Background/Purpose:

Depression is a known comorbidity of ankylosing spondylitis (AS) with over one-third of AS patients affected and increased depression-rate ratio compared to the general population. Predictors of depression in AS have not been comprehensively reported.  The purpose of this study is to examine the baseline demographic and clinical predictors of depression in addition to depression severity over time in AS.  The baseline predictors of depression in AS were also studied over time on depression severity. 

Methods:

Using the Center for Epidemiological Studies Depression (CES-D) Scale for depression severity, 622 patients who met modified New York criteria for AS and were followed at least two years were studied from the Prospective Study of Outcomes in Ankylosing Spondylitis cohort. Univariable/multivariable longitudinal analyses using mixed effect negative binomial regression models were conducted to assess the clinical/demographic features associated with CES-D score accounting for correlation of repeated CES-D measures over time. 

Results:

The cohort was 73% male and 81% of patients were white, having a mean age 42.4 years (SD= 13.8).  The mean disease duration at baseline was 17.5 years (SD±13.1).  The median CES-D score at baseline was 9 (IQR= [5, 17]). In the univariable longitudinal model,  independent factors positively associated with a higher CES-D score included: smoking , have multiple comorbidities, Hispanic ethnicity,  greater reported pain, disease activity (by the Bath Ankylosing Spondylitis Disease Activity Index), functional impairment (by the Bath Ankylosing Spondylitis Functional Index ), elevated CRP, NSAID use, DMARD use, anti-depressants use, narcotic use, and self-reported depression.  Inversely associated factors included: marital status (married), exercise and TNF –α inhibitor use.  Baseline radiographic severity did not have any statistically associations with CES-D scores. In multivariable analysis (Table 1), positive correlations with current smoking, functional impairment, disease activity, greater pain perception, usage of DMARDS and narcotic pain analgesics and self-reported depression. Inverse associations were found with being married, exercising > 120 minutes per week, and, curiously, greater baseline radiographic severity (as measured by the Bath Ankylosing Spondylitis Radiographic Index).

Conclusion:

This study identifies demographic and clinical factors that predict longitudinal depression severity and demonstrates potentially modifiable targets to treat depression in AS patients.

Table 1. Multivariable Associations of Selected Variables on Longitudinal CES-D Score