Use of oral health impact profile-14 (OHIP-14) in different contexts. what is being measured?

• Autor(a) principal: Campos, Lucas Arrais [UNESP]
• Data de Publicação: 2021
• Outros Autores: Peltomäki, Timo, Marôco, João, Campos, Juliana Alvares Duarte Bonini [UNESP]
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://dx.doi.org/10.3390/ijerph182413412; http://hdl.handle.net/11449/233910
• topic: Oral health; Psychometrics; Validation study

The Oral Health Impact Profile-14 (OHIP-14) has been used to assess the impact that oral health problems can have on an individual’s life. Different theoretical models were proposed to evaluate the results. The aims of this study were to evaluate the fit of different factorial models of the OHIP-14 to non-dental patients (NDP) and dental patients (DP) samples from Brazil and Finland and to estimate the differential functioning of the items in the OHIP-14 between the samples. Two studies were conducted, one in Brazil and the other in Finland, composed of five samples (Brazil—Sample 1 (S1): DP, n = 434, age: 25.3 [SD = 6.3] years; S2: NDP, n = 1486, age: 24.7 [SD = 5.6] years; S3: DP, n = 439, age: 29.0 [SD = 6.7] years; Finland—S4: DP, n = 482, age: 26.3 [SD = 5.4] years; S5: NDP, n =·2425, age: 26.7 [DP = 5.5] years). The fit of the OHIP-14 models to the data was estimated using a confirmatory strategy (validity based on the internal structure). Differential item functioning (DIF) between samples was estimated. For NDP from both countries, the response pattern severely violated the normality assumption in six items of the OHIP-14, indicating that the instrument does not fit for these samples. For DP, the model with the best fit was unifactorial, which deals with the estimation of the general impact of oral health on an individual’s life, without addressing specific dimensions. Configural invariance was refuted between samples. DIF indicated that the characteristic of the sample (NDP and DP) in both countries interfered in the response given to the items, with the response level being more adequate for the latent PD trait. The validity of data related to the impact of oral health problems on an individual’s life was confirmed through a unifactorial model. OHIP-14 works properly in DP samples and was limited in NDP samples, being also influenced by cultural context and age.