© 2008/JupiterImages Corporation/ Medioimages
Oral health means more than healthy teeth—it is integral to general health and essential for well-being (Bailey, Gueldner, Ledikwe, & Smiciklas-Wright, 2005; Coleman, 2002; Petersen, 2003). Older adults’ oral health status has been gaining more importance in developing and developed countries as the proportion of older adults in the population slowly but continuously increases, due to increase in life expectancy.
Turkey has the same issue on its agenda. According to the country’s 2000 census, the elderly population (those older than age 65) is 3,858,949, which is 5.7% of the total population, an increase of 32.6% since 1990 (State Institute of Statistics, 2003). Although the majority of older adults in Turkey live independently in the community, a growing number live in different kinds of institutions. A total of 12,359 and 8,481 older adults live in the country’s 103 public and 156 private nursing or residential homes, respectively (Social Services and Society for the Protection of Children Head Office, 2008). Most surveys indicate that the older adults who live in residential homes have the worst oral health. Unfortunately, data are lacking in Turkey regarding the oral health of those living in residential or nursing homes (Ünlüer, Gökalp, & Doǧan, 2007).
Oral health, although rarely life threatening, plays an essential role in older adults’ quality of life, management of medical problems, nutrition, and social interaction (Gil-Montoya, de Mello, Cardenas, & Lopez, 2006). Because of the elderly population’s increased longevity, the impact of oral health on the quality of life of older adults without their own teeth may have increased during the past decade (Gift & Redford, 1992).
Nurses have a professional duty to ensure basic oral health care for nursing home residents and have unique opportunities for oral health promotion, screening, early detection of disease, and referral to dentists (Coleman, 2004; Kayser-Jones, Bird, Paul, Long, & Schell, 1995). Nurses can identify oral health problems, initiate appropriate interventions (e.g., referral to a dentist or hygienist), and prevent or minimize the significant morbidity associated with poor oral health.
Oral Health Measures
Studies show that the use of a self-report oral health index improves the quality of nursing care (Bailey et al., 2005; Wårdh, Hallberg, Berggren, Andersson, & Sörensen, 2000). A variety of oral health-related quality of life instruments have been developed in the past 20 years as a result of increased concern about the impact of oral conditions on a person’s quality of life. Measures of oral health-related quality of life (OHRQL) assess the functional, psychological, and social effects of oral diseases and disorders. Prior to the development of these measures, subjective perceptions of oral health were usually elicited by means of single-item global indicators. These included self-rated oral health, self-perceived need for dental treatment, and satisfaction/dissatisfaction with oral health status.
Numerous multidimensional, multi-item scales and indexes are now available (Atchison, Der-Martirosian, & Gift, 1998; Dolan, Peek, Stuck, & Beck, 1998; Locker & Gibson, 2005). The majority of these measures have been shown to have adequate reliability and validity. The Geriatric Oral Health Assessment Index (GO-HAI) is one of the most comprehensive instruments available (Wong, Liu, & Lo, 2002). This instrument was developed in the United States in 1990 by Atchison and Dolan and has been psychometrically tested in different countries (Hägglin, Berggren, & Lundgren, 2005; Ikebe, Sajima, Nokubi, & Ettinger, 2003; Othman et al., 2006; Tubert-Jeannin, Riordan, Morel-Papernot, Porcheray, & Saby-Collet, 2003; Wong et al., 2002).
The purpose of this study was to translate the original English version of the GOHAI into a Turkish version and to test its reliability and validity. The reason for choosing the GOHAI was that it is an internationally well-established OHRQL instrument and is fairly compact, including only 12 items, which may improve response rates. The short length and self-administered format of the GOHAI may provide a quick and inexpensive method of evaluating the oral health condition of older adults. Using this newly translated and validated tool in Turkey to obtain data could better direct nursing care. In addition, the tool may be useful in a broad range of practice settings.
Design and Sample
A cross-sectional survey of a convenience sample was conducted to examine the reliability and validity of the GOHAI. The data were collected from residents of a residential home in the Izmir municipality of Turkey. Individuals were selected to participate in the study if they met the following inclusion criteria: willingness to participate, age 65 and older, and had no apparent cognitive impairment after assessment by the researcher. To maximize recruitment, potential participants were identified using three approaches: identifying older adults who met inclusion criteria, obtaining permission from the residential home, and contacting and enrolling those who were eligible.
