Psychometric properties of Modified Yale Food Addiction Scale 2.0 Malay version / Norsuhana Sainal
Background Certain food has been found to demonstrate addictive-like eating in some individuals, giving rise to the concept of food addiction. Its association with obesity has been demonstrated in many studies done previously. Obesity continues to grow as a major health risk globally and recently...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Published: |
2019
|
| Subjects: | |
| Online Access: | http://studentsrepo.um.edu.my/11541/ http://studentsrepo.um.edu.my/11541/4/norsuhana.pdf |
| Summary: | Background Certain food has been found to demonstrate addictive-like eating in some
individuals, giving rise to the concept of food addiction. Its association with obesity has
been demonstrated in many studies done previously. Obesity continues to grow as a major
health risk globally and recently, it has become a global epidemic. To date, there is no
established criteria to assess for food addiction yet as this topic of research is still
relatively in its early stage. The Modified Yale Food Addiction Scale 2.0 (mYFAS 2.0),
the latest and shorter version of the Yale Food Addiction Scale is the only self-report
measurement designed to assess addictive-like eating. This study aims to examine the
psychometric properties of mYFAS 2.0 Malay version. Method A cross-sectional study
conducted in 4 stages was carried out in the University Malaya Medical Center (UMMC).
For stage 1, mYFAS 2.0 was translated into the Malay language. For stage 2, a pilot test
involving 30 medical staff was performed. Stage 3 was conducted on 30 medical students
to assess internal consistency and test-retest reliability. Internal consistency was
determined using Cronbach’s alpha. Test-retest reliability was evaluated at two weeks
after the initial assessment and was determined by Intraclass Correlation (ICC). For stage
4, a validity test was carried out involving 144 overweight and obese participants
attending the Sports Medicine Outpatient Clinic. Convergent validity was done with the
Malay Binge Eating Scale (BES). Construct validity was determined with Confirmatory
Factor Analysis. Result The mYFAS 2.0 Malay version has good reliability with internal
consistency measured with Cronbach’s alpha = 0.92 and test-retest reliability measured
with Intraclass Correlation (ICC) ranged from 0.550 to 0.868. There is a significant strong
positive correlation between mYFAS 2.0 Malay version symptom score with BES Malay
version score (r = 0.545, p < 0.01). The mYFAS 2.0 Malay version is a 9-items one-factor
iv
model and it has failed to replicate the original 13-items one-factor model. The final
model achieved best model fit after removal of 2 items; the goodness of fit index (GFI) =
0.943; the adjusted goodness of fit index (AGFI) = 0.898; the comparative fit index (CFI)
= 0.984; the root means square error of approximation (RMSEA) = 0.05. Another 2 items
had to be removed to achieve ideal values for Composite Reliability (CR) = 0.902 and
Average Variance Extracted (AVE) = 0.507. The 9-items mYFAS 2.0 Malay version has
good sensitivity (92.3%) and specificity (96.9%) as a screening instrument. Conclusion
The 13-items mYFAS 2.0 Malay version is a reliable measure but has limitations in its
construct validity. The modified version of 9-items mYFAS 2.0 Malay version has better
validity and reliability. It also has good sensitivity and specificity as a screening
instrument. |
|---|