Next Article in Journal
Epidemiological Algorithm for Early Detection of COVID-19 Cases in a Mexican Oncologic Center
Previous Article in Journal
Measuring the Physical Activity of Seniors before and during COVID-19 Restrictions in the Czech Republic
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

The MOHIP-14PW (Modified Oral Health Impact Profile 14-Item Version for Pregnant Women): A Real-World Study of Its Psychometric Properties and Relationship with Patient-Reported Oral Health

by
Chengwu Yang
1,2,*,
Shulamite S. Huang
3,
Tiffany A. Moore Simas
1,2,
Hugh Silk
4,
Judith A. Savageau
4 and
Stefanie L. Russell
3,5,*
1
Department of Population and Quantitative Health Sciences, UMass Chan Medical School, Worcester, MA 01655, USA
2
Department of Obstetrics and Gynecology, UMass Chan Medical School, Worcester, MA 01655, USA
3
Department of Epidemiology and Health Promotion, College of Dentistry, New York University, New York, NY 10010, USA
4
Department of Family Medicine and Community Health, UMass Chan Medical School, Worcester, MA 01655, USA
5
Department of Oral Surgery, Bellevue Hospital, New York, NY 10016, USA
*
Authors to whom correspondence should be addressed.
Healthcare 2022, 10(3), 461; https://doi.org/10.3390/healthcare10030461
Submission received: 17 January 2022 / Revised: 18 February 2022 / Accepted: 23 February 2022 / Published: 1 March 2022
(This article belongs to the Topic State-of-the-Art Dentistry and Oral Health)

Abstract

:
Background: The 14-item version of the Oral Health Impact Profile (OHIP-14) has been widely used as a measure for oral health-related quality of life (OHQoL) since its publication in 1997. However, few studies have examined its psychometric properties and relationship with patient-reported oral health in pregnant women. Aim: To offer empirical evidence for appropriate use of the OHIP-14 among pregnant women in research and clinical practice. Objectives: (1) to empirically investigate the psychometric properties of the OHIP-14, (2) to modify it into the MOHIP-14PW (modified OHIP-14 for pregnant women), and (3) to compare their relationships with patient-reported oral health in pregnant women. Methods: In this real-world study (RWS) from suburban New York clinics, we collected OHIP-14 data from 291 pregnant women and assessed its psychometric properties at the item-, dimension-, and measure-level, including confirmatory factor analysis (CFA) and exploratory factor analysis (EFA). Accordingly, we modified the OHIP-14 into the MOHIP-14PW. Finally, we compared their correlations with patient-reported oral health scores. Results: All OHIP-14 items had severely skewed distributions, and two had a correlation with the patient-reported oral health < 0.1. All seven pairs of items correlated well (0.47 to 0.62), but the Cronbach’s alphas indicated suboptimal reliability, with two below 0.70. CFA results offered suboptimal support to the original structure, and EFA found a three-dimensional structure best fitted the data. Therefore, we modified the OHIP-14 into the MOHIP-14PW. CFA on the MOHIP-14PW offered stronger supports, and the Cronbach’s alphas increased to 0.92, 0.72, and 0.71. The MOHIP-14PW’s dimensions were more meaningful to pregnant women and had stronger relationships with patient-reported oral health than the OHIP-14; the average correlation coefficients increased by 26% from 0.19 in OHIP-14 to 0.24 in the MOHIP-14PW. Conclusions: The original OHIP-14 required modifications at the item-, dimension-, and measure- level, and the MOHIP-14PW had better psychometric properties, easier interpretation, and stronger correlation with patient-reported oral health in low-income pregnant women. Through an interdisciplinary RWS on a large sample of pregnant women, this study offers concrete empirical evidence for the advantages of the MOHIP-14PW over the original OHIP-14.

1. Introduction

Globally, oral diseases are highly prevalent and greatly reduce oral health-related quality of life (OHQoL) [1]. This relationship is particularly true for pregnant women, when their bodies undergo significant hormonal and physiological changes, affecting gingival inflammation [2,3]. The presence of dental and gingival disease negatively affects the self-perception of OHQoL in pregnant women, affecting their mental state and ability to achieve proper nutrition [2]. Low-income pregnant women are at particularly high risk of oral diseases and poor OHQoL [4].
While there are multiple OHQoL instruments [5,6,7,8,9], the Oral Health Impact Profile (OHIP) [10,11] is most widely used for both adults and pregnant women. According to Google Scholar, the original development article of the 49-item OHIP [10] and its shorter version, OHIP-14 [11], have been cited 2,873 and 2,590 times as of December 10, 2021, both with an average of approximately 100 citations per year. A newly published systematic review of adult OHQoL instruments found that the OHIP-14 was the most frequently used (26.8%), while the OHIP-49 ranked third [12]. In addition, the literature suggests that the “OHIP-14 appeared to be responsive to change” [13]. A recent systematic review of OHQoL during pregnancy included eight studies [2] and identified four OHQoL measures that have been used, with the OHIP-14 being the most common (62.5%). Additionally, some randomized controlled trials (RCTs) among pregnant women have used the OHIP-14 scores as the primary outcome [14].
However, we are uncertain if our tools for assessing OHQoL are accurate, both in the general population and in pregnant women. Despite its popularity, studies evaluating the psychometric properties of OHIP are rare [15,16]. Currently, it appears that there is no consensus regarding the dimensionality or the factor structure of the OHIP-14, i.e., which items belong to which dimensions [17]. For example, Santos et al. demonstrated the unidimensionality of the OHIP-14 in a Brazilian sample [18]. Brennan et al. found a multidimensional structure different from the original seven dimensions among a sample of South Australian dental patients [19]. In China, Xin and Ling identified a four-dimensional factor structure [20]. In addition, in another project [21] using the OHIP-14 in patients with severe periodontitis, the two corresponding authors (C.Y. and S.R.) and their team also found that the OHIP-14 had a factor structure other than the seven-dimensional structure proposed by the original author [11]. From a clinical aspect, the OHIP-14 offers seven dimension-scores and one total score, which is burdensome for busy clinicians and researchers. This burden can be eased in practical application if a simpler factor structure of the OHIP-14 can be identified for pregnant women resulting in fewer dimension-scores.
In this real-world study using empirical clinical data collected from two dental clinics, we aim to (1) thoroughly investigate the psychometric properties of the OHIP-14 in pregnant women at the item-, dimension-, and measure-level; (2) modify the scale accordingly into the MOHIP-14PW (modified OHIP-14 for pregnant women); and (3) assess the relationship of patient-reported oral health with the OHIP-14 in comparison with the MOHIP-14PW in low-income pregnant women. Finally, based on our findings, we propose further modifications of the MOHIP-14PW for studies examining OHQoL among pregnant women.