Of the 260 older adults living in the residential home, 100 were in the bedridden and emergency care unit. Of the remaining 160 residents (a group of independently living adults), 101 (63.1%) were able to respond to a questionnaire, were willing and gave fully informed consent, and were subsequently asked to participate in this study.
The questionnaire was composed of questions on sociodemographic data (e.g., age, gender, level of education, marital status) and the GOHAI. In addition, the participants were questioned on their perceptions of their general and oral health, their satisfaction with their oral health, and whether they had a perceived need for dental treatment related to their oral health. Data were collected at the participant’s residence by a dentist and a public health nurse (S.E. and G.C.A.) by means of a personal interview.
The 12 items of the GOHAI assess three dimensions of oral health (Atchison & Dolan, 1990):
- Physical function, including eating, speaking, and swallowing.
- Psychosocial function, including worry or concern about oral health, dissatisfaction with appearance, self-consciousness about oral health, and avoidance of social contacts because of oral problems.
- Pain or discomfort, including the use of medication to relieve pain or discomfort in the mouth. Participants were asked if they have always, often, sometimes, seldom, or never experienced any of these problems in the past 3 months. Questions were worded both positively and negatively, requiring the respondents to consider their answers.
Since its development, the GOHAI has been translated into Chinese (Wong et al., 2002), French (Tubert-Jeannin et al., 2003), and Swedish (Hägglin et al., 2005). The index has been found to be valid for use with younger adults (Hägglin et al., 2005; Tubert-Jeannin et al., 2003), with satisfactory psychometric properties among ethnically diverse samples (Hägglin et al., 2005) and in differing groups of older adults (Ikebe et al., 2003).
The Translation Process. In the first stage, the GOHAI was translated separately from English to Turkish by three teaching staff nurses, whose native language is Turkish. Subsequently, it was back translated from Turkish to English by three experts, whose native language is English. All translators worked independently and were not associated with the research in any other way. Once these forward and backward translations were completed, the original and back translations of both English and Turkish versions were carefully compared, word by word.
Once translated, the GOHAI was evaluated by five teaching staff (four in the field of nursing and one an expert in the field of dentistry), and the final version was made by the researchers according to the suggestions.
Ethics. Permission for use of the GOHAI was obtained by e-mail from Professor Kathryn Atchison. Written approval was obtained at the planning stage of the study from the Ethics Committee of the participating university. The study’s objective was explained orally to the residents at the study site.
SPSS, version 11, was used for data analysis. Responses on the GOHAI were scored on a scale ranging from 1 to 5. When data were transferred to the computer, the responses were recoded, and the GOHAI scores ranged from 0 to 60, with higher scores indicating greater self-reported oral health status. Reliability was assessed using the internal consistency approach. The Cronbach’s alpha coefficient was calculated to assess the degree of internal consistency and homogeneity between the items. Pearson’s correlation coefficient was used to measure the item-scale correlation. Concurrent validity was evaluated by examining the association between the GOHAI score and the global health/oral health rating questions.
In this study, it was hypothesized that for concurrent validity, lower GOHAI scores would be associated with perceived need for dental treatment, poor self-reported general health/oral health, and dissatisfaction with oral health status. General and oral health self-ratings were scored as 3 (good), 2 (fair), and 1 (poor). In addition, the GOHAI scores were compared with the sociodemographic variables. The relationship between scores and categorical variables was determined using appropriate nonparametric tests, including the Mann-Whitney and Kruskall-Wallis tests. The principal component factor analysis was performed to investigate the number of components contributing to the GOHAI responses. Factor analysis with varimax rotation was conducted. Factors with an eigenvalue greater than 1 were extracted.
Characteristics of the Sample
All participants were between ages 61 and 102, with a mean age of 75.51 (SD = 8.33 years). The majority (51.5%) were women. The education level of the participants was low: 63.4% had not received any formal education, 17.8% had attained primary education, 17.8% had attained secondary education, and only 1% had attained college education. Only 4% of the participants were married.
