the risk of developing axSpA in obese patients.
Obesity is a persistent and unresolved problem in patients with axSpA that needs to be addressed in future treatment recommendations and by initiating interventions that target socially disadvantaged patients in particular. In addition, future research should review our findings in other regions of the world and further investigate the direction of causality.
Footnotes
The Swiss Clinical Quality Management in Rheumatic Diseases (SCQM) foundation is supported by the Swiss Society of Rheumatology and by AbbVie, AstraZeneca, iQone, Janssen, Lilly, MSD, Novartis, Pfizer, Samsung Bioepis, Sandoz, and Biogen.
AC received honoraria for lectures or presentations from AbbVie, MSD, and Novartis. MJN has received consulting and/or speaking fees from AbbVie, Eli Lilly, Janssen, Novartis, and Pfizer, and a research grant from Novartis. OD reports receiving consulting fees from AbbVie. RM reports payment/honoraria from AbbVie, Eli Lilly, Gilead Sciences, and Pfizer. The remaining authors declare no conflicts of interest relevant to this article.
- Accepted for publication June 27, 2023.
- Copyright © 2023 by the Journal of Rheumatology