2. Methods

2.1. Participant Recruitment

We recruited low-income pregnant women aged 18 years and above from two prenatal clinics on Long Island, New York. These clinics, located in suburban Nassau County on Long Island, serve Medicaid-eligible women. The Center for Maternal Oral Health (CMOH), a program within the dental practice to which women are referred, is in Great Neck, NY. This study received approval from Institutional Review Boards (IRB) at the New York University Medical Center and Long Island Jewish, North Shore, NY.

2.2. Data Collection

2.2.1. Survey

We collected survey data via a written questionnaire provided to pregnant women as they visited the clinic for their prenatal care. We derived this questionnaire from those used previously [22]. Surveys were conducted in English and Spanish, and participants received a USD 5 gift card for participating and completing the survey. Aside from the Hispanic participants, all other pregnant women completed the survey in English. We included demographics in the questionnaire, such as age, race/ethnicity, place of birth, education, household income, marital status, and parity, i.e., number of live births. We also asked participants to complete the OHIP-14. In addition, all participants self-reported their overall condition of oral health by rating it into five categories: 1 = “excellent”, 2 = “very good”, 3 = “good”, 4 = “fair”, 5 = “poor”.

2.2.2. Data Management

We transferred all data from the paper survey into an electronic dataset using the SPSS statistical software, version 25 [23]. To verify and clean all data, we used double-entry and comparisons.

2.3. Data Analysis

2.3.1. Item-Level Analysis of the OHIP-14 and the Patient-Reported Oral Health

For the OHIP-14 and patient-reported oral health data, we calculated means, standard deviations (SD), medians, and distribution of responses (%) for each item. We considered an item as “good” if (1) responses were distributed well across the different categories without a severe floor or ceiling effect, i.e., less than 15% of participants choosing either the lowest or the highest response, and (2) the mean or median were close to the center of the range of response values [24]. For example, the OHIP-14 items should have a mean or median that is close to 2, because these items have response values as 0, 1, 2, 3, and 4.

2.3.2. Dimension-Level Analysis of the OHIP-14 and the Patient-Reported Oral Health

We investigated the correlation matrix of the 14 OHIP items and patient-reported oral health, as well as the internal consistency of the seven dimensions using Cronbach’s alpha [25,26]. Prior literature indicates that a correlation coefficient of less than 0.1 is deemed small, between 0.1 and 0.3 medium, between 0.3 and 0.5 large [27]. Moreover, items measuring the same construct (dimension) should be more correlated than items outside the construct. Prior literature indicates that a Cronbach’s alpha between items measuring the same construct of greater than 0.7, 0.8, and 0.9 is deemed, respectively, as adequate, good, or excellent [27].

2.3.3. Measure-Level Analysis of the OHIP-14

We investigated the factor structure of the OHIP-14 items by using confirmatory factor analysis (CFA) and exploratory factor analysis (EFA) techniques. Specifically, we used a CFA-EFA-ReCFA procedure [28], starting from a CFA according to the original factor structure of the OHIP-14 [11]. If the CFA results were not optimal, we ran an EFA to explore the potential factor structure that best fit this sample. If a new factor structure was found to better fit our sample, we then used an additional CFA (i.e., ReCFA) to confirm the newly identified factor structure. The following model fit index was used in the CFA-EFA-ReCFA procedure: normalized chi-square (NC) < 3.0, root mean square error of approximation (RMSEA) < 0.05, comparative fit index (CFI) > 0.95, Tucker–Lewis index (TLI) > 0.95, and standardized root mean square residual (SRMR) < 0.10 [27,29].

2.3.4. Comparisons of the Relationship of Patient-Reported Oral Health with Dimension Scores from OHIP-14 versus MOHIP-14PW

After identification of the best factor structure for the data, we then calculated dimension scores using the mean of the items scores within each dimension. The correlation of the dimension scores with patient-reported oral health was then calculated and compared between the OHIP-14 and the MOHIP-14PW.
We used the Mplus software, Version 8 [30], for CFA and EFA, and the SPSS software, Version 25.0 [23], for all other statistical analysis.

