Identifying the unseen: cross-cultural adaptation and psychometric properties of the persian dental neglect scale (P-DNS) to uncover oral health neglect in Iranian seniors

Abstract

Background

Good oral health is vital for quality of life, yet many older adults neglect their oral hygiene and dental visits. As no specific tool adapted to the Iranian cultural and socioeconomic context exists to assess oral health neglect in this population, this study aimed to evaluate the psychometric properties and cross-cultural adaptation of the Persian version of the Dental Neglect Scale (P-DNS) for older adults.

Methods

This methodological study used convenience sampling at Tooba Clinic in Sari, Iran, from May to July 2024. Data were collected via face-to-face interviews using the P-DNS, the Geriatric Oral Health Assessment Index (GOHAI), the Decayed, Missing, and Filled Teeth (DMFT) index, the Gingival Index (GI), and a demographic questionnaire. P-DNS translation and cultural adaptation followed Beaton et al.‘s guidelines. Psychometric properties were analyzed using SPSS v.25.0, with statistical significance set at p < 0.05.

Results

The study included 200 participants (mean age: 65.5 ± 5.7 years). Mean (± SD) scores were: P-DNS 19.5 ± 5.5, GOHAI 23.5 ± 5.1, DMFT 23.2 ± 8.3, and GI 1.43 ± 0.45. Experts confirmed face, content (scale-level CVI = 0.89), and cross-cultural validity. Confirmatory factor analysis indicated a good model fit (GFI = 0.99, CFI = 1). Convergent validity was supported by a significant positive correlation between P-DNS and GOHAI (r = 0.57, p < 0.05). Divergent validity was supported by significant negative correlations between P-DNS and the DMFT (r = − 0.42, p < 0.05) and GI (r = − 0.46, p < 0.05) indices. Internal consistency (Cronbach’s α = 0.91) and test-retest reliability (ICC = 0.92) were high. Using ROC curve analysis with the GOHAI as a criterion, an optimal cut-off score of 13.5 was established (AUC = 0.82), yielding 88.1% sensitivity and 65.5% specificity. No significant floor (2%) or ceiling (2.5%) effects were observed. Higher P-DNS scores (indicating less neglect) were positively associated with higher education, better perceived economic and oral health status, having insurance, and more frequent dental visits and brushing (p < 0.05).

Conclusion

The P-DNS demonstrated high validity and reliability, making it a valuable tool for assessing dental neglect among older adults in Iran. However, the findings’ generalizability may be limited due to the use of convenience sampling.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12877-025-06459-7.

Introduction

As the global population ages, researchers and policymakers are paying greater attention to its associated challenges. These include the management of multimorbidity, functional decline, and increased socioeconomic vulnerability. Iran is similarly undergoing a demographic transition, shifting from a predominantly young population to an increasingly older one [1].

The oral health of many older adults is often compromised by a progressive decline in physical health and the cumulative effects of untreated dental conditions over time [2]. Furthermore, the high prevalence of chronic diseases in this demographic means that both degenerative changes and the management of these conditions can adversely affect oral health.

Although good oral health is a crucial indicator of health-related quality of life, many older adults neglect their oral hygiene and dental visits. Dental neglect (DN) is defined as an individual’s failure to take preventive measures to maintain oral health, seek necessary dental care, and regard the importance of the teeth and related tissues [3]. DN often leads to short-term complications such as cavities, pain, and infections in the oral cavity [4]. However, these immediate effects can progress to long-term complications, including tooth loss, difficulties with speech and chewing, intraoral abscesses, systemic infections, and weight loss, all of which ultimately decrease an individual’s quality of life [5, 6]. For instance, the consequences of such neglect are not isolated; a strong bidirectional relationship exists between poor oral health, particularly periodontal disease, and diabetes. Chronic oral inflammation resulting from neglect can worsen glycemic control in diabetic patients, demonstrating how oral health is integral to managing systemic diseases [7].

For decades, the Decayed, Missing, and Filled Teeth (DMFT) index and the Gingival Index (GI) have been cornerstone tools in evaluating oral health within clinical and epidemiological studies. Both indices leverage the advantages of direct visual-tactile examination: the DMFT provides a clinical assessment of decayed, missing, and filled teeth, while the GI quantifies gingival inflammation. Notably, numerous studies consistently show that older adults tend to present with higher mean DMFT and GI scores [8, 9]. Furthermore, recent research has increasingly highlighted the intricate relationship between various oral health issues, particularly tooth loss, and cognitive decline [10].

Therefore, it is essential to identify older adults who neglect oral hygiene and regular dental visits, as an individual’s overall health and well-being are closely linked to their oral health. Simply put, good oral health allows an individual to speak freely, eat comfortably, and engage in social interactions without discomfort or embarrassment. The literature indicates that an individual’s self-assessment of their oral health is a key factor that affects their use of dental care services, highlighting the need for validated self-report measures [11].