Thirty-one percent perceived that they needed dental treatment, 47% rated their general health as poor, and 57% rated their oral health as poor. Similarly, 51.5% of the participants were dissatisfied with their oral health. The mean GOHAI score was 42.5 (SD = 7.4) with a median of 42 and a score range of 19 to 60.
The Cronbach’s alpha coefficient for the Turkish version of the GOHAI was 0.75. This measure indicated a high degree of internal consistency and homogeneity among items. In new instruments, an internal consistency ⩾0.50 is thought to be satisfactory (Erefe, 2002). All item-scale correlations were ⩾0.3 (range = 0.30 to 0.67). Item 5 (“Able to eat without discomfort”) had the lowest value (0.30).
The concurrent validity evaluation results are shown in Table 1. No significant differences in the GOHAI scores were found for gender or age (Mann-Whitney, p > 0.05), and no difference was determined between the GOHAI scores for perceived need for dental treatment (Mann-Whitney, p > 0.05) and perceived general health (Kruskall-Wallis, p > 0.05). Low GOHAI scores were associated with perceived poor oral health (Kruskall-Wallis, p < 0.01) and dissatisfaction with oral health (Mann-Whitney, p < 0.05).
Table 1: Concurrent and Discriminant Validity of GOHAI Scores and Selected Independent Variables (N = 101)
The principal component analysis using varimax rotation extracted four components with eigenvalues above 1. The total variance explained was 64%. The Kaiser-Meyer-Olkin measure for sampling adequacy was acceptable at 0.71. All items loaded higher than 0.40. Factor 1 is related to the theoretical construct of psychosocial functioning. Factor 2 is related mainly to the construct of physical functioning. Factors 3 and 4 were performed by using the 5 items about pain and discomfort and the other two constructs, as shown in Table 2.
Table 2: Principal Components and Factor Loadings of the Rotated Factors Resulting from Two Separate Factor Analyses (N = 101)
Because of uncertainty regarding the nature of factors 3 and 4 and a scree plot, the slope began to decrease after the first two factors. That is, the generation of a scree plot suggested a two-factor solution. Therefore, a new factor analysis with only two factors was performed. These two factors explained 43.57% of the total variance of self-reported oral health as reported by the translated GOHAI. The first factor included the items related to social dimension: “Limit contact with people,” “Pleased with look of teeth,” “Worried about teeth, gums, dentures,” “Self-conscious of teeth, gums, dentures,” and “Uncomfortable eating in front of people.” Only one item, “Sensitive to hot/cold/sweet foods,” is related to pain and discomfort. The second factor consisted of six items about the physical dimension, which refers to the theoretical constructs of pain and discomfort and physical functioning. These items were “Limit kinds of food,” “Trouble biting or chewing,” “Able to swallow comfortably,” “Unable to speak clearly,” “Able to eat without discomfort,” and “Used medication to relieve pain.”
This article describes the initial reliability and validity testing of the 12-item GOHAI in Turkish. An instrument adapted for use in another country or culture should be culturally relevant and valid for the local population, while demonstrating acceptable psychometric properties. Thus, it is important to carry out a rigorous translation and validation process before an instrument developed in one culture is used in a different cultural setting (Hägglin et al., 2005; Othman et al., 2006; Tubert-Jeannin et al., 2003).
In this study, the first step consisted of using a standardized translation process (Aksayan & Gözüm, 2002). Translation and back translation were conducted to ensure the accuracy and interpretability of the questions. This allowed the creation of a Turkish version, which exhibits satisfactory psychometric properties. The internal reliability was successfully tested in various ways (Erefe, 2002). In the current study, reliability of the index was measured with Cronbach’s alpha coefficient and item-scale correlation. The Cronbach’s alpha coefficient was 0.75. Although there is no actual lower limit to the coefficient, this value is higher than the recommended value of 0.70 (Erefe, 2002). It is similar to that in a study by Locker, Matear, Stephens, Lawrence, and Payne (2001), in which the Cronbach’s alpha coefficient of the GOHAI was 0.75. In previous surveys, the Cronbach’s alpha coefficient varied from 0.74 to 0.86 (Hägglin et al., 2005; Ikebe et al., 2003; Kressin, Atchison, & Miller, 1997; Othman et al., 2006; Tubert-Jeannin et al., 2003). This good internal consistency supports the use of the GOHAI in a variety of samples.