3. Results

3.1. Demographics of Participants

In total, 291 pregnant women across two clinics participated in this study. We summarized their demographic properties in Table 1. On average, they were in their mid-twenties (26.5 ± 5.4 years old) and self-reported their race/ethnicity as Hispanic (42.3%), Black (32.1%), Asian (14.0%), and White (11.7%). Roughly half (53.1%) were U.S. immigrants. Most women had completed high school or above (82.1%), but more than two-thirds had a household income below USD 40,000/year. Over three-quarters were married or living as married, and roughly half (51.4%) did not have any previous live births.

3.2. Psychometric Properties of the OHIP-14 and the MOHIP-14PW

3.2.1. Item Level

Table 2 displays the distribution of responses to the OHIP-14 items and the patient-reported oral health. All items of the OHIP-14 had severely skewed distributions among the five response categories (from 0 = “never” to 4 = “very often”), with a median of 0.0. The patient-reported oral health item showed excellent variability, with only one of the five response categories having either less than 5% or greater than 95% of participants, and with a median of 3.0 and mean of 2.94, just at the very center of the five categories (1, 2, 3, 4, 5).

3.2.2. Dimension Level

Table 3 summarizes the Cronbach’s alpha, Spearman correlation matrix of the OHIP-14 items, and patient-reported oral health. All pairs of items within each of the original seven dimensions correlated well, with Spearman correlation coefficients ranging from 0.47 to 0.62. Cronbach’s alphas ranged from 0.64 to 0.77, with two dimensions (functional limitation and psychological disability) below 0.70, indicating suboptimal reliability. Two items (2 and 14) had Spearman correlation with the patient-reported oral health less than 0.1 (“small”).

3.2.3. Measure-Level

Table 4 shows the factor analysis results. CFA results offered suboptimal support to the original conceptual seven-dimensional factor structure of the OHIP-14 in this sample: NC = 2.06, RMSEA = 0.061 (95% CI: 0.045, 0.077), CFI = 0.990, TLI = 0.984, and SRMR = 0.040. The RMSEA was greater than 0.05, which indicated the measurement model could be further optimized. Results from the EFA (data not shown, but available by request) indicated that at most four factors could be extracted from the 14 items and that a three-dimensional factor structure was the best fit for the data, with a 9-3-2 item allocation on the three new factors (Table 4).
Following this new three-dimensional factor structure, we modified the OHIP-14 into the modified OHIP-14 for pregnant women (MOHIP-14PW). We then ran another CFA for the MOHIP-14PW and confirmed this new three-dimensional factor structure with stronger supports: NC = 1.56, RMSEA = 0.044 (95% CI: 0.028, 0.059), CFI = 0.993, TLI = 0.992, SRMR = 0.039. Each and all the model-fit indexes showed improvement, and the RMSEA in particular was below the cut-off value of 0.05. In addition, the three new dimensions in the MOHIP-14PW had Cronbach’s alphas of 0.92, 0.72, and 0.71, indicating good to excellent reliability. We renamed the three new dimensions of the MOHIP-14PW as “Physical Impact” (nine items), “Psychological Impact” (three items), and “Pain Impact” (two items).

3.3. Relationships of the Dimension Scores with Patient-Reported Oral Health

Table 5 displays the correlation of the MOHIP-14PW with patient-reported oral health, compared to the correlation of the original OHIP-14 with patient-reported oral health. The MOHIP-14PW was more strongly related to patient-reported oral health than the OHIP-14. The average correlation coefficients increased from 0.19 (range: 0.13–0.28) in OHIP-14 to 0.24 (range: 0.21–0.29) in MOHIP-14PW, a 26% improvement.

4. Discussion

In this real-world study, we found that the original factor structure of the OHIP-14 was not optimally suited for this surveyed sample of low-income pregnant women. We therefore modified the OHIP-14 into the MOHIP-14PW, which has a three-dimensional factor structure that was more suitable for the underlying data. This led to changes of the OHIP-14 scores in the following ways: (1) number of subscales, from seven to three; (2) re-grouping of the items included in OHIP-14; and (3) range of the subscale scores, from 0–8 for all the original seven subscales to 0–8, 0–12, and 0–36 for the three new subscales.
Scale adaptation and modification is a cutting-edge science [31], and it is common practice in scale development [24,32]. A good example is the modified dental anxiety scale (MDAS) [33], which was modified from the original DAS [34,35]. Moreover, modification of a scale can be critical at resecuring positive findings, such as a large randomized controlled trial (RCT) on Parkinson’s disease [36].

4.1. Interpretation of the New Three-Dimensional Factor Structure of the MOHIP-14PW

From a content perspective, the primary change from the OHIP-14 to the MOHIP-14PW is how items are grouped into domains. Specifically, both items 9 and 10 (“difficulty in relaxing due to dental problems” and “embarrassment due to dental issues”) were originally part of the Psychological Disability domain in the OHIP-14. In the MOHIP-14PW, item 9 was included instead in the Physical Impact domain and item 10 into Psychological Impact instead of Psychological Disability.
From a technical and psychometrical perspective, the separation of items 9 and 10 is not surprising if we have a closer look at the item matrix (Table 3). Item 9 is more correlated with items from other dimensions than Item 10. Similarly, Item 10 is more correlated with Items 5 (0.55) and 6 (0.54) than Item 9 (0.47). These correlation patterns are precisely captured in the new dimension structure of the MOHIP-14PW. As a result, both Items 9 and 10 had higher factor loadings in their new dimensions of the MOHIP-14PW compared to their factor loadings in the OHIP-14 (Table 4)—the factor loading on Item 9 increased from 0.870 to 0.906, while the factor loading on Item 10 increased from 0.761 to 0.947. All these technical outputs offered empirical supports for the new dimension structure of the MOHIP-14PW.
A key question is whether Item 9 (“difficulty in relaxing due to dental problems”) conceptually is a psychological or physical attribute. From the content of Item 9, there can be an argument made either way. Hence, which domain Item 9 is grouped into likely depends strongly on the underlying empirical data. The data here suggest that at least for our sample of low-income pregnant women, difficulty in relaxing due to dental problems is primarily a physical rather than psychological issue.

