Article Text

FRI0750-HPR Construct validity of the international physical activity questionnaire – long form in patients with axial spondyloarthritis
  1. TW Swinnen1,2,3,
  2. S Willemijns2,
  3. W Dankaerts1,
  4. R Westhovens2,3,
  5. K de Vlam2,3
  1. 1Research Group for Musculoskeletal Rehabilitation
  2. 2Skeletal Biology and Engineering Research Center, KU Leuven
  3. 3Division of Rheumatology, University Hospitals Leuven, Leuven, Belgium


Background Physical activity (PA) is increasingly recognized as an important outcome measure in patients with axial spondyloarthritis (axSpA). Indeed, PA interventions in axSpA have shown to improve clinical status and to reduce comorbidity. The International Physical Activity Questionnaire-Long Form (IPAQ-LF) may be a feasible self-reported PA measurement tool, but its validity is unknown in axSpA.

Objectives To establish the convergent construct validity of the IPAQ-LF in axSpA.

Methods Forty patients with axSpA (Male/Female: 24/16; Mean±SD, Age:44.38±11.30 yrs, BMI: 26.27±5.11 kg/m2, disease duration: 11.40±9.50 yrs, disease activity (BASDAI): 3.69±2.59) completed the IPAQ-LF and wore the SenseWear Pro3 Multisensor Armband during five consecutive days (three weekdays and two weekend days). A priori, significant directional associations evidenced by Spearman correlation coefficients of ≥.50 were hypothesized between IPAQ-LF total or its subscales and the armband to conform convergent construct validity (p<.05).

Results IPAQ-LF total PA (Median (IQR): 16.71 (5.91–45.15) MET.hrs/day) was associated with Physical Activity Level (r=.434, p<.01), moderate to (very)vigorous PA (MET≥3, r=.439, p<.01), moderate PA (MET ≥3<6, r=.432, p<.01) and inactive time (MET≤1.8, r=-.382, p<.05) obtained with the armband. Similar, IPAQ-LF moderate PA (Median (IQR): 10.39 (2.41–23.71) MET.hrs/day) was related with PAL (r=.492, p<.01), moderate to (very)vigorous PA (r=.456, p<.01), moderate PA (r=.444, p<.01) and inactive time (r=-.491, p<.05). Also, IPAQ sitting (Median (IQR): 14.91 (10.89–20.80) hrs/day) was correlated to PAL (r=-.461, p<.01), moderate to (very)vigorous PA (r=-.391, p<.05), moderate PA (r=-.386, p<.05) and inactive time (r=.496, p<.01). No relevant nor significant correlations were found for the other IPAQ-LF subscales. Taken together, no hypothesis could be confirmed.

Conclusions Even at a group level, the convergent construct validity of IPAQ-LF in axSpA was not confirmed. Self-reported PA outcomes may provide important contextual information on PA, but perform poor at quantifying PA levels in axSpA. Future research on a feasible self-reported PA measurement tool for these patients is required.

Disclosure of Interest None declared

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.