Article Text

Concise report
Comorbidity profiles among patients with ankylosing spondylitis: a nationwide population-based study
  1. Jiunn-Horng Kang1,
  2. Yi-Hua Chen2,
  3. Herng-Ching Lin3
  1. 1Department of Physical Medicine and Rehabilitation, Taipei Medical University Hospital, Taipei, Taiwan
  2. 2School of Public Health, Taipei Medical University, Taipei, Taiwan
  3. 3School of Health Care Administration, Taipei Medical University, Taipei, Taiwan
  1. Correspondence to Professor Herng-Ching Lin, School of Health Care Administration, College of Medicine, Taipei Medical University, 250 Wu-Hsing Street, Taipei 110, Taiwan; henry11111{at}


Objective Ankylosing spondylitis (AS) is a systemic inflammatory disease that can result in chronic pain and disability. This study aimed to analyse the prevalence and risk of medical comorbidities in patients with AS compared with the general population.

Methods 11 701 patients with AS and 58 505 matching controls were selected for analysis from the National Health Insurance Research Dataset (NHIRD) in Taiwan. The Elixhauser comorbidity index was used for selecting medical comorbidities. Pearson χ2 tests and conditional logistic regression a nalyses were performed to examine the prevalence and risk of comorbidities between these two groups.

Results Patients with AS were at increased risk for multiple systemic comorbidities including cardiovascular, neurological, pulmonary, gastrointestinal, endocrine, haematological and mental illness. The most prevalent comorbidities in patients with AS were hypertension (16.4%), peptic ulcers (13.9%) and headaches (10.2%).

Conclusion The results show that patients with AS have a higher prevalence of multiple comorbidities than the general population in Taiwan. These findings are consistent with previous studies done in Western populations. The results could be useful for both the clinical management of patients with AS and for researching the underlying pathological mechanisms.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Ankylosing spondylitis (AS) is a chronic inflammatory disease that involves mainly the sacroiliac joints, axial structure and enthesis.1 2 The clinical course of AS is highly variable, but many patients with AS experience chronic pain, impaired physical function and a variety of disabilities. Specific diseases have been recognised as being associated with AS, including uveitis, heart conduction problems, aortic and valve disease and IgA nephropathy.2,,4 Increased mortality among patients with AS compared with the general population has been reported in previous studies.5 6 Some comorbidities related to increased mortality from AS—such as cardiovascular, gastrointestinal, renal diseases and violence—have also been reported.7,,9

Conclusions based on current data collected from referral centres may be limited owing to the relatively small sample sizes and selection bias. Moreover, the data regarding medical comorbidities associated with AS in previous studies were mainly from Western populations. To our knowledge, no previous report has analysed comorbid diseases among patients with AS in a Chinese population. AS is a heterogeneous disease and ethnic differences in its presentation have been noted in previous studies.10 Exploring the comorbidity profiles of AS in a Chinese population would therefore be useful for documenting possible ethnic effects. We conducted a nationwide population-based study to clarify the issues above, aiming to explore the prevalence of comorbidities among patients with AS in Taiwan.



We used the National Health Insurance Research Dataset (NHIRD), which includes all medical claims for inpatient and ambulatory care services as well as a registry of beneficiaries including all enrollees. Since the NHIRD consists of de-identified secondary data released to the public for research purposes, this study was exempt from full review by the institutional review board.

Study sample

The study cohort comprised all patients who sought ambulatory care during 2007 and received a diagnosis of AS (ICD-9-CM codes 720 or 720.0; n=30 058). In order to ensure diagnostic validity and patient homogeneity, we selected only patients who had at least two consensus AS diagnoses for the study group (see further details and table S1 in online supplement), excluding those who had a diagnosis of rheumatoid arthritis (ICD-9-CM code 714.0) or systemic lupus erythematosus (ICD-9-CM code 710.0; n=679). Furthermore, we included only those who had been receiving ambulatory treatment for AS since 2005 or before 2005, in order to ensure that the selected patients were not newly diagnosed cases. Ultimately, 11 701 patients were eligible for inclusion in the study group.

We chose our comparison group from the 2007 registry for beneficiaries of the Longitudinal Health Insurance Database by randomly selecting 58 505 enrollees (five for every patient with AS) matched with the study group in terms of gender, age, monthly income and level of urbanisation where the patient resided.

The Elixhauser Comorbidity Index was modified as a basis for selecting medical comorbidities for analysis. In this study, comorbid conditions were counted if the condition either occurred in the inpatient setting or in two or more ambulatory care claims coded during 2007 (see online supplement for more details of the methodology).

Statistical analysis

The SAS statistical package was used to perform the analysis. Conditional logistic regression analyses conditioned on gender, age, monthly income and level of urbanisation of the patient's community were performed to investigate the risk of comorbidities for the two groups. Adjusted ORs are presented with 95% CIs. The α level was set at p<0.05.


Table 1 shows the distribution of characteristics of the sampled patients; more than half (66.8%) of the patients were aged <45 years and 79.1% were men. Table 2 shows the prevalence of comorbidities according to patient group. Compared with patients in the comparison group, patients with AS had a significantly higher prevalence of hypertension, ischaemic heart disease, hyperlipidaemia, congestive heart failure, cardiac arrhythmias, peripheral vascular disorder, headache, migraine, chronic obstructive pulmonary disease (COPD), hypothyroidism, liver disease, peptic ulcers, hepatitis, tuberculosis, deficiency anaemia, depression and psychoses. No significant difference was observed between the two groups in the prevalence of stroke, paralysis, other neurological disorders, asthma, restrictive lung disease, diabetes without complications, diabetes with complications, obesity, weight loss, renal failure, fluid and electrolyte disorders, HIV, alcohol abuse, drug abuse, metastatic cancer and solid tumour.