4.2. Advantages and Innovations of the MOHIP-14PW

Several studies have used the OHIP-14 to investigate the OHQoL among pregnant women [2]. As evidenced in our study, the MOHIP-14PW has the following advantages and innovations over the original OHIP-14 among the pregnant women: (1) improved psychometric properties; (2) easier interpretation for pregnant women; and (3) increased relationship with patient-reported oral health. We hope the field will benefit in the future from these important innovations and advantages of the MOHIP-14PW.

4.3. Future Directions

Obtaining data from several hundred pregnant women (n = 291) participating in this study is a meaningful achievement. We are fortunate and uniquely advantaged by having access to large numbers of pregnant women seeking dental care in a metropolitan area. In the future, we plan to collect surveys from other patient groups.
We suggest that modifying the response to the 14 items, from the 5-point Likert type (0, 1, 2, 3, 4) to a continuous scale, such as 0–100, where 0 = “never”, and 100 = “all the time”, would improve the MOHIP-14PW further. The 14 items in the original OHIP-14 all exhibited severely skewed response distributions with a median of 0.0 in this population, which limits the ability of future studies to detect any changes in oral health-related quality of life. However, this severe skewness issue would be easily rectified by modifying the responses to all 14 items from a five-point Likert scale to a continuous scale, such as 0 to 100.
Such a modification to the response scale has been used previously by the lead and corresponding author (C.Y.), which was proven to be critical in avoiding the very likely failure of a large RCT [37]. In that RCT, although with positive findings (i.e., p < 0.05) on the primary outcome measure, the improvements are all from above 90 out of the 100-point scale, such as from 90.2 to 92.1. Had the original primary measure, the Nolan self-efficacy scale [38], not been modified from the original five-point (1, 2, 3, 4, 5) Likert-scale to the 100-point continuous scale [37], the large RCT would not have been able to detect any changes in the primary outcome. As Kline (2015) ([27], p. 91) illustrated, reliability of a dimension in a measure, as assessed by Cronbach’s alpha, is a function of two parameters: (1) number of items within the dimension; and (2) average Pearson correlation between all pairs of items within the dimension. By definition, the five-point Likert responses are categorical, not continuous. Thus, Pearson correlation and Cronbach’s alpha are not as appropriate for a five-point Likert scale. Instead, a 0–100 continuous scale is much more appropriate for Pearson correlation and Cronbach’s alpha. Hence, we consider that future work examining changes in OHQoL should consider further modifying the MOHIP-14PW to allow for a continuous response scale.
We plan to eliminate the items that showed low correlation with the patient-reported oral health, especially items 2 and 14. Close correlation with the content measure (patient-reported oral health) is the fundamental criteria for the inclusion or exclusion of an item. In a similar study [39] on the National Institutes of Health Stroke Scale (NIHSS [40]), 1 of the 15 items was negatively correlated with all other items. After that item was eliminated, the patient-reported performance of the measure improved substantially. Similarly, in another study of the Repeatable Battery for the Assessment of Neuropsychological Status (RBANS [41]) in Parkinson’s disease, half of the 12 items were eliminated due to low correlation with other items. The resulting product exhibited much better performance than the original RBANS [28].
Many factors are related to oral health, including but not limited to age, race/ethnicity, education, diet, and income. Specifically, oral health in low-income pregnant women might be significantly different from that seen in higher-income pregnant women. Moreover, other factors may be associated with more changes to oral health in low-income pregnant women compared to higher-income pregnant women. Therefore, our team has started another study to investigate these potential relationships. Since the original OHIP-14 appeared to be responsive to change [13], we believe that the MOHIP-14PW will maintain this important feature. In the future, we also plan to add a new survey to assess these pregnant women’s perceptions of changes in their oral health following treatment at the clinics, so that we can evaluate the response of MOHIP-14PW to change. In addition, future work examining the MOHIP-14PW for pregnant women should consider avoiding the result deviation caused by language problems across different populations, cultures, and countries by offering the tool in more languages.
We also plan to apply the further-refined MOHIP-14PW onto national and international samples of pregnant women. Broader sampling across sociodemographic backgrounds and geography is warranted to further refine the MOHIP-14PW for this population. Since the women recruited in this sample may only be representative of low-income pregnant women in suburban New York, future work examining the MOHIP-14PW for pregnant women should consider recruiting nationally and internationally representative populations. Such data would enable researchers to examine differential item functioning (i.e., measurement bias at the item level) across different populations, cultures, and countries.