In this study, the item-scale correlation range was 0.30 to 0.67, and in a manner parallel to this, the study by Wong et al. (2002) had an item-scale correlation range of 0.28 to 0.61. In the Swedish version of the GOHAI, the correlation values were found to be somewhat higher (0.50 to 0.83) (Hägglin et al., 2005).
In the current study, the lowest item-scale rating was for item 5. The problem of item 5 has been discussed previously. The low correlation value observed for item 5 can be explained by the complex sentence structure used: The sentence has both a positive (“able to eat anything”) and a negative (“without feeling discomfort”) statement. Participants with poor education (as in our study) may have had some difficulty interpreting it. Because similar misunderstanding was predicted by Othman et al. (2006) and Wong et al. (2002), the reversed items (3, 5, 7) were therefore negatively worded in their studies. We agree such an alteration would ensure easier comprehension of these questions by Turkish older adults.
The GOHAI score of the current study (42.5, SD = 7.4) was similar to that in both the Chinese (48.9, SD = 7.2) and Malaysian studies (46.2, SD = 9.7). In accordance with other studies, no significant differences in GOHAI scores were found for gender (Hägglin et al., 2005; Kressin et al., 1997; Tubert-Jeannin et al., 2003). The analyses showed no significant differences for age and GOHAI scores; however, previous studies showed diverse results about age and GOHAI scores (Kressin et al., 1997; Othman et al., 2006; Tubert-Jeannin et al., 2003).
Robinson, Gibson, Khan, and Birnbaum (2003) discussed the varying results for age and GOHAI scores and stated that many studies (such as this one) are conducted among narrow age spans, resulting in no or small age differences. A lower impact in older adults may be explained by the fact that older adults’ expectations were found to be low when oral health was poor and dental care was less effective (Robinson et al., 2003). It may also be the result of people changing their minds as they get older.
In the current study, no statistically significant associations were found for the self-perceived need for dental treatment and perceived general health. Ikebe et al. (2003) reported contrary findings in a population with a mean age of 65, as significant associations were found for these self-reported measures. On the other hand, it was found that the mean GOHAI scores were lower for participants with poorer perceived oral health and dissatisfaction with oral health. However, the literature contains examples of significant relationships between the GOHAI and all self-reported measures. In a study by Locker et al. (2001), both measures demonstrated good concurrent validity, as they did in the current study. The results show that two hypotheses related to concurrent validity were met.
Factor analyses of the GOHAI seem to reveal different dimensional structures (one, two, three) in published studies, which are probably due to the characteristics of the samples and populations (Hägglin et al., 2005; Ikebe et al., 2003; Tubert-Jeannin et al., 2003). Only one factor emerged in the factor analysis of the original English GOHAI version (Atchison & Dolan, 1990). In the Chinese and Swedish versions, the factor analyses revealed three factors (Hägglin et al., 2005; Wong et al., 2002). Two factors emerged in the Turkish version when using the principal components with eigenvalues greater than 1. However, in contrast to the Turkish version, the three factors of the Swedish and Chinese versions did not support the theoretical construct reported by the GOHAI. Those authors suggested that the GOHAI is better viewed as measuring one or two dimensions rather than three separate dimensions of impact (Kressin et al., 1997).
It is also important to note that Atchison and Dolan (1990) did not specify the items belonging to each dimension of the theoretical construct of the index. Therefore, the assignment of items to different dimensions varies between studies of the GOHAI (Kressin et al., 1997; Wong et al., 2002). In the current study, the item “Unable to speak clearly” was viewed as a physical functioning dimension, as it was in other research (Hägglin et al., 2005; Othman et al., 2006), but some studies labeled it as a social dimension (Kressin et al., 1997; Wong et al., 2002).
Limitations to this study include the small sample, limiting options for data analyses, and the convenience of selecting older adults. Another limitation is the use of self-report assessed oral health measures.
