Background
Quality of Life (QoL) is defined as individuals’ perception of their position in life with regard to the culture and value systems where they live, and in relation to their goals, expectations, standards and concerns [
1]. It is not a simple concept; instead, it is a broad-ranging concept influenced in a complex way by the individuals’ physical health, psychological state, level of dependence, social relationships, and their relationships to salient features of the surrounding environment [
1‐
3]. Primarily and directly oral health can affect the general health resulting in positive or negative impact on the individuals’ QoL, that is, they can eat, talk and conduct daily social activities without discomfort or embarrassment [
1,
4]. When QoL is linked to health and disease, it is referred to as Health-Related Quality of Life (HRQoL). This term is useful to differentiate it from other QoL aspects. HRQoL is a multidimensional concept including physical, mental/emotional, and social functioning domains. It refers to the impact of health status on the individual’s quality of life [
5‐
7].
A part of HRQoL is Oral Health-Related Quality of Life (OHRQoL), which is a multidimensional concept referring to patient’s perceptions when eating, sleeping, or engaging in social activities with respect to oral health [
8‐
10]. The Oral Health Impact Profile (OHIP) is the most widely used instrument to measure OHRQoL [
8]. Based on Locker’s conceptual model of oral health [
11], Slade and Spencer developed and validated the original 49-item OHIP (OHIP49) [
12]. It has been validated culturally and linguistically; its psychometric properties have been tested and it has been used in both cross-sectional and longitudinal studies [
13‐
16]. The OHIP is usually used to evaluate the impact of oral disease on quality of life and to measure the outcomes of clinical interventions [
17‐
19]. Despite the fact that it has been widely used and accepted, in addition to being comprehensive and precise when measuring OHRQoL, it is long and hence it is time-consuming, more prone to missing data, inconvenient, it costs more, and may causes problems especially to elderly respondents [
20]. For the afore-mentioned reasons, several short forms of OHIP were created [
20‐
23].
The shortest version of this instrument is the 5-item OHIP which, as its name implies, consists of 5 questions representing the four suggested dimensions: Oral Function, Orofacial Pain, Orofacial Appearance and Psychosocial Impact [
24,
25]. It has only 10% of the original instrument items but can capture almost 90% of the information. This makes it a more accepted and attractive instrument for efficient OHRQoL assessment. It has been tested and validated in a Swedish general population [
14], Japanese prosthodontic patients [
26], Dutch temporomandibular dysfunction (TMD) patients [
27], German general population and TMD patient [
23], and a US adult general population [
28]. So far, there has not been any information about the Arabic version of this ultra-short instrument.
The aim of this study, therefore, was to translate an Arabic version of the 5-item OHIP and to test its psychometric properties among an Arabic-speaking population.
Discussion
This is the first study assessing the psychometric properties of the OHIP5 in its Arabic version. Findings indicate satisfactory validity of the instrument suggesting its use as one total score to assess the OHRQoL in Arabic-speaking population. Our findings are well comparable to those from other studies on the 5-item OHIP. The dimensionality test of OHIP5-Ar revealed a one-dimensional instrument containing five items, each referring to at least one of the four suggested dimensions of the longer versions. Similar results were found in the English [
28] and German [
23] versions of the OHIP5. The internal consistency of the OHIP5-Ar was 0.78 indicating satisfactory construct validity of the instrument. This value is higher than that of the Dutch version [
27] which was 0.67, lower than that of the Japanese version [
26] which was 0.81, but it is close to the German [
23], Swedish [
14], and English [
28] versions (0.76, 0.77, and 0.75, respectively). The average inter-item correlation was 0.42 which is similar to the Swedish version (0.41) [
14]. The ICC of 0.88 was somewhat higher than that of the Dutch [
27], German [
23], and Japanese [
26] versions. But differences were not considered relevant.