Table 1

Sociodemographic characteristics of patients with ankylosing spondylitis compared with controls (n=70 206)

Table 2

Prevalence of medical comorbidities among patients with ankylosing spondylitis compared with controls (n=70 206)

Table 3 shows the ORs of comorbidities in patients with AS compared with the comparison group. Conditional regression analyses showed that the ORs of hypertension, ischaemic heart disease, hyperlipidaemia, congestive heart failure, cardiac arrhythmias, COPD and peripheral vascular disorder were 1.87, 2.74, 1.46, 1.42, 1.82, 3.24 and 1.37, respectively, that of the control group. Patients with AS also had higher ORs of mental illness, including depression and psychoses, than the comparison group. Table 3 also presents gender differences in ORs for medical comorbidities between patients with and without AS.

Table 3

ORs (95% CI) for medical comorbidities among patients with ankylosing spondylitis (AS) compared with controls using conditional regression analysis


Our data show that patients with AS are at a higher risk for comorbidities involving multiple systems than the general population. We found the comorbidity profiles were largely similar between Chinese and Western populations with AS. The use of a nationwide database in our study provides sufficient statistical power to detect subtle differences. Nevertheless, it is unrealistic to assume all significant differences reflecting the direct pathological mechanism of AS and comorbidities. Several circumstances could explain the association between AS and the comorbidities we observed including: (1) direct casualty in the pathological mechanism; (2) shared common pathway or biological background; and (3) occult irrelevant variables beside the ‘true’ pathological mechanism. Our study provides fundamental data exploring a wide spectrum of comorbidities among the patients with AS. Careful interpretation and further clarification of our findings are suggested.

In addition, the differences between statistical significance and impact in interpreting study results should be noted for more detailed consideration of our findings. Specifically, even if the relative risk is minor, the association between AS and some comorbidities such as hypertension could still have a great impact due to its relatively high prevalence among a large affected population. On the other hand, despite the higher relative risk, ischaemic heart disease might not possess much clinical significance because the prevalence is relatively low (0.3%). Both prevalence and relative risk should thus be considered in investigating the impacts of comorbidities among patients with AS. Gender difference in comorbidity risk profiles was also notable in our study. Gender differences in clinical course and disease presentation of AS have been previously recognised.2 11 We propose that gender may be also an important factor in comorbidities associated with AS.

Inflammatory processes in autoimmune diseases are considered a factor in the initiation and progression of atherosclerosis.12 13 Studies have shown that atherosclerosis and cardiovascular risks are also increased in populations with AS.8 14 15 Our data show that cardiovascular comorbidities are also increased in a Chinese population with AS. However, we found no increased risk for stroke among patients with AS. Data for exploring the association between stroke and AS remain lacking. Establishing the risk of stroke for a Western population with AS should be considered for further research. Recently, the European League Against Rheumatism has published recommendations for cardiovascular risk management in patients with inflammatory arthritis. Adequate control of disease activity, regular cardiovascular risk assessment, controlling risk factors and pharmacological considerations are all recommended to reduce cardiovascular risk for these patients.16 Cardiovascular diseases can be associated with increased mortality and morbidity in patients with inflammatory arthritis. Adequate cardiovascular risk management can improve care for these patients.

In patients with AS who had received radiotherapy previously, a higher risk for malignancies has been reported, particularly leukaemia.17 However, radiotherapy is no longer applied AS treatment. We did not find that the risk of malignancy increased in patients with AS, consistent with a study performed in patients with AS who did not receive radiotherapy.18 We should note that our findings may be confounded by the relatively young age of our subjects. A considerable proportion of patients with AS in our study had headache. Cervical spondylitis, psychological stress or chronic use of non-steroidal anti-inflammatory drugs may be contributing factors to headache in patients with AS. Interestingly, we also found the risk of migraine was increased in patients with AS. To our knowledge, this is the first report to demonstrate an increased prevalence of migraine in the population with AS. An association between headache and migraine and other autoimmune diseases has been noted previously.19 20 Further study is suggested to confirm our findings and explore possible the pathological mechanisms.

Our study has several limitations. First, the validity of diagnoses cannot be confirmed in our study. However, we analysed the stability of AS diagnoses by our selection criteria and found the result to be acceptable. Second, because our goal is to determine the overall risk for various comorbidities in the population with AS, we did not review the subjects' medications as a confounding factor. Third, cause–effect relationships cannot be determined in our study. A longitudinal study may be needed. Finally, the design of this current study cannot preclude possible investigation bias—that is, if a patient is receiving medical care for a particular disease that increases the likelihood of finding another subclinical disease, or that the patient seeks attention for an otherwise unrelated condition. The prevalence of medical comorbidities among patients in the comparison group may therefore be underestimated.


Increased risk of multiple comorbidities was noted among Chinese patients with AS compared with a general Chinese population in Taiwan. We found novel associations between AS and comorbidity such as migraine. Further studies are needed to explore the mechanisms and develop specific interventions for the comorbidities in patients with AS.


This study is based in part on data from the National Health Insurance Research Database provided by the Bureau of National Health Insurance, Department of Health, Taiwan and managed by the National Health Research Institutes. The interpretations and conclusions contained herein do not represent those of the Bureau of National Health Insurance, Department of Health or the National Health Research Institutes.


Supplementary materials


  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.