One of the most frequently cited scales for assessing this construct is the Dental Neglect Scale (DNS), which comprises six items rated on a five-point scale [3]. The DNS has been translated and validated in multiple languages, including Romanian [12] and Norwegian [13]. Its key feature is the ability to provide valuable information about individuals who are avoiding dental care and, consequently, not attending clinical screenings.

Our comprehensive literature search confirmed that no specific validated instrument exists in the Persian language for measuring dental neglect. Addressing this gap served as the primary motivation for the current study. Given the significant impact of oral health on the quality of life of older adults, this study aimed to examine the psychometric properties of the Persian version of the Dental Neglect Scale (P-DNS) among Iranian older adults.

Materials and methods

Study design

This cross-sectional methodological study was conducted in two phases: cross-cultural adaptation and psychometric evaluation. The cross-cultural adaptation followed Beaton et al.‘s six-step guideline [14], ensuring linguistic and cultural validity. For the psychometric evaluation, the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) framework was employed [15, 16].

Sample size and setting

This study was conducted from May to July 2024 at Tooba Clinic, a dental clinic in Sari, Iran. According to the rule-of-thumb for sample size [17], at least 10 to 15 participants per item are recommended [18]. Given that the DNS consisted of 6 items, a sample of 60 to 90 participants would have been sufficient; however, we recruited a total of 200 older adults to achieve more reliable results.

Inclusion and exclusion criteria

The study’s inclusion criteria were: being 60 years of age or older; exhibiting normal cognitive function (a score of 7 or higher on the AMT questionnaire); being capable of communicating in Persian; and expressing a willingness to participate. This threshold was chosen as a score of 6 or lower was considered indicative of cognitive impairment in the validated Persian version of the AMT [19].

Exclusion criteria included any physical or medical condition that prevented the clinical oral examination (e.g., severe trismus or uncontrolled movements), any communication disorder that would significantly impede the interview process (e.g., severe aphasia or hearing loss), as well as incomplete surveys.

Participants and data collection

Data were gathered using the convenience sampling method. Participants were selected from older adults who visited Tooba Dental Clinic and met the inclusion criteria. A dentist specializing in oral care, who received training prior to the study, recruited the 200 participants. This trained dentist conducted all face-to-face interviews and performed the clinical oral examinations to determine the DMFT and Gingival Index scores. Each interview and examination took approximately 20 min. For participants unable to read or write, the questionnaire was read aloud and filled out by the data collector (the trained dentist). The data collector ensured that all items were answered during the interview, and sampling was continued until 200 complete questionnaires were obtained; therefore, there were no missing data in the final analysis.

Measures

Dental neglect scale (DNS)

The DNS is a 6-item assessment tool designed to quantify the level of oral health neglect. It helped professionals identify individuals who may struggle with or disregard proper dental care. Each item was rated on a 5-point scale (1 = Definitely No to 5 = Definitely Yes), with total scores ranging from 6 to 30. A lower score signified greater dental neglect [3]. The Persian version of the DNS, used for the first time in this study, can be found in the Supplementary Materials.

The oral health assessment index (GOHAI)

We used the GOHAI to assess concurrent validity. The 12 items of the index assessed oral health problems across three dimensions: physical function, psychosocial function, and pain or discomfort. Responses were scored on a 6-point Likert scale, with higher scores indicating better oral health-related quality of life. The Persian version of the index has excellent reliability and validity [20].

The DMFT index

This index was used to assess the number of decayed, missing, and filled teeth. For this study, decayed teeth referred to the number of teeth that had caries. Missing teeth were those extracted due to decay or other factors, and filled teeth were those that had been treated for caries.

The gingival index (GI)

The GI, developed by Silness and Löe [21], is a standardized measure of gingival inflammation. It assessed four surfaces per tooth on a scale from 0 (healthy) to 3 (severe inflammation). Scores of 1 and 2 indicated mild and moderate inflammation, respectively. The final score was calculated as the mean of all site scores [22].

Abbreviated mental test (AMT)

Participants were required to complete this questionnaire to meet the inclusion criteria. The AMT, a 10-item tool, swiftly evaluated older individuals for potential cognitive impairments. The total score ranged from 0 to 10, with a score of 6 or lower suggesting cognitive impairment [19]. The Persian version of the AMT is valid and reliable in Iran [19].

Medical and sociodemographic characteristics

The participants’ demographic and medical information were collected, including age, gender, living arrangements, marital status, educational level, chronic diseases, occupation, perceived economic status, smoking status, tooth brushing frequency, insurance status, denture quality, dental visiting patterns, loneliness, life satisfaction, gum health, and self-rated oral health status.

DNS cross-cultural adaptation

Before commencing the translation process, permission was formally obtained from the original author of the questionnaire. Following this step, the translation and cultural adaptation procedures were carried out in accordance with the established guidelines by Beaton et al. [14], ensuring that both the linguistic accuracy and cultural relevance of the content were maintained. The steps involved in the translation process are outlined in detail as follows:

• Step 1. Preliminary Translation: Two bilingual translators with expertise in both English and Persian, as well as knowledge of the medical field, independently translated the questionnaire. This step ensured that both linguistic and cultural nuances were preserved.