5. Conclusions

In a real-world study with empirical data from a relatively large sample, we modified the popular OHIP-14 into the MOHIP-14PW through full-scale psychometric assessment and refinement. This modification improved both the psychometric properties and relationship with patient-reported oral health for a sample of low-income pregnant patients. This new MOHIP-14PW should be considered for use in future studies in pregnant women for more accurate and meaningful measure of OHQoL, especially if further tested in more diverse populations. The clinical aspects of this research include a decreased burden through the reduction of dimension-scores from seven to three, as well as easier understanding and easier interpretation of the dimensions for pregnant women. The purpose of collecting the MOHIP-14PW is to investigate OHQoL among pregnant women. This scale can be used either in clinical practice or research to ease the burden for clinicians, researchers, and pregnant women.

Author Contributions

Conceptualization: C.Y., S.L.R., S.S.H., T.A.M.S. and H.S. Data curation: S.L.R. Formal analysis: C.Y. and S.L.R. Funding acquisition: S.L.R. Investigation: C.Y., S.L.R., S.S.H., T.A.M.S., H.S. and J.A.S. Methodology: C.Y., S.L.R., S.S.H., T.A.M.S. and H.S. Project administration: S.L.R. and C.Y. Software: C.Y. and S.L.R. Supervision: C.Y. and S.L.R. Writing—original draft: C.Y., S.L.R. and S.S.H. Writing—review and editing: C.Y., S.L.R., S.S.H., T.A.M.S., H.S. and J.A.S. All authors have read and agreed to the published version of the manuscript.

Funding

This study was funded by an internal grant for faculty from the New York College of Dentistry, “Dean’s Award,” May 26, 2010, ($5390.19). PI: Dr. Stefanie L. Russell, titled “Pilot and Exploratory Studies on Oral Health, Access and Utilization of Dental Care in Low-income, Urban Pregnant Women”, 2010.

Institutional Review Board Statement

Prior to the implementation of this study, this study was judged as exempt under the Institutional Review Boards (IRB) at both New York University (15 July 2010) and Long Island Jewish, North Shore Medical Center (9 July 2010). All participants agreed to participate and understood their participation was voluntary. Moreover, we performed this study in accordance with the ethical standards as laid down in the 1964 Declaration of Helsinki.

Informed Consent Statement

Oral informed consent was obtained from all subjects involved in the study.

Data Availability Statement

The data presented in this study are available on request from the corresponding authors. The data are not publicly available due to other ongoing investigations and manuscripts.

