The factor structure of the Oral Health Impact Profile [presentation] Presentation uri icon

abstract

  • Objective: The Oral Health Impact Profile (OHIP) is currently the most widely used oral health-related quality of life (OHRQoL) questionnaire, but findings about is structural validity are conflicting. It is not clear how many dimensions are represented in the questionnaire and therefore, it is not known, how many scores sufficiently describe the construct OHRQoL measured with OHIP. How many OHIP scores should be used and what these scores represent is fundamental for the assessment of OHRQoL in patient care and research. It was the aim to investigate OHIP’s factorial structure in an international study of general population subjects and prosthodontics patients. Method: The study is a secondary data analysis of 4736 general population subjects and prosthodontics patients with complete OHIP-49 data (excluding the three denture-specific items) from Croatia, Germany, Hungary, Japan, Slovenia, and Sweden. Polychoric correlations among the 46 OHIP items were submitted to exploratory factor analysis (EFA). Factors were extracted using the principal factor method and rotated using promax, an oblique rotation technique. Result: The unrotated EFA showed a dominating first factor. The first eigenvalue of 26.9 was substantially larger than the second eigenvalue of 2.6, suggesting that data were essentially unidimensional. When four factors were extracted and rotated, the factors correlated between 0.55 and 0.69 with each other. Based on items with salient loadings > 0.5, these factors were named Psychosocial Impact, Orofacial Pain, Oral Function, and Orofacial Appearance. Conclusion: OHIP was unidimensional enough to be sufficiently described with one summary score. When OHRQoL dimensions were assessed using OHIP, a four-factor structure of Psychosocial Impact, Orofacial Pain, Oral Function, and Orofacial Appearance as dimensions appeared as an OHRQoL model consistent with our theoretical understanding of perceived oral health.