Correlation between summary score of the OHIP5-Ar and global perceived oral health question was close to that resulted from the latent variable analysis, indicating fit of the model for the target population. Similarly, the correlations between the summary score of the OHIP5-Ar and global general health question and that resulted from SEM analysis were close to each other. This confirms that the individual items of the OHIP5-Ar are able to measure the construct OHRQoL. The correlation between the OHIP5-Ar and global oral health is similar to those observed in the German (
r = 0.50) [
23] and Japanese (
r = 0.48) [
26] versions and higher than that found in the English version (
r = 0.30) [
28]. The SEM-based correlation, however, is close to that of the English version (
r = 0.46) [
28]. The OHIP5-Ar could clearly and significantly differentiate between different groups that have potential differences. It is well-known in the dental literature [
56] that patients with more remaining teeth (≥20 teeth) have better oral function than those with fewer remaining teeth (< 20 teeth), and this will be primarily mirrored in their OHRQoL. In our study, patients with ≥20 teeth had lower values of the OHIP5-Ar score (better oral health) than those who had < 20 teeth. Esthetic appearance can also affect the individual’s performance and thus the quality of life [
61‐
63]. The OHIP5-Ar includes a question (Q3) related to esthetic which represents the orofacial appearance domain in the longer versions. Accordingly, patients who presented with missing posterior teeth reported better oral health than patients with missing anterior teeth. Similarly, patients with unhealthy periodontium reported worse OHRQoL than those did with healthy periodontium. Pain, bleeding, tooth mobility, calculus deposition, and gum recession were some of the clinical features of the diseased periodontium which reflected negatively on the patient’s response. Increasing age is considered one of the contributing factors for poor oral health. This was obvious with our sample where patients ≤40 years had lower values (positive response) of the OHIP5-Ar than those with age > 40 years. Different values of oral health were observed among subjects with different educational levels. This was not surprising as an educated subject is likely to know more about oral health and care.
Sensitivity to change of the OHIP5-Ar was clearly confirmed in our sample. The effect of dental treatments on patients’ response was moderate to large. This effect is similar to that obtained in the German [
23] and Japanese [
26] versions. The relevant sample in our study received different types of dental treatments which is in contrast with the previous versions of OHIP5 where patients received the same type of dental treatment (provision of new or remake a denture in the Japanese version and patient’s with TMD pain in the German version). The ability of OHIP5-Ar to predict the changes of different types of dental treatment has been proven in our study. It makes this instrument more suitable to be used in everyday dental practice. For assessing sensitivity to change, the time between intervention and recall assessment ranged from 10 days to 1 month. Even though the OHIP usually has a recall period of 1 month [
23,
26], i.e., respondents are asked what they have experienced during the previous month and requiring at least 1 month between an intervention and the follow-up assessment to capture the entire treatment effect, changes in OHIP5-Ar summary scores were statistically significant and of moderate to large effect size. This also suggests that a shorter recall period (7 days) as suggested by Waller et al. [
68], where patients are asked about their experiences during the previous week, could also be applied for the OHIP5-Ar. However, we believe that the ability of OHIP to detect the changes during this short period depends on the type of treatment itself. For example, adaptation to 3-unit fixed partial denture will take shorter time than adaptation to lower complete denture, and this will be definitely mirrored in the patient’s response.
In comparison with the longer Arabic versions of this instrument which were validated by Al-Jundi et al. (OHIP49-Ar) [
30] and Khalifa et al. (OHIP14-Ar), [
31] the internal consistency of our version is higher than that of the OHIP49-Ar version (0.74), [
30] and similar to that of the OHIP14-Ar (0.80). [
31] The inter-item correlation and ICC agreement are also higher in our version. However, the correlation between OHIP5-Ar and both global oral and general health questions are lower than those found in the OHIP49-Ar version. [
30] The OHIP14-Ar version [
31] also reported highly significant correlations between these variables but no coefficients (
r) were mentioned. The higher correlation coefficients in the OHIP49-Ar version might be related to the number of questions (49 questions compared with 5 question) which can influence the relationships between these variables.
Dental status as well as dental treatment can greatly affect the individual’s QoL. Dental caries [
69], teeth missing [
56], gum disease [
70], aphthous ulcers [
71], and temporomandibular disorders [
72] are some examples of such pathologies that affect the QoL. Moreover, dental anxiety can also affect the individual perception of QoL [
73]. On the other hand, many or almost all dental treatments can significantly improve the individual’s QoL [
74‐
77]. The current study has strengths and limitations. We used subjective measures of perceived oral and general health and objective measures of physical oral health status for validity assessments, what is considered a strength of the study. Furthermore, we applied sophisticated statistical analyses, i.e., SEM analysis and latent factor model. The sample size was sufficiently large to unambiguously prove reliability, validity, and sensitivity to change of the OHIP5-Ar. Our sample was recruited from specialist and general dental clinics attenders. A longer version of OHIP (OHIP14) was also validated among general dental clinics attenders in Scotland [
78]. Although our study population could be considered a representative for dental patients, a community-based study with large sample size is recommended to validate the instrument among the general population. We didn’t correlate the OHIP5-Ar to the long Arabic version (OHIP49-Ar). Since prior studies have reported a high correlation between short and long versions [
14,
26,
28], therefore this study did not focus on highlighting associations on this subject. Only one component (missing teeth) of the DMFT index was assessed, further studies could include this index to further validate the instrument.