Acknowledgments

We are thankful to participants in this study. The lead author (C.Y.) thanks Alina Niu for her assistance on part of the statistical analyses.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Peres, M.A.; Macpherson, L.M.; Weyant, R.J.; Daly, B.; Venturelli, R.; Mathur, M.R.; Listl, S.; Celeste, R.K.; Guarnizo-Herreño, C.C.; Kearns, C.; et al. Oral diseases: A global public health challenge. Lancet 2019, 394, 249–260. [Google Scholar] [CrossRef]
  2. Fakheran, O.; Saied-Moallemi, Z.; Khademi, A.; Sahebkar, A. Oral Health-Related Quality of Life during Pregnancy: A Systematic Review. Curr. Pharm. Des. 2020, 26, 4014–4021. [Google Scholar] [CrossRef] [PubMed]
  3. Acharya, S.; Bhat, P.V. Oral-health-related quality of life during pregnancy. J. Public Health Dent. 2009, 69, 74–77. [Google Scholar] [CrossRef] [PubMed]
  4. Russell, S.L.; Kerpen, S.J.; Rabin, J.M.; Burakoff, R.P.; Yang, C.; Huang, S.S. A Successful Dental Care Referral Program for Low-Income Pregnant Women in New York. Int. J. Environ. Res. Public Health 2021, 18, 12724. [Google Scholar] [CrossRef] [PubMed]
  5. Slade, G.D.; Strauss, R.P.; Atchison, K.; Kressin, N.; Locker, D.; Reisine, S.T. Conference summary: Assessing oral health outcomes--measuring health status and quality of life. Community Dent. Health 1998, 15, 3–7. [Google Scholar] [PubMed]
  6. Thomson, W.M.; Broder, H.L. Oral–Health–Related Quality of Life in children and adolescents. Pediatric Clin. 2018, 65, 1073–1084. [Google Scholar] [CrossRef]
  7. Yang, C.; Crystal, Y.; Ruff, R.; Veitz-Keenan, A.; McGowan, R.; Niederman, R. Quality Appraisal of Child Oral Health–Related Quality of Life Measures: A Scoping Review. JDR Clin. Transl. Res. 2020, 5, 109–117. [Google Scholar] [CrossRef]
  8. Haag, D.; Peres, K.; Balasubramanian, M.; Brennan, D. Oral Conditions and Health-Related Quality of Life: A Systematic Review. J. Dent. Res. 2017, 96, 864–874. [Google Scholar] [CrossRef]
  9. Tsakos, G.; Allen, F. Oral Health-Related Quality of Life, in Oral Epidemiology; Peres, M.A., Watt, R.G., Eds.; Springer: Berlin/Heidelberg, Germany, 2021; pp. 319–332. [Google Scholar]
  10. Slade, G.D.; Spencer, A.J. Development and evaluation of the oral health impact profile. Community Dent. Health 1994, 11, 3–11. [Google Scholar] [PubMed]
  11. Slade, G.D. Derivation and validation of a short-form oral health impact profile. Community Dent. Oral Epidemiol. 1997, 25, 284–290. [Google Scholar] [CrossRef] [PubMed]
  12. Riva, F.; Seoane, M.; Reichenheim, M.E.; Tsakos, G.; Celeste, R.K. Adult oral health-related quality of life instruments: A systematic review. Community Dent. Oral Epidemiol. 2021, 1–6. [Google Scholar] [CrossRef] [PubMed]
  13. Locker, D.; Jokovic, A.; Clarke, M. Assessing the responsiveness of measures of oral health-related quality of life. Community Dent. Oral Epidemiol. 2004, 32, 10–18. [Google Scholar] [CrossRef] [PubMed]
  14. Musskopf, M.L.; Milanesi, F.C.; da Rocha, J.M.; Fiorini, T.; Moreira, C.H.C.; Susin, C.; Rösing, C.K.; Weidlich, P.; Oppermann, R.V. Oral health related quality of life among pregnant women: A randomized controlled trial. Braz. Oral Res. 2018, 32, e002. [Google Scholar] [CrossRef] [PubMed]
  15. Schierz, O.; Baba, K.; Fueki, K. Functional Oral Health-Related Quality of Life Impact: A Systematic Review in Populations with Tooth Loss. J. Oral Rehabil. 2021, 48, 256–270. [Google Scholar] [CrossRef] [PubMed]
  16. John, M.T.; Reißmann, D.R.; Feuerstahler, L.; Waller, N.; Baba, K.; Larsson, P.; Čelebić, A.; Szabo, G.; Rener-Sitar, K. Factor analyses of the Oral Health Impact Profile–Overview and studied population. J. Prosthodont. Res. 2014, 58, 26–34. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  17. Soares, G.; Santiago, P.; Werneck, R.; Michel-Crosato, E.; Jamieson, L. A psychometric network analysis of OHIP-14 across Australian and Brazilian populations. JDR Clin. Transl. Res. 2021, 6, 333–342. [Google Scholar] [CrossRef] [PubMed]
  18. Santos, C.M.d.; de Oliveira, B.H.; Nadanovsky, P.; Hilgert, J.B.; Celeste, R.K.; Hugo, F.N. Oral Health Impact Profile-14::¿ una escala unidimensional? Cad. D Saúde. Pública. 2013, 29, 749–757. [Google Scholar] [CrossRef]
  19. Brennan, D.S.; Spencer, A.J. Dimensions of oral health related quality of life measured by EQ-5D+ and OHIP-14. Health Qual. Life Outcomes 2004, 2, 35. [Google Scholar] [CrossRef] [Green Version]
  20. Xin, W.; Ling, J. Validation of a Chinese version of the oral health impact profile-14. Chin. J. Stomatol. 2006, 41, 242–245. [Google Scholar]
  21. Sousa, F.; Yang, C.; Oliveria, V.B.; Russell, S.L.; Niederman, R.; Rego, R.O. Severe Periodontitis and OHRQoL Measured by the OHIP-14. In Proceedings of the 2020 IADR/AADR/CADR General Session, Washington, DC, USA, 18 March 2020. [Google Scholar]
  22. Boggess, K.A.; Urlaub, D.M.; Massey, K.E.; Moos, M.-K.; Matheson, M.B.; Lorenz, C. Oral hygiene practices and dental service utilization among pregnant women. J. Am. Dent. Assoc. 2010, 141, 553–561. [Google Scholar] [CrossRef] [Green Version]
  23. IBM. IBM SPSS Statistics for Window; IBM Corp: Armonk, NY, USA, 2017. [Google Scholar]
  24. DeVellis, R.F.; Thorpe, C.T. Scale Development: Theory and Applications; Sage Publications: Sauzende Oaks, CA, USA, 2021. [Google Scholar]
  25. Cronbach, L.J. Coefficient alpha and the internal structure of tests. Psychometrika 1951, 16, 297–334. [Google Scholar] [CrossRef] [Green Version]
  26. Cronbach, L.J.; Shavelson, R.J. My current thoughts on coefficient alpha and successor procedures. Educ. Psychol. Meas. 2004, 64, 391–418. [Google Scholar] [CrossRef]
  27. Kline, R.B. Principles and Practice of Structural Equation Modeling, 4th ed.; Guilford Publications: New York, NY, USA, 2015. [Google Scholar]
  28. Yang, C.; Garrett-Mayer, E.; Schneider, J.S.; Gollomp, S.M.; Tilley, B.C. Repeatable battery for assessment of neuropsychological status in early Parkinson’s disease. Mov. Disord. 2009, 24, 1453–1460. [Google Scholar] [CrossRef] [PubMed]
  29. Brown, T.A. Confirmatory Factor Analysis for Applied Research; Guilford Publications: New York, NY, USA, 2015. [Google Scholar]
  30. Muthén, L.K.; Muthén, B.O. Mplus User’s Guide (1998–2017), 8th ed.; Muthén & Muthén: Los Angeles, CA, USA, 2017. [Google Scholar]
  31. Stewart, A.L.; Thrasher, A.D.; Goldberg, J.; Shea, J.A. A framework for understanding modifications to measures for diverse populations. J. Aging Health 2012, 24, 992–1017. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  32. Yang, C.; Ford, M.E.; Tilley, B.C.; Greene, R.L. Religiosity in black and white older Americans: Measure adaptation, psychometric validation, and racial difference. Medicine 2016, 95, e4257. [Google Scholar] [CrossRef] [PubMed]
  33. Humphris, G.M.; Morrison, T.; Lindsay, S. The Modified Dental Anxiety Scale: Validation and United Kingdom norms. Community Dent. Health 1995, 12, 143–150. [Google Scholar]
  34. Corah, N.L.; Gale, E.N.; Illig, S.J. Assessment of a dental anxiety scale. J. Am. Dent. Assoc. 1978, 97, 816–819. [Google Scholar] [CrossRef]
  35. Corah, N.L. Development of a dental anxiety scale. J. Dent. Res. 1969, 48, 596. [Google Scholar] [CrossRef]
  36. Yang, C.; Vrana, K.E. Rescuing Suboptimal Patient-Reported Outcome Instrument Data in Clinical Trials: A New Strategy. Healthcare 2018, 6, 27. [Google Scholar] [CrossRef] [Green Version]