• Step 2. Merging of Translations: The two independent translations were reviewed by the translators and research team to assess accuracy and cultural appropriateness. A single unified version (Version 1) was developed by combining the best elements of both translations.

• Step 3. Back Translation: Two independent translators, who were fluent in both languages but not specialists in the medical field, translated Version 1 back into English. They worked without access to the original questionnaire to ensure objectivity. This step helped identify discrepancies and ensure that the translated content retained the original meaning.

• Step 4. Expert Panel Review: A panel of professionals, including medical experts, geriatric specialists, and experienced translators, reviewed the translated and back-translated versions. They assessed linguistic accuracy, cultural relevance, and medical appropriateness, leading to the refinement of the initial Persian version.

• Step 5. Testing the Initial Version: The Persian version was tested with 20 doctoral-level volunteers, including faculty members and doctoral students. They provided feedback on clarity, comprehensibility, and any ambiguities they encountered. Their insights helped refine the questionnaire further.

• Step 6. Finalization: Based on expert feedback, necessary modifications were made. The final Persian version was reviewed to ensure it accurately reflected the original English questionnaire while being culturally suitable for Persian-speaking individuals.

This meticulous process ensured that the translation was not only linguistically accurate but also culturally adapted, making the questionnaire suitable for Persian-speaking individuals.

Statistical analysis

Descriptive data were analyzed using means, medians, standard deviations, and frequencies. The P-DNS score distribution was assessed for normality. Since the normality assumption was not met, non-parametric tests were chosen for group comparisons (Mann-Whitney U and Kruskal-Wallis H tests). Spearman’s rank correlation coefficient was used for analyzing relationships between quantitative variables. Generalized Linear Models were utilized for multivariate analysis, including variables with a p-value < 0.2 from univariate models. All analyses were conducted using SPSS version 25.0, with statistical significance set at p < 0.05.

Face validity

Face validity referred to the extent to which the instrument appeared appropriate for its purpose, as judged by respondents. After the P-DNS was approved, 30 older adults examined it for ambiguity and difficulty. Each item’s Impact Score (IS) was then calculated, and items with an IS of 1.5 or higher were retained for further analysis [23].

Content validity

To ensure content validity, both qualitative and quantitative methods were used. Qualitatively, the items were reviewed by 10 experts (5 dental specialists and 5 gerontology specialists). Quantitatively, the Content Validity Index (CVI) and Content Validity Ratio (CVR) were calculated.

CVR

To determine the CVR, 10 experts independently assessed the necessity of each item on a three-point Likert scale. Based on the Lawshe table for a panel of 10 experts, any item with a CVR value lower than 0.62 was considered inappropriate [24, 25].

CVI

To determine the CVI, the experts assessed the relevance, clarity, and simplicity of each item. Values greater than 0.79 for the item-level CVI and 0.80 for the scale-level CVI were deemed acceptable [26].

Construct validity

Confirmatory factor analysis

Confirmatory Factor Analysis (CFA) is the statistical method used to assess construct validity and confirm the questionnaire’s factor structure. Its primary goal was to evaluate the fit between the theoretical model and the empirical data. AMOS software, version 24, was used for this purpose.

Convergent validity

Convergent validity was assessed by examining the correlation between the P-DNS questionnaire and the GOHAI questionnaire. The GOHAI was chosen as a relevant measure of oral health-related quality of life, a construct expected to be related to dental neglect.

Divergent validity

Divergent validity was evaluated by analyzing the correlations between the P-DNS questionnaire and the DMFT and GI indices. These clinical indices were selected as measures of dental caries experience and gingival inflammation, respectively, representing distinct constructs from dental neglect.

Reliability

Internal consistency

Internal consistency was defined as the extent to which scale items measure the same construct. It assessed the degree of interrelatedness among the items. Cronbach’s alpha was the method used for evaluation. A value closer to 1 indicated a higher level of internal consistency, whereas a value below 0.70 might indicate that the scale required revision [27].

Stability (test-retest reliability)

Stability referred to the tool’s ability to consistently generate comparable outcomes. A reliable instrument guaranteed that its measurements were not affected by random errors. To determine stability, the P-DNS was re-administered to 30 participants after a two-week interval, and reliability was assessed using the intraclass correlation coefficient. A value exceeding 0.80 signified adequate reliability [28].

Measurement error

The Standard Error of Measurement (SEM) was calculated to estimate the amount of random error in individual P-DNS scores.

Assessment of score distribution (floor and ceiling effects)

This effect was considered to occur when more than 15% of respondents achieved either the highest or lowest possible score. The presence of these effects indicates that the questionnaire may have been unable to distinguish individuals at the highest or lowest levels of the construct [29].

To provide a clear understanding of the study’s methodology, Fig. 1 illustrates the comprehensive validation process of the P-DNS.

Fig. 1.

The comprehensive validation process for the P-DNS

Results

The study included 200 participants, with a near-even gender distribution (49% male, 51% female). The mean age of participants was 65.5 ± 5.7 years, with ages ranging from 60 to 81. The majority of participants were married (88.5%), and 59.5% reported their economic status as average. Key oral health indicators were assessed, revealing mean scores of 19.5 ± 5.5 for P-DNS, 23.5 ± 5.1 for GOHAI, 23.2 ± 8.3 for the DMFT index, and 1.43 ± 0.45 for the GI index. Additional demographic and health-related factors, along with their influence on P-DNS scores, are presented in Table 1.

Table 1.

Sociodemographic and medical characteristics of the participants

Variable		N (%)	Mean ± SD	Test statistics	P-value
The P-DNS		200 (100)	19.5 ± 5.5	-	-
The GOHAI		200 (100)	23.5 ± 5.1	0.573a	0.025
The DMFT index		200 (100)	23.2 ± 8.3	−0.525a	0.015
The GI index		200 (100)	1.43 ± 0.45	−0.610a	0.002
Age	60	45 (22.5)	18.69 ± 5.05	1.86c	0.395
	61–69	114 (57)	19.83 ± 5.63
	≤ 70	41 (20.5)	19.59 ± 5.45
Gender	Male	98 (49)	19.95 ± 5.89	−1.546b	0.122
	Female	102 (51)	19.12 ± 5.01
Educational level	Elementary school	42 (21)	14.83 ± 4.87	37.634c	< 0.001
	Middle school	17 (8.5)	18.88 ± 4.85
	High school	81 (40.5)	20.54 ± 5.03
	Academic degree	60 (30)	21.62 ± 4.66
Marital status	Single	23 (11.5)	19.04 ± 4.69	−0.432b	0.666
	Married	177 (88.5)	19.59 ± 5.56
Occupation	Housewife	24 (12)	19.67 ± 5.23	13.266c	0.001
	Unemployed	103 (51.5)	20.82 ± 5.43
	Self-employed	73 (36.5)	17.66 ± 5.09
Chronic disease	Yes	118 (59)	20.15 ± 5.2	−1.644b	0.100
	No	82 (41)	18.62 ± 5.72
Smoking	Yes	54 (27)	18.33 ± 5.37	−1.776b	0.076
	No	146 (73)	19.97 ± 5.44
Living arrangements	With children	7 (3.5)	13.43 ± 0.53	12.281c	0.006
	With spouse	57 (28.5)	19.21 ± 4.54
	With family (children and spouse)	120 (60)	19.77 ± 5.99
	Alone	16 (8)	21.5 ± 3.31
Perceived economic status	Good	52 (26)	21.23 ± 4.01	7.735c	0.021
	Average	119 (59.5)	19.25 ± 5.47
	Poor	29 (14.5)	17.59 ± 6.85
Loneliness	Yes	57 (28.5)	18.72 ± 5.07	−0.966b	0.334
	No	143 (71.5)	19.85 ± 5.59
Life satisfaction	Yes	135 (67.5)	19.57 ± 5.56	−0.064b	0.949
	No	65 (32.5)	19.43 ± 5.28
Insurance	Yes	160 (80)	20.05 ± 5.37	−2.839b	0.005
	No	40 (20)	17.43 ± 5.39
Oral health status	Good	44 (22)	25.43 ± 2.51	83.202c	< 0.001
	Poor	74 (37)	16.14 ± 4.08
	No opinion	82 (41)	19.41 ± 5.04
Dental visiting pattern	Never	81 (40.5)	16.1 ± 4.89	66.983c	< 0.001
	Once a year	96 (48)	20.96 ± 4.45
	Regularly	23 (11.5)	25.61 ± 2.54
Gum health	Healthy	104 (52)	19.83 ± 6.1	−0.8b	0.424
	Swollen or bleeding	96 (48)	19.2 ± 4.68
Denture quality	Good	45 (22.5)	18.89 ± 5.12	2.756c	0.097
	Poor	37 (18.5)	16.81 ± 3.14
	No denture	118 (59)	20.62 ± 5.84
Frequency of tooth brushing	Never	51 (25.5)	13.86 ± 3.07	55.349c	< 0.001
	Once a day	99 (49.5)	19.58 ± 3.91
	Two times or more	50 (25)	25.2 ± 3.82

P-DNS Persian Dental Neglect Scale, GOHAI Oral Health Assessment Index, DMFT index Decayed, Missing, and Filled Teeth, GI index Gingival Index

^aSpearman’s rank correlation

^bMann-Whitney U

^cKruskal-Wallis H

Demographic and health-related factors significantly influenced P-DNS scores. Educational attainment (P < 0.001) showed a strong correlation, with higher education levels leading to increased P-DNS scores (e.g., academic degree: 21.62 ± 4.66 vs. elementary school: 14.83 ± 4.87). Occupation (P = 0.001) also played a role, with unemployed individuals having a mean score of 20.82 ± 5.43, while self-employed individuals had a lower mean of 17.66 ± 5.09. Perceived economic status (P = 0.021) demonstrated a positive trend, with good perceived status yielding higher scores (21.23 ± 4.01) compared to poor status (17.59 ± 6.85). Insurance coverage (P = 0.005) was also a significant factor, with insured individuals reporting higher scores (20.05 ± 5.37) than uninsured individuals (17.43 ± 5.39).

Oral health status (P < 0.001), dental visiting patterns (P < 0.001), and frequency of tooth brushing (P < 0.001) were strongly associated with P-DNS scores. Participants reporting good oral health had markedly higher scores (25.43 ± 2.51) than those with poor oral health (16.14 ± 4.08). Similarly, regular dental visits (25.61 ± 2.54) and more frequent tooth brushing (e.g., ≥ 2 times/day: 25.2 ± 3.82) were linked to significantly higher scores compared to never visiting the dentist (16.1 ± 4.89) or never brushing (13.86 ± 3.07). Living arrangements also significantly impacted P-DNS scores (P = 0.006), with individuals living alone showing a higher mean score (21.5 ± 3.31) compared to those living with children (13.43 ± 0.53). Finally, GOHAI, DMFT, and GI indices demonstrated significant inverse correlations with P-DNS scores (P = 0.025, P = 0.015, and P = 0.002 respectively), as shown by Spearman’s rank correlation coefficient.

Generalized Linear Models were utilized to assess the simultaneous association between study variables and P-DNS scores. Variables showing a p-value less than 0.2 in initial univariate analyses were included in the multiple regression model; “occupation” was excluded due to high collinearity with “educational level”. Table 2 presents these independent effects on P-DNS scores.

Table 2.

Results of the generalized linear model examining the simultaneous association of study variables with P-DNS

Variable		OR (95% CI)	Wald statistic	p-value
Educational level	Elementary school	0.725 (0.675,0.779)	78.083	< 0.001
	Middle school	0.865 (0.794,0.943)	10.875	0.001
	High school	0.932 (0.883,0.984)	6.467	0.011
	Academic degree
Chronic disease	Yes	1.031 (0.983,1.082)	1.623	0.203
	No
Smoking	Yes	1.128 (1.068,1.191)	18.814	< 0.001
	No
Living arrangements	With spouse	0.995 (0.906,1.094)	0.01	0.920
	With family (children and spouse)	0.988 (0.895,1.091)	0.057	0.812
	Alone
Perceived economic status	Good	1.097 (1.012,1.19)	5.073	0.024
	Average	1.019 (0.945,1.098)	0.239	0.625
	Poor
Insurance	Yes	1.068 (1.01,1.13)	5.329	0.021
	No
Oral health status	Good	1.025 (0.959,1.095)	0.535	0.464
	Poor	0.933 (0.887,0.98)	7.582	0.006
	No opinion
Dental visiting pattern	Never	0.841 (0.77,0.918)	14.982	< 0.001
	Once a year	0.983 (0.912,1.06)	0.2	0.655
	Regularly
Gum health	Healthy	1.026 (0.983,1.071)	1.377	0.241
	Swollen or bleeding
GOHAI	-	1.05 (1.03, 1.07)	35.00	< 0.001
DMFT	-	0.97 (0.96, 0.99)	15.00	0.001
GI	-	0.92 (0.89, 0.95)	28.00	< 0.001

GOHAI Oral Health Assessment Index, DMFT index Decayed, Missing, and Filled Teeth, GI index Gingival Index

The multivariate Generalized Linear Model revealed several independent factors significantly associated with P-DNS scores. Lower educational attainment (e.g., elementary school: OR = 0.725, P < 0.001) was associated with lower P-DNS scores, indicating lower odds of higher scores compared to those with an academic degree. Conversely, smoking (OR = 1.128, P < 0.001), good perceived economic status (OR = 1.097, P = 0.024), and having insurance (OR = 1.068, P = 0.021) were positively associated with P-DNS scores.

Oral health indicators and behaviors also played a significant role. Poor oral health status (OR = 0.933, P = 0.006) and never visiting a dentist (OR = 0.841, P < 0.001) were inversely associated with P-DNS scores. Conversely, increased frequency of tooth brushing (OR = 1.229, P < 0.001) showed a positive association. Additionally, GOHAI (OR = 1.05, P < 0.001) demonstrated a significant positive association with P-DNS scores, while DMFT (OR = 0.97, P = 0.001) and GI (OR = 0.92, P < 0.001) indices showed significant inverse associations with P-DNS scores.

Face and content validity

The face validity of the P-DNS was quantitatively assessed using the item impact score. The results showed that all items had scores greater than 1.5; consequently, all items were retained. Following an expert evaluation for qualitative content validity, the items were approved with only minor revisions. The CVR for each item exceeded 0.77, suggesting the questionnaire possessed adequate content validity. Additionally, the CVI was used to assess the relevance, clarity, and simplicity of the items. The item-level CVI for each item was greater than 0.84, and the overall scale-level CVI (S-CVI) was 0.89, indicating an adequate level of relevance.

Construct validity (confirmatory factor analysis)

The structure of the DNS questionnaire was defined based on a single dimension. To confirm this structure and assess construct validity, CFA was employed. Given the non-normal distribution of the data, the Diagonally Weighted Least Squares (DWLS) estimator, a robust method against non-normality, was utilized for parameter estimation.

Initially, the proposed P-DNS model did not demonstrate acceptable fit indices. The initial model showed χ2/df = 13.78, RMSEA = 0.25 (90% CI: 0.21, 0.29), GFI = 0.83, and P < 0.001. Following structural modifications, the final model achieved a statistically acceptable fit. The final values of the fit indices are reported in Table 3, and the final model with standardized coefficients is presented in Fig. 2.

Table 3.

Goodness-of-fit indices for the P-DNS model

Step	Model	/df	CFI	GFI	RMSEA (90%CI)	SRMR	P
	Recommended value	< 5	≥ 0.9	≥ 0.9	< 0.1	< 0.1	> 0.05
1	Initial model	13.78	0.92	0.83	0.25 (0.21, 0.29)	0.067	< 0.001
2	Final model (structural model)	1.14	1	0.99	0.024 (0.0, 0.10)	0.018	0.351

/df Chi-square test, CFI Comparative fit index, GFI Goodness-of-fit index, RMSEA Root mean square error of approximation, 90%CI 90% Confidence Interval, SRMR Standardized root mean square residual, P-DNS Persian version of the Dental Neglect Scale Questionnaire

Fig. 2.

Single-factor model of the P-DNS with the standardized factor loading. DNS: Dental Neglect Scale

As shown in Table 4, all questionnaire items were directly and significantly associated with DNS, based on their regression coefficients. Standardized coefficients indicated that items two and four were the most influential, while item three was the least influential in the DNS construct.

Table 4.

Regression coefficients of the paths

Path	Regression coefficient	Standard error	t-value	P value	Standardized regression coefficient
P-DNS1<--- P-DNS	1.000				0.85
P-DNS2<--- P-DNS	1.05	0.048	21.79	< 0.001	0.90
P-DNS3<--- P-DNS	0.76	0.072	10.60	< 0.001	0.65
P-DNS4<--- P-DNS	1.10	0.042	26.28	< 0.001	0.89
P-DNS5<--- P-DNS	0.55	0.041	13.46	< 0.001	0.75
P-DNS6<--- P-DNS	0.66	0.050	13.29	< 0.001	0.77

P-DNS Persian version of the Dental Neglect Scale

Construct validity (convergent and divergent validity)

Convergent validity was assessed by examining the correlation between the P-DNS and the GOHAI (Table 5). The GOHAI was selected as a validated measure of oral health-related quality of life, a construct theoretically associated with dental neglect. As anticipated, the P-DNS showed a significant positive correlation with the GOHAI (r = 0.57, p < 0.05), supporting its convergent validity. This finding indicated that individuals with lower levels of dental neglect (higher P-DNS scores) reported better oral health-related quality of life (higher GOHAI scores).

Table 5.

A correlation matrix of the study’s main variables

Scale	P-DNS	GOHAI	GI index	DMFT index
DMFT index	−0.42	−0.51	0.77	1
GI index	−0.46	−0.59	1	0.77
GOHAI	0.57	1	−0.59	−0.51
P-DNS	1	0.57	−0.46	−0.42

Correlation is significant at p < 0.05

Divergent validity was evaluated by analyzing the correlations between the P-DNS and two clinical indices: the DMFT and the GI. These indices were chosen because they measure clinical oral health outcomes that are conceptually distinct from the behavioral construct of dental neglect. The results demonstrated significant negative correlations between the P-DNS and both the DMFT (r = −0.42, p < 0.05) and GI (r = −0.46, p < 0.05) indices. Specifically, lower P-DNS scores (indicating greater neglect) were associated with higher levels of dental caries experience and gingival inflammation. These findings suggested that the P-DNS successfully measures the behaviors and attitudes that can lead to poor oral health (the cause), while the DMFT and GI indices measure the physical consequences of those behaviors (the effect).

Reliability

The internal consistency of the P-DNS was evaluated using Cronbach’s alpha. The resulting value was 0.91, indicating excellent internal consistency. Stability was assessed via the test-retest method, in which thirty participants were re-interviewed after a two-week interval. The ICC between the two administrations was 0.92 (95% CI: 0.87–0.95, p < 0.01). Additionally, the SEM for the P-DNS score was approximately 1.56. These results confirmed the strong reliability of the P-DNS.

Floor and ceiling effects

The score distribution was examined for floor and ceiling effects. In total, 2% of participants scored the lowest possible score, while 2.5% achieved the highest. Since both percentages were well below the 15% threshold, no significant floor or ceiling effects were observed for the P-DNS.

Cut-off point

The cut-off point for the P-DNS was determined using a ROC curve analysis with the GOHAI as the criterion. The optimal cut-off score was 13.5, which yielded a sensitivity of 88.1% and a specificity of 65.5%, with an Area Under the Curve (AUC) of 0.82 (95% CI: 0.75–0.89). The Kappa agreement coefficient between the P-DNS and GOHAI was 0.43, indicating moderate agreement. Based on this cut-off point, 177 participants (88.5%) were categorized as having dental neglect.

Discussion

This study successfully adapted and validated the P-DNS among Iranian older adults, demonstrating its robust psychometric properties. The findings confirm the P-DNS as a reliable and valid instrument for assessing oral health neglect, contributing a crucial self-report measure where none previously existed in Iran.

The determined cut-off point of 13.5 for the P-DNS, with its high sensitivity (88.1%), is particularly valuable for screening purposes. This high sensitivity means the tool is effective at identifying older adults who likely have dental neglect, minimizing false negatives. While the specificity (65.5%) is moderate, the overall utility of this cut-off is high for initial screening in clinical and public health settings. In practice, the moderate specificity indicates that some individuals without neglect may be flagged as at-risk (false positives); therefore, individuals who screen positive would require a brief secondary assessment (e.g., a clinical interview) to confirm the finding before an intervention is planned.

The application of this cut-off revealed that a substantial proportion of participants (88.5%) were categorized as having dental neglect, underscoring the high prevalence of this issue among Iranian older adults in this study sample. This high prevalence becomes more striking when compared to international data; for instance, the mean score for Iranian seniors (19.5 ± 5.5) differs considerably from that of a younger Romanian population (25.29 ± 3.59) [12] and Norwegian adults (10.6 ± 3.3) [13]. These differences likely reflect national contexts, as the Norwegian authors suggested their population’s score might stem from a well-organized dental care system where neglect is a less relevant concept. Conversely, the high prevalence in Iran highlights the issue’s public health significance and may be driven by systemic financial barriers, such as significant out-of-pocket costs for dental care, and cultural attitudes, like a belief that tooth loss is an inevitable part of aging, which reduces the perceived importance of prevention.

Cross-cultural adaptation and psychometric properties of the DNS

The present study followed the standardized procedure proposed by Beaton et al. for the cross-cultural translation and adaptation of the P-DNS [14]. This approach encompassed both linguistic translation and cultural modification of items, where needed, to enhance their contextual relevance and ensure conceptual equivalence. This methodology has previously demonstrated its effectiveness in similar research conducted in Iran [30, 31].

Regarding face and content validity, our study employed rigorous quantitative and qualitative methods. While other validation studies, such as the Norwegian [13] and Romanian [12] versions, mentioned aspects related to face validity, detailed quantitative reports on item IS, CVR, or CVI are not consistently available in their published articles. This highlights the comprehensive nature of our validation process.

Given that the DNS was developed based on a predefined theoretical framework with a single domain, an Exploratory Factor Analysis (EFA) was not conducted. Instead, the CFA supported the hypothesized single-factor structure, consistent with the Norwegian study that extracted one factor explaining 37% of the total scale variance [13]. The original DNS development also confirmed a latent dental neglect variable accounting for 43.6% of the variance [3]. Although the initial P-DNS model’s fit was poor, structural modifications led to excellent fit indices (χ2/df = 1.14, CFI = 1.00, GFI = 0.99, RMSEA = 0.024, SRMR = 0.018, p = 0.351). These indices are notably superior to those reported in the Romanian study’s CFA (χ2/df = 1.13, CFI = 0.99, RMSEA = 0.017, SRMR = 0.059) [12], with our lower RMSEA and SRMR values indicating a strong model fit. This robust factor structure allows for confident interpretation of the total P-DNS score as a measure of a unified construct.

The assessment of construct validity yielded compelling results. Convergent validity was supported by the significant positive correlation between P-DNS scores and GOHAI scores (r = 0.57, p < 0.05). This aligns with theoretical expectations, as individuals with less dental neglect (higher P-DNS scores) logically experience better oral health-related quality of life (higher GOHAI scores). The Romanian study also found a positive association between their DNS version and the Oral Health Values Scale (r = 0.37, p < 0.001) [12].

Divergent validity was effectively demonstrated by the significant negative correlations of the P-DNS with both the DMFT (r = − 0.42, p < 0.05) and GI (r = − 0.46, p < 0.05) indices. The negative correlation, where higher P-DNS scores (less neglect) correspond to better clinical indices (lower DMFT and GI scores), reinforces the clinical relevance of assessing dental neglect as a precursor to observable oral health problems. This differentiation is crucial, as the P-DNS provides complementary information on the underlying behaviors leading to the clinical conditions measured by the DMFT and GI. The Romanian study also reported negative associations between DNS and other instruments, including the Oral Health Impact Profile-14 (measuring the impact of oral health on quality of life) and the Dental Indifference Scale (measuring a reduced need for oral care) [12].

The psychometric evaluation of the P-DNS revealed strong reliability. The internal consistency (Cronbach’s alpha = 0.91) was notably higher than that reported in the Norwegian (α = 0.57 in the general population) [13] and Romanian (α = 0.70) [12] validation studies. The stability of the P-DNS was further supported by a high 2-week test-retest ICC of 0.92, which also compares favorably to the test-retest reliability found in the Norwegian study (Spearman’s rho = 0.60) [13]. The calculated Standard Error of Measurement (SEM) of approximately 1.56 further reinforces the precision and stability of the P-DNS scores, suggesting that the instrument provides consistent measurements with minimal random error. The negligible floor (2%) and ceiling (2.5%) effects indicate that the P-DNS effectively captures the full spectrum of the construct in this population.

Factors affecting dental neglect

The significant associations identified in the multivariate analysis with demographic and health-related factors provide further insights. Educational level, socioeconomic status, and insurance coverage consistently influenced P-DNS scores. This highlights potential disparities in dental care access and practices, suggesting that those with lower education, poorer economic status, or lack of insurance may be at higher risk for dental neglect. Similarly, the strong links between P-DNS and self-reported oral health status, dental visiting patterns, and frequency of brushing reinforce the behavioral nature of the construct and emphasize the importance of promoting regular dental habits and professional care in this age group. The positive association with GOHAI and inverse associations with DMFT and GI in the multivariate model further emphasize the comprehensive role of dental neglect in overall oral health, extending beyond simple clinical indicators. Our findings are consistent with a study by Bhattarai et al. on a Nepalese adult population, which also established a significant link between higher dental neglect and worse clinical outcomes. Specifically, their research showed that increased neglect was significantly associated with a higher DMFT score (p = 0.022) and poorer oral hygiene (p = 0.001), directly supporting the inverse relationships we identified [32].

The Romanian study also found significant differences in DNS scores based on age and education level, with higher education correlating with higher scores, supporting the role of these factors in dental neglect [12]. Gender differences were also noted in the Romanian study, with females exhibiting greater care for oral health than males. However, in our study, this relationship was not observed. This discrepancy could be attributed to the differing cultural and socioeconomic contexts of the study populations.

Limitations and future directions

Despite its strengths, this study has several limitations. The cross-sectional design precludes the establishment of causal relationships, and the use of convenience sampling limits generalizability. While the sample size was sufficient for psychometric validation, a larger, population-based sample from diverse geographical regions would strengthen the findings. Furthermore, our single-center design did not allow for an exploration of key social determinants, such as urban-rural disparities, which may significantly influence neglect. The study’s reliance on self-reported data may also introduce recall or social desirability biases. As a self-report measure, the P-DNS captures perceived neglect, which may not always align perfectly with neglect determined by clinical examination. Future research should consider longitudinal designs, population-based sampling, and the integration of objective clinical assessments.

Conclusion

In conclusion, the P-DNS has demonstrated strong psychometric properties, confirming its reliability and validity as a valuable tool for assessing oral health neglect among Iranian older adults. Its ability to identify a significant proportion of at-risk individuals, combined with its established validity, supports its use in both clinical and research settings to better understand and address oral health challenges in this growing demographic.

Authors’ contributions

S.P., S.M.T. and H.N. were responsible for the initiation of the study’s conceptualization and design. S.A. gathered and obtained the data. A.F. and T.M. conducted the data analysis and provided an interpretation of the results. H.N. and N.H. ultimately authored the text. The paper has been read and approved by all the authors.

Funding

The author(s) stated that there was no financial support linked to the research presented in this publication.

Data availability

The corresponding author can provide the datasets created or analyzed during the current study upon reasonable request.

Declarations

Ethics approval and consent to participate

Since this study involves human subjects, we followed the principles described in the 1964 Helsinki Declaration and its subsequent revisions, or other comparable ethical standards. Participants were provided with assurances regarding the anonymity and confidentiality of their information. This commitment was emphasized to ensure that participants felt secure in sharing their responses, knowing that their identities and personal data would be protected throughout the research process. To ensure this, a coding system was employed, replacing each participant’s name with a unique code. All individuals provided written and verbal informed consent prior to data collection phase. This study was approved by the research ethics committee of Mazandaran University of Medical Sciences (ethics code: IR.MAZUMS.REC.1403.20070).

Competing interests

The authors declare no competing interests.

Conflict of interest

The author(s) did not report any potential conflict of interest.