Nursing Implications and Conclusion
We found that the Turkish version of the GOHAI showed acceptable reliability and validity. The Turkish version of the GOHAI seems to provide valuable information about oral health conditions and psychosocial and functional problems in older adults. In general, the GOHAI can be recommended as a quick and easy instrument for use in residential homes, especially by nurses. Nevertheless, a review of some of the items used in the original GOHAI questionnaire may be needed. The effect of wording some items in a positive format and others in a negative format should be studied. It would also be interesting to use the Turkish version with other populations, such as adults or members of different social class, to assess its generalizability. Further research using test-retest procedures is also needed to examine the stability of the GOHAI over time.
- Aksayan, S & Gözüm, S2002. Guidebook I for the adaptation of intercultural criteria: The stages of adapting criteria and language adaptation. The Turkish Journal of Research and Development in Nursing, 4, 9–14.
- Atchison, KA, Der-Martirosian, C & Gift, HC1998. Components of self-reported oral health and general health in racial and ethnic groups. Journal of Public Health Dentistry, 58, 301–308. doi:10.1111/j.1752-7325.1998.tb03013.x [CrossRef]
- Atchison, KA & Dolan, TA1990. Development of the Geriatric Oral Health Assessment Index. Journal of Dental Education, 54, 680–687.
- Bailey, R, Gueldner, S, Ledikwe, J & Smiciklas-Wright, H2005. The oral health of older adults: An interdisciplinary mandate. Journal of Gerontological Nursing, 317, 11–17.
- Coleman, P2002. Improving oral health for the frail elderly: A review of widespread problems and best practices. Geriatric Nursing, 23,189–199. doi:10.1067/mgn.2002.126964 [CrossRef]
- Coleman, PR2004. Promoting oral health in elder care: Challenges and opportunities. Journal of Gerontological Nursing, 304, 3.
- Dolan, TA, Peek, CW, Stuck, AE & Beck, JC1998. Three-year changes in global oral health rating by elderly dentate adults. Community Dentistry and Oral Epidemiology, 26, 62–69. doi:10.1111/j.1600-0528.1998.tb01926.x [CrossRef]
- Erefe, I. (Ed.). 2002. The characteristics of data gathering instruments. In The principles, process and methods of research in nursing (pp. 169–188). Istanbul, Turkey: Odak Publications.
- Gift, HC & Redford, M1992. Oral health and the quality of life. Clinics in Geriatric Medicine, 8, 673–683.
- Gil-Montoya, JA, de Mello, AL, Cardenas, CB & Lopez, IG2006. Oral health protocol for the dependent institutionalized elderly. Geriatric Nursing, 27, 95–101. doi:10.1016/j.gerinurse.2005.12.003 [CrossRef]
- Hägglin, C, Berggren, U & Lundgren, J2005. A Swedish version of the GOHAI index: Psychometric properties and validation. Swedish Dental Journal, 29, 113–124.
- Ikebe, K, Sajima, H, Nokubi, T & Ettinger, RL2003. Application of Geriatric Oral Health Assessment Index to elderly Japanese. Retrieved May 20, 2007, from the International Association for Dental Research Web site: http://iadr.confex.com/iadr/2003Goteborg/techprogram/abstract_37183.htm
- Kayser-Jones, J, Bird, WF, Paul, SM, Long, L & Schell, ES1995. An instrument to assess the oral health status of nursing home residents. The Gerontologist, 35, 814–824.
- Kressin, NR, Atchison, KA & Miller, DR1997. Comparing the impact of oral disease in two populations of older adults: Application of the geriatric oral health assessment index. Journal of Public Health Dentistry, 57, 224–232. doi:10.1111/j.1752-7325.1997.tb02979.x [CrossRef]
- Locker, D & Gibson, B2005. Discrepancies between self-ratings of and satisfaction with oral health in two older adult populations. Community Dentistry and Oral Epidemiology, 33, 280–288. doi:10.1111/j.1600-0528.2005.00209.x [CrossRef]
- Locker, D, Matear, D, Stephens, M, Lawrence, H & Payne, B2001. Comparison of the GOHAI and OHIP-14 as measures of the oral health-related quality of life of the elderly. Community Dentistry and Oral Epidemiology, 29, 373–381. doi:10.1034/j.1600-0528.2001.290507.x [CrossRef]
- Othman, WN, Muttalib, KA, Bakri, R, Doss, JG, Jaafar, N & Salleh, N et al. 2006. Validation of the Geriatric Oral Health Assessment Index (GOHAI) in the Malay language. Journal of Public Health Dentistry, 66, 199–204. doi:10.1111/j.1752-7325.2006.tb02580.x [CrossRef]
- Petersen, PE2003. The world health report 2003: Continuous improvement of oral health in the 21st century–The approach of the WHO Global Oral Health Programme. Community Dentistry and Oral Epidemiology, 31Suppl. 1, 3–23. doi:10.1046/j..2003.com122.x [CrossRef]
- Robinson, PG, Gibson, B, Khan, FA & Birnbaum, W2003. Validity of two oral health-related quality of life measures. Community Dentistry and Oral Epidemiology, 31, 90–99. doi:10.1034/j.1600-0528.2003.00051.x [CrossRef]
- Social Services and Society for the Protection of Children Head Office (T.C. Basbakanlik Sosyal Hizmetler ve Çocuk Esirgeme Kurumu). 2008. Elderly care services. Retrieved July 28, 2008, from http://www.shcek.gov.tr/hizmetler/yasli/Yasli_Bakim_Hizmetleri.asp
- State Institute of Statistics. 2003. 2000 census of population (Publication No. 2759). Ankara, Turkey: Author.
- Tubert-Jeannin, S, Riordan, PJ, Morel-Papernot, A, Porcheray, S & Saby-Collet, S2003. Validation of an oral health quality of life index (GOHAI) in France. Community Dentistry and Oral Epidemiology, 31, 275–284. doi:10.1034/j.1600-0528.2003.t01-1-00006.x [CrossRef]
- Ünlüer, Ş, Gökalp, S & Do an, BG2007. Oral health status of the elderly in a residential home in Turkey. Gerodontology, 24, 22–29. doi:10.1111/j.1741-2358.2007.00136.x [CrossRef]
- Wårdh, I, Hallberg, LR, Berggren, U, Andersson, L & Sörensen, S2000. Oral health care: A low priority in nursing. In-depth interviews with nursing staff. Scandinavian Journal of Caring Sciences, 14, 137–142. doi:10.1080/02839310050162370 [CrossRef]
- Wong, MC, Liu, JK & Lo, EC2002. Translation and validation of the Chinese version of GOHAI. Journal of Public Health Dentistry, 62, 78–83. doi:10.1111/j.1752-7325.2002.tb03426.x [CrossRef]
Concurrent and Discriminant Validity of GOHAI Scores and Selected Independent Variables (N = 101)
|Variable||Median GOHAI Score||Mann-Whitney Test|
| Men (n = 49)||48|
| Women (n = 52)||54|
| 60 to 75 (n = 53)||42|
| >76 (n = 48)||43.5|
|Perceived need for dental treatment||−0.19||0.842|
| Yes (n = 31)||51|
| No (n = 70)||50|
|Satisfied with oral health||−2.85||0.004*|
| Yes (n = 49)||45|
| No (n = 52)||41.5|
|Perceived general health||0.52||0.770|
| Good (n = 21)||44|
| Fair (n = 33)||42|
| Poor (n = 47)||42|
|Perceived oral health||20.73||0.000*|
| Good (n = 9)||52|
| Fair (n = 34)||43|
| Poor (n = 58)||42|
Principal Components and Factor Loadings of the Rotated Factors Resulting from Two Separate Factor Analyses (N = 101)
|1. Limit kinds of food||0.82||0.70|
|2. Trouble biting or chewing||0.76||0.61|
|3. Able to swallow comfortably||0.58||0.63|
|4. Unable to speak clearly||0.40||0.59||0.33|
|5. Able to eat anything without feeling discomfort||0.70||0.51|
|6. Limit contact with people||0.74||0.39|
|7. Pleased with look of teeth||0.52||0.42||0.60|
|8. Used medication to relieve pain||0.57||0.43|
|9. Worried about teeth, gums, dentures||0.70||0.70|
|10. Self-conscious of teeth, gums, dentures||0.89||0.89|
|11. Uncomfortable eating in front of people||0.90||0.89|
|12. Sensitive to hot/cold/sweet foods||0.53||0.55|