  37. Green, M.J.; van Scoy, L.J.; Foy, A.J.; Stewart, R.R.; Sampath, R.; Schubart, J.R.; Lehman, E.B.; Dimmock, A.E.F.; Bucher, A.M.; Lehmann, L.S.; et al. A randomized controlled trial of strategies to improve family members’ preparedness for surrogate decision-making. Am. J. Hosp. Palliat. Med. 2018, 35, 866–874. [Google Scholar] [CrossRef]
  38. Nolan, M.T.; Hughes, M.T.; Kub, J.; Terry, P.B.; Astrow, A.; Thompson, R.E.; Clawson, L.; Texeira, K.; Sulmasy, D.P. Development and validation of the family decision-making self-efficacy scale. Palliat. Supportive Care 2009, 7, 315–321. [Google Scholar] [CrossRef] [PubMed]
  39. Yang, C.; Zhao, W. On the relationship between the modified Rankin scale (mRS) and the NIH stroke scale (NIHSS). In Proceedings of the 36th Annual Meeting of the Society for Clinical Trials (SCT), The Society for Clinical Trials (SCT), Arlington, VI, USA, 17–20 May 2015. [Google Scholar]
  40. Brott, T.; Adams, H.P.; Olinger, C.P.; Marler, J.R.; Barsan, W.G.; Biller, J.; Spilker, J.; Holleran, R.; Eberle, R.; Hertzberg, V. Measurements of acute cerebral infarction: A clinical examination scale. Stroke 1989, 20, 864–870. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  41. Randolph, C.; Tierney, M.C.; Mohr, E.; Chase, T.N. The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS): Preliminary clinical validity. J. Clin. Exp. Neuropsychol. 1998, 20, 310–319. [Google Scholar] [CrossRef] [PubMed]
Table 1. Demographic characteristics of the study participants (N = 291).
Table 1. Demographic characteristics of the study participants (N = 291).
Demographic CharacteristicsMean ± SD/n (%)
Age (years)26.5 ± 5.4
Race/ethnicity
 White31 (11.7%)
 Black85 (32.1%)
 Asian37 (14.0%)
 Hispanic112 (42.3%)
Place of birth
 USA136 (46.9%)
 Puerto Rico (USA)9 (3.1%)
 Caribbean (other than Puerto Rico)40 (13.8%)
 Central/South America59 (20.3%)
 South Asia22 (7.6%)
 Other Asia16 (5.5%)
 Other8 (2.8%)
Education
 Lower than high school52 (17.9%)
 High school/some college182 (62.8%)
 College graduate or above56 (19.3%)
Household income
 ≤USD 15,000/year88 (30.2%)
 USD 15,001–40,000/year109 (37.5%)
 ≥USD 40,001/year94 (32.3%)
Marital status
 Married/living as married225 (77.6%)
 Others65 (22.4%)
Parity
 No previous live births148 (51.4%)
 1 previous live birth73 (25.3%)
 ≥2 previous live births67 (23.3%)
Table 2. Distribution of responses to the OHIP-14 items among the study participants.
Table 2. Distribution of responses to the OHIP-14 items among the study participants.
OHIP-14 ItemBrief DescriptionNMeanSDMedianDistribution of Responses (%)
01234
1Difficulty in pronunciation2830.160.480.088.08.82.50.70.0
2Deterioration of taste2780.230.620.085.39.04.01.40.4
3Pain in the mouth2770.560.890.066.814.116.22.20.7
4Discomfort while chewing food2770.430.860.076.29.710.81.81.4
5Self-conscious about teeth, mouth, and dentures2750.631.090.068.711.313.12.54.4
6Tensed due to problems of your teeth and mouth2770.380.820.077.312.36.92.21.4
7Diet unsatisfactory due to dental problems2770.180.580.088.17.23.60.40.7
8Interruption during meals due to dental problems2760.330.770.081.28.78.00.71.4
9Difficulty in relaxing due to dental problems2770.220.560.084.510.54.01.10.0
10Embarrassment due to dental issues2780.420.850.074.812.99.01.41.8
11Irritability with others due to dental problems2770.180.570.088.46.53.21.80.0
12Difficulty in doing usual jobs due to dental problems2780.130.460.091.45.81.81.10.0
13Less satisfaction in life due to dental issues2780.170.530.088.86.83.60.40.4
14Total unable to function due to dental issues2780.120.440.091.74.73.20.40.0
Patient-reported oral healthOverall condition of your oral health 12345
2902.940.903.006.920.047.922.42.8
Likert-type response coding (Slade, 1997): 0 = “never”, 1 = “hardly ever”, 2 = “occasionally’, 3 = “fairly often”, and 4 = “very often”. Overall condition of your oral health: 1 = “excellent”, 2 = “very good”, 3 = “good”, 4 = “fair”, 5 = “poor”.
Table 3. OHIP-14 in the study participants: Cronbach’s alpha of the original 7 dimensions, and Spearman correlation matrix of the 14 items and the overall health.
Table 3. OHIP-14 in the study participants: Cronbach’s alpha of the original 7 dimensions, and Spearman correlation matrix of the 14 items and the overall health.
OHIP-14 Item1234567891011121314
1 (0.67)
20.50
30.280.34 (0.72)
40.360.380.56
50.340.400.220.44 (0.70)
60.400.450.320.530.54
70.450.520.290.460.390.50 (0.72)
80.460.500.390.580.400.450.56
90.580.560.410.510.350.510.510.63 (0.64)
100.270.340.310.510.550.540.400.380.47
110.470.520.260.470.440.560.550.610.590.49 (0.74)
120.570.480.330.450.300.420.630.510.530.400.59
130.470.450.270.400.360.450.640.490.570.530.580.64 (0.77)
140.500.520.340.430.320.430.610.490.540.440.610.670.62
Overall oral health0.140.070.220.150.210.210.120.160.110.280.140.170.180.09
Values in bold are Spearman correlation coefficients within the same original OHIP-14 dimensions. Values in italic indicate Spearman correlation coefficients higher than those within the same original OHIP-14 dimensions. Values in parentheses are the Cronbach alphas for each of the seven original OHIP-14 dimensions.
Table 4. Factor structures of the OHIP-14 in pregnant women: original vs. modified.
Table 4. Factor structures of the OHIP-14 in pregnant women: original vs. modified.
Item #Original Dimension Modified Dimension
Dimension NameFactor LoadingDimension NameFactor Loading
1Functional Limitation0.834Physical Impact0.804
2Functional Limitation0.867Physical Impact0.834
3Physical Pain0.723Pain Impact0.720
4Physical Pain1.004Pain Impact1.008
5Psychological Discomfort0.734Psychological Impact0.724
6Psychological Discomfort0.895Psychological Impact0.889
7Physical Disability0.859Physical Impact0.854
8Physical Disability0.900Physical Impact0.897
9Psychological Disability0.870Physical Impact0.906
10Psychological Disability0.761Psychological Impact0.853
11Social Disability0.922Physical Impact0.924
12Social Disability0.961Physical Impact0.963
13Handicap0.904Physical Impact0.889
14Handicap0.961Physical Impact0.947
Model Fit Indexes
Chi-Square115.364 115.276
Df56 74
NC2.060 1.558
RMSEA
(95% CI)
0.061
(0.045, 0.077)
0.044
(0.028, 0.059)
CFI0.990 0.993
TLI0.984 0.992
SRMR0.040 0.039
NC: normalized chi-square; RMSEA: root mean square error of approximation; CI: confidence interval; CFI: comparative fit index; TLI: Tucker–Lewis index; SRMR: standardized root mean square residual.
Table 5. Comparisons of the relationship of overall oral health with dimension scores from OHIP-14 versus MOHIP-14 in pregnant women (MOHIP-14PW).
Table 5. Comparisons of the relationship of overall oral health with dimension scores from OHIP-14 versus MOHIP-14 in pregnant women (MOHIP-14PW).
Original: OHIP-14Modified: MOHIP-14PW
Dimension NameSpearman Correlation with Overall Oral HealthDimension NameSpearman Correlation with Overall Oral Health
Functional Limitation0.13Physical Impact0.23
Physical Pain0.21Psychological Impact0.29
Psychological Discomfort0.23Pain Impact0.21
Physical Disability0.16
Psychological Disability0.28
Social Disability0.16
Handicap0.19
Average0.19Average0.24
Range(0.13, 0.28)Range(0.21, 0.29)
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Yang, C.; Huang, S.S.; Moore Simas, T.A.; Silk, H.; Savageau, J.A.; Russell, S.L. The MOHIP-14PW (Modified Oral Health Impact Profile 14-Item Version for Pregnant Women): A Real-World Study of Its Psychometric Properties and Relationship with Patient-Reported Oral Health. Healthcare 2022, 10, 461. https://doi.org/10.3390/healthcare10030461

