Journal Logo

A-49 Free Communication/Poster - Objective Monitoring and Biostatistics Wednesday, May 27, 2015, 7: 30 AM - 12: 30 PM Room: Exhibit Hall F

Agreement Between The Single-item Measure And IPAQ Using Contemporary Physical Activity Recommendations

425 Board #276 May 27, 9

30 AM - 11

00 AM

Zwolinsky, Stephen; McKenna, Jim; Pringle, Andy; Widdop, Paul; Griffiths, Claire

Author Information
Medicine & Science in Sports & Exercise: May 2015 - Volume 47 - Issue 5S - p 112
doi: 10.1249/01.mss.0000476715.14418.f7
  • Free

The aggregation of marginal gains in our knowledge of physical activity measurement is central to the development of effective public health policy and programming. Use of the single-item physical activity measure has become increasingly widespread. However, it was originally developed and validated against the 2004 UK physical activity recommendations, which were updated in 2011. Whilst the changes are modest, they may alter an individual’s activity categorisation compared to the preceding guidelines.

PURPOSE: This study assesses the levels of agreement, sensitivity and specificity for the single-item measure against the IPAQ using contemporary physical activity recommendations.

METHODS: Validation was undertaken using a quota sampling system of 7650 adults. The sensitivity and specificity, and κ statistic were used to assess agreement between the tools for classifying participants as sufficiently active for health (≥150 minutes of physical activity per week) or not, and being inactive (<30 of minutes of physical activity per week) or not. Spearman’s rank correlation coefficients assessed concurrent validity of the single-item measure against IPAQ.

RESULTS: Overall agreement between the single-item measure and IPAQ at categorising sufficiently active participants was weak, sensitivity was 18.7% (κ= 0.13, 95% CI 0.12 to 0.14). For the classification of inactive participants agreement between the two measures was moderate, sensitivity was 74.2% (κ= 0.45, 95% CI 0.43 to 0.47). The single-item measure’s concurrent validity was fair when classifying sufficiently active participants (rs = 0.24). However, a stronger, moderate correlation was found for the assessment of inactive participants (rs = 0.47).

CONCLUSIONS: Compared to IPAQ, the single-item measures capacity to detect participants achieving current physical activity guidelines was weak. However, in this cohort, its ability to identify inactive participants was more encouraging showing acceptable levels of agreement and sensitivity. Therefore, it could be legitimately included in large scale surveys and interventions for adults, where identifying inactive individuals is the goal, or is a requirement for entry to an intervention.

© 2015 American College of Sports Medicine