AMA Style

Yang C, Huang SS, Moore Simas TA, Silk H, Savageau JA, Russell SL. The MOHIP-14PW (Modified Oral Health Impact Profile 14-Item Version for Pregnant Women): A Real-World Study of Its Psychometric Properties and Relationship with Patient-Reported Oral Health. Healthcare. 2022; 10(3):461. https://doi.org/10.3390/healthcare10030461

Chicago/Turabian Style

Yang, Chengwu, Shulamite S. Huang, Tiffany A. Moore Simas, Hugh Silk, Judith A. Savageau, and Stefanie L. Russell. 2022. "The MOHIP-14PW (Modified Oral Health Impact Profile 14-Item Version for Pregnant Women): A Real-World Study of Its Psychometric Properties and Relationship with Patient-Reported Oral Health" Healthcare 10, no. 3: 461. https://doi.org/10.3390/healthcare10030461

APA Style

Yang, C., Huang, S. S., Moore Simas, T. A., Silk, H., Savageau, J. A., & Russell, S. L. (2022). The MOHIP-14PW (Modified Oral Health Impact Profile 14-Item Version for Pregnant Women): A Real-World Study of Its Psychometric Properties and Relationship with Patient-Reported Oral Health. Healthcare, 10(3), 461. https://doi.org/10.3390/healthcare10030461

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop