Associations of nutritional status and oral health-related quality of life with early childhood caries of children aged 4–5 years: a hospital-based cross-sectional analytical study

Abstract

Background

Regular growth and development during early childhood can be hindered by different communicable diseases and chronic severe early childhood caries (S-ECC). Comprehensive studies on the factors associated with ECC, nutritional status, and oral health-related quality of life (OHRQoL) are scarce. Thus, this study aimed to investigate dietary, oral hygiene, and socio-demographic factors associated with these conditions and their impact on OHRQoL in a hospital-based cohort of Sri Lankan children.

Methods

This cross-sectional analytical study included 545 children aged 4–5 years attending the Pedodontic Clinic at the Dental Teaching Hospital, University of Peradeniya, Sri Lanka. After obtaining parental consent, data were collected using a pre-validated questionnaire, and oral examinations were conducted by three trained and calibrated dentists. Binary logistic regression analysis and negative binomial regression analysis were used to examine the effects of independent variables on nutritional status, ECC status, and OHRQoL.

Results

Within this clinical sample, higher S-ECC severity assessed via the defs index and the consumption of some sugary snacks and sweets were significantly associated with poor nutritional status (P < 0.05). Conversely, higher weight-for-age, BMI-for-age, and maternal education levels were significantly associated with a lower likelihood of S-ECC. Adverse dietary habits (frequent consumption of sweetened snacks), prolonged breastfeeding (> 2 years), bottle-feeding, and inadequate oral hygiene practices (infrequent or unsupervised brushing and use of non-fluoridated toothpaste) were identified as key behavioral indicators significantly associated with increased S-ECC (P < 0.05). Furthermore, lower nutritional status and higher defs indices were strongly associated with poorer OHRQoL scores (P < 0.001).

Conclusions

S-ECC and poor nutrition were prevalent among 4-5-year-olds, strongly associated with dietary habits, oral hygiene practices, and maternal education, while both conditions correlated with lower oral health-related quality of life. Importantly, higher ECC and poor nutrition were interrelated, highlighting the potential utility of integrated oral health and nutritional interventions in clinical pediatric settings.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s41043-026-01407-6.

Background

Early childhood caries (ECC) is identified as the presence of one or more decayed teeth, missing teeth due to caries, or filled tooth surfaces in any primary tooth in a child aged 72 months or younger. Severe early childhood caries (S-ECC) is identified when a child aged 4 years has ≥ 5 decayed, missing, or filled tooth surfaces, and a 5-year-old child has ≥ 6 such surfaces [1]. Serious complications of ECC include tooth sensitivity to thermal changes, pain, fever, dental abscesses, difficulty chewing, prolonged food retention, and impaired brushing [2]. Oral health-related quality of life (OHRQoL) is defined as “a multidimensional construct that reflects people’s comfort when eating, sleeping, and engaging in social interactions; their self-esteem; and their satisfaction concerning their oral health” [3]. The Early Childhood Oral Health Impact Scale (ECOHIS) is a validated tool that captures oral health problems and treatment experiences that may affect the daily lives and overall well-being of children and their families [4].

ECC is increasingly recognized as a multifactorial condition influenced by biological, behavioral, dietary, and socioeconomic determinants. Dietary practices such as frequent sugar consumption, inadequate nutritional intake, and poor oral hygiene behaviors contribute to ECC development, while socioeconomic disadvantage may further exacerbate both poor oral health and malnutrition [5]. At the same time, ECC itself may negatively affect children’s nutritional status through pain, difficulty chewing, and impaired eating behaviors. These consequences may ultimately interfere with growth and development [6]. In addition, ECC-associated issues adversely affect children’s and families’ oral health-related quality of life. Therefore, ECC, nutritional status, and OHRQoL can be considered interconnected dimensions of child well-being rather than isolated health outcomes [7, 8].

Borderline nutritional status and ECC remain major health issues in low- and middle-income countries [9, 10]. In Sri Lanka, 60.7% of five-year-old children have been reported to have active dental caries, of which 96.2% remain untreated [11]. At the same time, stunting, wasting, and underweight affect 17.3%, 15.1%, and 20.5% of children under five years of age, respectively [12]. Globally, ECC affects a substantial proportion of preschool children, with reported estimates ranging from 20% to over 70%, reflecting considerable variability across regions and socioeconomic groups [13]. Poor nutritional status, including underweight, stunting, and wasting, contributes to increased morbidity, delayed growth, and impaired cognitive development [14, 15]. Published research indicates that ECC not only directly affects children’s oral health but also has broader implications for their quality of life and nutritional well-being [7]. Systematic reviews and population studies consistently show that greater ECC severity is significantly associated with poor OHRQoL in preschool children, affecting functional, psychological, and social domains measured by validated tools such as the ECOHIS, and intensifying negative impacts as caries increases in severity [16, 17].

Despite growing recognition of these interrelationships, relatively few studies have examined ECC, multiple anthropometric indicators for nutritional status, OHRQoL, dietary factors, oral hygiene behaviors, and sociodemographic variables together in preschool children [8, 18–20], and such comprehensive evidence is particularly limited in Sri Lanka [7, 21–23]. Furthermore, while many studies have focused on the impact of poor nutritional status on ECC, limited evidence is available on how ECC may, in turn, influence the nutritional status of children [24–26]. This study offers a novel approach by simultaneously evaluating all of these interconnected health domains within a single cohort, providing the first comprehensive analysis of how dental caries, nutritional status, behavioral factors, and quality of life concurrently impact preschool children in Sri Lanka.

The gaps in the literature also highlight the importance of the present study. Therefore, we hypothesized that higher ECC status is significantly associated with poor nutritional status among Sri Lankan children aged 4–5 years, as assessed by height-for-age (HA), weight-for-age (WA), weight-for-height (WH), body mass index-for-age (BMIA), mid-upper arm circumference-for-age (MUACA), head circumference-for-age (HCA), and triceps skinfold thickness-for-age (TSFTA), with the corresponding null hypothesis stating that no such association exists. In addition, we hypothesized that higher ECC status is significantly associated with poorer OHRQoL in this age group, with null hypotheses indicating no association. Overall, the primary objective of this study was to investigate the complex associations among nutritional status, S-ECC, and OHRQoL among Sri Lankan children within an integrated framework. Furthermore, this study aimed to identify the behavioral, environmental, and structural determinants affecting these primary outcomes, focusing on how dietary habits, socio-economic factors, and oral hygiene practices concurrently influence child health within the study sample.

Methods

Selection of the sample

A cross-sectional analytical study design was employed. The study was conducted at the teaching hospital of the Faculty of Dental Sciences, University of Peradeniya, Sri Lanka. The study population comprised children aged 4–5 years with severe, moderate, or no ECC who attended the dental clinic during the study period.

Children were excluded if they were taking medications known to affect salivary flow; had diagnosed systemic, genetic, or metabolic diseases causing secondary malnutrition (such as celiac disease, congenital heart defects, Type 1 diabetes, or severe chronic infections); had psychological impairments; exhibited congenital dental anomalies; had erupted permanent teeth; or were uncooperative during the clinical examination. The required sample size for assessing the incidence of the disease/condition was calculated using the following Cochran’s formula [27].

where n is the sample size, Z is the standard normal deviation (1.96 for 95% CI), P is 68.8% [28], and D is 5%. An additional 10% was added to account for potential non-response. Since the study was conducted in a hospital-based setting rather than through population-based random sampling, a conservative design effect of 1.5 was applied during sample size calculation to account for potential homogeneity among participants attending the same institution and to minimize underestimation of the required sample size [29]. Therefore, the final sample size was 545 ([330 + 33] × 1.5). Furthermore, the final sample size was sufficient for regression analyses, as the number of outcome events relative to the number of predictors met established recommendations for events-per-variable (EPV ≥ 10), supporting the robustness of the fitted models [30].

A systematic sampling technique was used for sampling. Every other child attending the Dental Teaching Hospital, Faculty of Dental Sciences, University of Peradeniya, for routine check-ups or dental treatment and fulfilling the inclusion criteria was enrolled in this cross-sectional analytical study during the period of January 2022 to November 2022. The Faculty of Dental Sciences was selected as the study site because it manages nearly all ECC conditions and offers a wide range of treatment modalities. Therefore, only a limited number of case scenarios were likely to be missed at enrollment, and almost all relevant cases were expected to be captured. Children registered at the Dental Teaching Hospital were subsequently enrolled in the study. This study was not pre-registered with the journal, however the study was pre-registered at the Faculty of Dental Sciences, University of Peradeniya, Sri Lanka, as a Master’s study. The study protocol was developed prior to data collection, and all planned objectives and analyses were followed as described, with no changes to the primary research questions after the study commenced. This study was conducted and reported in accordance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines for observational studies. A STROBE flow diagram detailing participant recruitment, eligibility assessment, and final inclusion is provided in Fig. 1.

Fig. 1.

STROBE flow diagram of participant recruitment and inclusion in the study

Obtaining informed consent

This study was designed and conducted following ethics review committee guidelines, following the principles of the Declaration of Helsinki [31] and standards for professional conduct. The ethical clearance certificate (ERC/FDS/UOP/I/2021/45) was obtained from the Ethical Review Committee, Faculty of Dental Sciences, University of Peradeniya, on 12th October 2021. All information regarding the research was conveyed to parents or guardians. The consent was obtained only after parents or guardians had received and sufficiently understood all important information. Given that intraoral photographs were obtained, parents provided explicit, separate written consent authorizing the publication of clinical photographs in scientific journals. Additionally, formal child assent was obtained verbally from each preschooler before any clinical examination or photographic documentation, ensuring the child’s voluntary cooperation and comfort.

Clinical examination of the oral cavities

Clinical examination of oral cavities was carried out by only three trained and calibrated dentists, according to the World Health Organization criteria [32] of the four-point scale caries lesion extension [33]. The cavitated carious lesions were classified according to a 4-point scale: D1- clinically detectable enamel caries with a sound surface, inclusive of white and brown spot lesions, D2 - cavitated enamel caries, D3 - carious lesion involving dentin, D4 - lesions involving pulp [33].

Before data collection, examiner calibration was performed. Inter-examiner and intra-examiner reliability for caries diagnosis were assessed using Cohen’s kappa statistic calculated in IBM SPSS Statistics for Windows (Version 25; IBM, Armonk, NY, USA). Kappa values ranged from 0.81 to 0.85, indicating substantial agreement. Dental examinations were carried out using a sterile plane dental mouth mirror and a blunt-tipped probe/explorer under white LED light, with participants seated upright, in accordance with World Health Organization diagnostic criteria. The caries status of the primary teeth of children was recorded using the deft (sum of decayed, extracted due to caries, and filled teeth) and the defs (sum of decayed, extracted due to caries, and filled surfaces) indices [32, 34]. According to the WHO diagnostic criteria, as decayed teeth, only cavitated teeth were counted, meaning that non-cavitated white or brown spot lesions are explicitly excluded from this category.

Determination of the nutritional status of children through anthropometric data

Anthropometric measurements of patients were taken using standard techniques. Each measurement was taken twice, and the mean was used for calculations. Data such as HA, WA, WH, BMIA, MUACA, HCA, and TSFTA of children were obtained using standard normative charts. The body weights were measured using a portable calibrated scale (T-79–100; Tsutsumi, Sanjo, Niigata, Japan) with 100 g precision, with the child standing barefoot in minimal clothing, and height was measured using a tape measure. TSFTA was measured using a skinfold caliper (Slim Guide; Body Care, Kenilworth, UK). The WHO child growth standards were used to evaluate the nutritional status of selected children. The child’s nutritional status was determined using Z-scores, where a Z-score below − 2 was considered indicative of poor nutritional status in terms of HA, WA, WH, BMIA, MUACA, HCA, and TSFTA [35, 36].

Collecting information through questionnaires

All enrolled children were interviewed using parent/guardian-administered questionnaires. Information about children’s demographics, oral and general health status, and eating habits was collected using a previously validated and published questionnaire [37], which was translated into Sinhala (the main language of Sri Lanka) and verified by an expert panel within the Ethics Review Committee. The dietary component operated as an adapted Food Frequency Questionnaire (FFQ) designed to assess the habitual intake of core food groups and cariogenic items. To capture established dietary patterns, the FFQ structured food consumption frequencies into five standardized categorical responses: ‘more than 1 time/day’, ‘6–7 times/week’, ‘3–5 times/week’, ‘1–2 times/week’, and ‘fewer than 1 time/week or never’. While the tool was originally validated and implemented to assess socio-behavioral caries risk factors in preschool cohorts within a project framework [37], its application in the current study was adapted to align with the regional dietary habits of Sri Lankan preschoolers. Content and face validity were verified by a panel of local pediatric dental experts before administration. The validated ECOHIS was used to assess the OHRQoL of children [22].

Statistical analysis

IBM SPSS Statistics for Windows (Version 25; IBM, Armonk, NY, USA) was used for statistical analyses. The first primary outcome was dental caries, which was subdivided into the following three levels of severity: mild (scores from 1 to 3), moderate (scores from 4 to 5), and severe ECC (scores similar to or above 6) [38]. Based on the defs index, the severity of dental caries was subdivided into ECC (defs < 6) and S-ECC (defs ≥ 6) [39]. The term S-ECC refers to the atypical, progressive, acute, or rampant pattern of decay, which is taken as defs ≥ 6 in this study [39]. The associations between caries experience and each of the independent variables were evaluated.

The second primary outcome was nutritional status assessed by HA, WA, WH, BMIA, MUACA, HCA, and TSFTA. The associations between nutritional status and each of the independent variables were evaluated. The third primary outcome was OHRQoL assessed by the child impact section (CIS), family impact section (FIS), and overall ECOHIS scores. The associations between OHRQoL and each of the independent variables were evaluated.

The independent variables in each dependent variable category included nutritional status, ECC status, demographics, eating habits, oral health, and general health. The statistical analyses included descriptive information and frequency distributions. Binary logistic regression analysis (using the enter method) was used to evaluate the association of independent variables on dichotomous dependent variables (e.g., presence of dental caries, dichotomized as “yes” or “no”). Dental caries severity was measured using two levels of severity of deft and defs: defs < 6 and defs ≥ 6. This single, conservative cutoff was selected to maintain statistical homogeneity and prevent the artificial inflation of S-ECC cases, given that the study sample was tightly clustered and chronologically advanced within the target range, with a mean age of 55.8 ± 5.2 months (representing children predominantly approaching or entering their fifth year). A minor proportion of records (27 children) containing missing or incomplete questionnaire responses, dental examination data, and anthropometric data were excluded from the final multivariable models, which is reflected in the provided participant flow diagram in Fig. 1.

To evaluate the ECOHIS counts (CIS, FIS, and overall scores), data were first tested for overdispersion. Due to significant overdispersion, negative binomial regression analysis was performed instead of a standard Poisson regression to estimate associations with the independent variables. The independent variables were entered into the regression analysis if they exhibited P values < 0.05. The level of significance was set as α ≤ 0.05. To account for the inflated risk of type I errors associated with evaluating multiple dietary habits simultaneously, a multiple-testing correction was implemented across the dietary panel. The false discovery rate (FDR) was controlled at 5% (Q = 0.05) utilizing the Benjamini–Hochberg procedure. Raw P-values for dietary variables were ranked, and adjusted P-values were computed to identify robust statistical associations. Model stability and sample size adequacy across the three regression models were verified using events-per-variable (EPV) metrics (Additional file 1: Table S3).

Results

ECC status

Overall, 77.6% of children showed evidence of ECC, and among those with dental caries, 98.0% were untreated. The number of deft per child ranged from 0 to 20, with a mean (± SD) deft of 6.25 (± 5.07). Mild, moderate, and severe ECC conditions were observed in 14.5%, 10.8%, and 52.3% of children, respectively (Table 1). The clinical presentation and progressive stages of S-ECC are illustrated in Fig. 2. Approximately 3 out of 5 children (63.1%) had four or more decayed teeth. The number of defs ranged from 0 to 82 per child, with a mean (± SD) defs of 15.76 (± 15.61). Approximately 3 out of 5 children (65.5%) had six or more decayed tooth surfaces.

Table 1.

Frequency of ECC status among children

Dental caries status		Mean (± SD) or Number (%)
deft	Mean deft	6.25 (± 5.07)
	Decayed teeth	2996 (88.0)
	Extracted teeth	340 (10.0)
	Filled teeth	70 (2.0)
	Untreated caries	3336 (98.0)
	No ECC (deft = 0)	122 (22.4)
	Mild ECC (0 < deft ≤ 3)	79 (14.5)
	Moderate ECC (4 ≤ deft ≤ 5)	59 (10.8)
	S-ECC (deft ≥ 6)	285 (52.3)
defs	Mean defs	15.76 (± 15.61)
	Decayed surfaces	6965 (81.3)
	Extracted surfaces	1501 (17.5)
	Filled surfaces	105 (1.2)
	Untreated surfaces	8466 (98.8)
	No ECC (defs = 0)	122 (22.4)
	ECC (defs < 6)	66 (12.1)
	S-ECC (defs ≥ 6)	357 (65.5)
Caries lesion extensions on the tooth surfaces D1 caries	Mean D1	3.66 (± 4.80)
	D1 = 0	193 (35.4)
	0 < D1 ≤ 5	218 (40.0)
	6 ≤ D1 ≤ 15	117 (21.5)
	16 ≤ D1 ≤ 25	15 (2.7)
	D1 ≥ 26	2 (0.4)
D2 caries	Mean D2	3.21 (± 4.56)
	D2 = 0	217 (39.8)
	0 < D2 ≤ 5	216 (39.6)
	6 ≤ D2 ≤ 15	90 (16.5)
	16 ≤ D2 ≤ 25	21 (3.9)
	D2 ≥ 26	1 (0.2)
D3 caries	Mean D3	3.88 (± 5.37)
	D3 = 0	220 (40.4)
	0 < D3 ≤ 5	184 (33.8)
	6 ≤ D3 ≤ 15	114 (20.9)
	16 ≤ D3 ≤ 25	23 (4.2)
	D3 ≥ 26	4 (0.7)
D4 caries	Mean D4	4.83 (± 8.22)
	D4 = 0	287 (52.7)
	0 < D4 ≤ 5	125 (22.9)
	6 ≤ D4 ≤ 15	76 (13.9)
	16 ≤ D4 ≤ 25	36 (6.6)
	D4 ≥ 26	21 (3.9)

Frequency of ECC (early childhood caries) status of studied children in the population. Abbreviations: S-ECC - severe early childhood caries, deft - sum of decayed, extracted due to caries, and filled teeth, defs - sum of decayed, extracted due to caries, and filled surfaces, D0 - no caries, D1- clinically detectable enamel caries with a sound surface, inclusive of white and brown spot lesions, D2 - cavitated enamel caries, D3 - carious lesion involving dentin, and D4 - lesions involving pulp

Fig. 2.

Severe stages of ECC among children aged 4–5 years. Clinical presentation of severe early childhood caries (S-ECC) among pediatric participants aged 4–5 years. Separate written informed consent was obtained from parents/guardians before photographic documentation and publication. All photographs have been tightly cropped to the intraoral field to guarantee complete participant anonymization and confidentiality. Standard anatomical proportions of the primary maxillary teeth serve as an internal scale reference across panels

When all examined carious lesions were considered were categorized as D1, D2, D3, and D4 caries, and their proportions are given in Fig. 3. The D1 carious lesions mean (± SD) per child was 3.66 (± 4.80), with 40.0% of children having 1–5 D1 caries. Means (± SD) of D2 and D3 lesions were 3.21 (± 4.56) and 3.88 (± 5.37), respectively. The D4 carious lesions mean (± SD) per child was 4.83 (± 8.22), with 22.9% of children having 1–5 D4 caries. Twenty-one children presented with extremely severe ECC, having ≥ 26 tooth surfaces affected with D4 caries.

Fig. 3.

Observed percentages of caries lesion extensions. Percentage distribution of dental caries lesion extensions observed among the study population of children aged 4–5 years. Clinical severity is classified according to the World Health Organization criteria, a 4-point scale tracking progressive depth of decay: D1 (clinically detectable enamel caries with a sound surface, inclusive of white and brown spot lesions), D2 (cavitated enamel caries), D3 (carious lesion involving dentin), and D4 (lesions involving pulp)

Nutritional status by anthropometric measurements

The children’s mean (± SD) anthropometric measurements were as follows: height 104.47 (± 5.70) cm, weight 15.17 (± 2.53) kg, BMI 13.84 kg/m² (± 1.43), MUAC 15.47 (± 1.46) cm, HC 49.33 (± 2.08) cm, and TSFT 8.68 (± 2.78) mm. Though most children had height-for-age Z-scores (HAZ) around the median (−1 to + 1 SD), their weight-for-age (WAZ), weight-for-height (WHZ), and BMI-for-age (BMIAZ) Z-scores tended to shift between the median and − 3 SD. Mid-upper arm circumference-for-age (MUACAZ), head circumference-for-age (HCAZ), and triceps skinfold thickness-for-age (TSFTAZ) mostly fell between − 1 and + 1 SD. The percentages of children who had adequate and mildly, moderately, and severely depleted nutritional status are illustrated in Fig. 4.

Fig. 4.

Distribution of nutritional status parameters among children aged 4–5 years based on World Health Organization (WHO) Z-score classifications. Height-for-age Z-score > + 2 SD tall-for-age, + 1 to >−1 SD normal, <−1 to −2 SD mild stunting, <−2 to −3 SD moderate stunting, <−3 SD severe stunting; Weight-for-age Z-score > + 2 SD overweight, + 1 to >−1 SD normal, <−1 to −2 SD mild underweight, <−2 to −3 SD moderate underweight, <−3 SD severe underweight; weight-for-height Z-score > + 2 SD overweight, + 1 to >−1 SD normal, <−1 to −2 SD mild wasting, <−2 to −3 SD moderate wasting, <−3 SD severe wasting; BMI-for-age Z-score score > + 2 SD overweight, + 1 to >−1 SD normal, <−1 to −2 SD mild thinness, <−2 to −3 SD moderate thinness, <−3 SD severe thinness; mid-upper arm circumference (MUAC)-for-age, head circumference (HC)-for-age, and triceps skinfold thickness (TSFT)-for-age Z-score score > + 2 SD higher, + 1 to >−1 SD adequate, <−1 to −2 SD mildly depleted, <−2 to −3 SD moderately depleted, <−3 SD severely depleted

Demographics

The mean (± SD) age of children was 55.78 (± 5.21) months; 282 (51.7%) were male, and 263 (48.3%) were female. Most children were Sinhalese (87.5%), followed by Muslims (9.6%) and Tamils (2.9%). Nearly half (45.3%) had one sibling, while only 0.4% had four siblings. The largest proportion of children (41.1%) was second-born, and only 1.8% were fourth-born. A small percentage (1.1%) lived with grandparents, although 27.3% were cared for by grandparents during parents’ working hours. Among mothers, 50.8% had a collegiate education and 13.6% had a tertiary education, although 62.2% were unemployed. The mean (± SD) maternal age was 33.83 (± 4.98) years. Among fathers, 46.9% had collegiate, and 11.6% tertiary education, and only 0.4% were unemployed. The highest proportion of children (43.1%) lived in households earning < 50,000 LKR per month.

Factors associated with the nutritional status of children

To statistically analyze the association between nutritional status (BMIA as the dependent variable) and the other independent variables studied, factors were grouped into four categories: ECC status, demographics, eating habits, and general health status. Dietary patterns and the FFQ structured food consumption frequencies are summarized in Additional file 1: Table S1. For each level, unconditional logistic regression analysis was performed to exclude variables with a P-value of ≥ 0.20. Explanatory variables were selected for the final binary logistic regression analysis model only if they had a P-value of ≤ 0.05 after adjustment for variables from the same or prior levels of determinants. This stepwise approach allows researchers to quantify the contribution of each level of adjustment and to understand the model-building strategy, as well as to interpret the independent associations.

The regression analysis for the first horizontal level selected the defs index as the ECC-related factor associated with poor nutritional status (Z-score < −2) of children (P < 0.20). Among demographic factors, the person responsible for childcare, maternal and paternal education up to secondary and collegiate levels, and household income were associated with poor nutritional status (P < 0.20). Consumption of bakery items, fruit juice, ice cream, candy, and tea with sugar ≤ 2 or > 2 days/week were identified as the eating habits associated with the poor nutritional status of children (P < 0.20). General health–related factors associated with poor nutritional status (P < 0.20) included birth weight, gestational age at birth, and breastfeeding duration.

The results of this final regression analysis indicated that there was a significant positive correlation between the defs index (OR = 1.546, CI: 1.022–2.352) and the incidence of poor nutritional status (BMIA Z-score < −2) (P = 0.042). Children whose mothers had been educated up to grade 13 were less likely (P = 0.047) to experience poor nutritional status (BMIA Z-score < −2) as evidenced by the odds ratio of 0.698 (CI: 0.411–0.917). Among eating habits, frequent consumption of ice cream (OR = 1.689, CI: 1.031–2.772, P = 0.047) and tea with sugar (OR = 1.965, CI: 1.056–3.655, P = 0.042) was significantly and positively associated with the occurrence of poor nutritional status. In addition, the birth weight of children was negatively associated with the poor nutritional status of children (OR = 0.459, CI: 0.285–0.739, P = 0.010).

Factors associated with the ECC status of children

To investigate the association between ECC status (with defs as the dependent variable) and the other independent variables studied, an initial unconditional logistic regression analysis was performed to screen variables with a P value ≥ 0.20. The screened explanatory variables were then included in the final binary logistic regression analysis model, and associations were considered statistically significant only if the P value was ≤ 0.05 (Table 2).

Table 2.

Binary logistic regression analysis model for factors associated with S-ECC status (defs ≥ 6)

Variable	Raw count (%)	OR	95% CI	P-value
Weight-for-age (Z-score < −2); Ref: Z-score > −2	114 (20.9)	1.320	1.010–1.730	0.040
BMI-for-age (Z-score < −2); Ref: Z-score > −2	141 (25.9)	1.118	1.010–1.237	0.038
Mother’s education - up to collegiate education (Grade 13); Ref: up to Grade 11	277 (50.8)	0.439	0.122–0.957	0.024
Muffins/donuts/cakes/pancakes/waffles/buns/bread or other bakery items (> 2 days/week); Ref: ≤ 2 days/week	258 (47.3)	1.478	1.138–1.921	0.040*
Sugary spreaders (Nutella/jam/adding sugar on butter/margarine) (> 2 days/week); Ref: ≤ 2 days/week	164 (30.1)	2.243	1.135–4.432	0.036*
Ice cream (> 2 days/week); Ref: ≤ 2 days/week	258 (47.3)	1.121	1.001–1.241	0.043*
Chocolates (> 2 days/week); Ref: ≤ 2 days/week	333 (61.1)	1.302	1.084–1.563	0.030*
Candy/sweets (other than chocolate) (> 2 days/week); Ref: ≤ 2 days/week	384 (70.5)	1.128	1.012–1.259	0.042*
Breast-fed duration (> 2 years); Ref: ≤ 2 years	318 (58.4)	6.258	1.956–17.989	0.001
Bottle-fed (Ref: not feeding)	120 (22.0)	2.051	1.044–4.138	0.032
Frequency of tooth brushing (once); Ref: twice	206 (37.8)	1.865	1.087–3.199	0.024
Who brushes your child’s teeth? (the child); Ref: an adult/both	131 (24.0)	2.116	1.164–3.847	0.014
Type of toothpaste (non-fluoridated); Ref: fluoridated	76 (13.9)	2.899	1.177–7.143	0.021

Final binary logistic regression model identifying factors associated with severe early childhood caries (S-ECC), defined as defs (sum of decayed, extracted due to caries, and filled surfaces index) ≥ 6. *Adjusted P-values: Benjamini–Hochberg multiple-testing correction was applied specifically across the variables within the dietary panel to control the False Discovery Rate at 5% (Q = 0.05)

The regression analysis for the first horizontal level selected all anthropometric parameters (Z-score < −2), except HCA, as the nutritional status-related factors associated with S-ECC status (defs ≥ 6) of children (P < 0.20). Gender, ethnicity, mother’s occupation, mother’s and father’s education, and household income were screened as the demographic factors associated with S-ECC status (defs ≥ 6) of children (P < 0.20). Consumption of yogurt, flavored milk, bakery items, savory snacks, biscuits, dried and fresh fish, Thriposha with sugar, healthy and sugary spreads, nuts, soft drinks, ice cream, chocolate, other candy, and tea with sugar ≤ 2 or > 2 days/week were identified as the eating habits associated with S-ECC status (defs ≥ 6) of children (P < 0.20). Breastfeeding duration and bottle feeding were recognized as general health-related factors associated with S-ECC status (defs ≥ 6) of children (P < 0.20). Frequency of tooth brushing, the person who brushes the child’s teeth, the use of fluoridated or non-fluoridated toothpaste, and the type of container used to drink sugary liquids were observed as oral health-related factors associated with S-ECC status (defs ≥ 6) of children (P < 0.20).

The results of this final regression analysis indicated that there was a significant relationship between the incidence of S-ECC (defs ≥ 6) and WAZ (OR = 1.320, CI: 1.010–1.730, P = 0.040) as well as BMIAZ (OR = 1.118, CI: 1.010–1.237, P = 0.038). The final regression analysis also indicated that there was a significant relationship between the occurrence of S-ECC (defs ≥ 6) and mother’s education up to collegiate education (OR = 0.439, CI: 0.122–0.957, P = 0.024). Among eating habits, children who consumed bakery products (P = 0.040), sugary spreads (P = 0.036), ice cream (P = 0.043), chocolates (P = 0.030), and other candies (P = 0.042) more than twice per week were more likely to have experienced S-ECC, as shown by odds ratios of > 1.000. In addition, Children breastfed for more than two years (OR = 6.258, CI: 1.956–17.989, P = 0.001) and those who were bottle-fed (OR = 2.051, CI: 1.044–4.138, P = 0.032) were more likely to have S-ECC. Moreover, Oral health behaviors such as brushing only once per day (OR = 1.865, CI: 1.087–3.199, P = 0.024), brushing by the child alone (OR = 2.116, CI: 1.164–3.847, P = 0.014), and using non-fluoridated toothpaste (OR = 2.899, CI: 1.177–7.143, P = 0.021) were significantly associated with S-ECC, as shown in Table 2.

Factors associated with the OHRQoL of children

Additional file 1: Table S2 shows ECOHIS responses and mean scores for the total ECOHIS and each section. The associations between OHRQoL (using the CIS, FIS, and overall ECOHIS scores as the dependent variables) and other studied independent variables were assessed by negative binomial regression analysis (Table 3). It was found that HAZ (RR = 1.109, CI: 1.003–1.226, P = 0.043), WAZ (RR = 1.784, CI: 1.319–1.854, P = 0.002), and BMIAZ (RR = 1.231, CI: 1.091–1.392, P = 0.001) were significantly associated with the CIS scores. Among the ECC status-related factors, the defs index (P = 0.002), D1 (P = 0.001), D2 (P = 0.002), D3 (P = 0.003), and D4 (P = 0.008) caries were significantly and positively associated with the CIS scores. Among demographic factors, being a male child (P = 0.021), birth order (P = 0.046), the mother being unemployed (P = 0.002), the mother’s age being ≤ 30 years (P = 0.014), and the father being unemployed/daily paid-employee (P = 0.008) were positively associated with the CIS scores (P < 0.05). In addition, toothbrushing only once a day (P = 0.016), toothbrushing by the child alone (P = 0.003), and using non-fluoridated toothpaste (P = 0.045) appeared to result in higher CIS scores as demonstrated by rate ratios (RR) of 1.068, 1.114, and 1.059, respectively.

Table 3.

Negative binomial regression analysis of the association between independent variables and sections and the total ECOHIS score

Variable	Child impact section RR (95% CI)	P-value	Family impact section RR (95% CI)	P-value	Overall ECOHIS RR (95% CI)	P-value
Nutritional status (Ref: Z-score > −2)
Height-for-age (Z-score < −2)	1.109 (1.003–1.226)	0.043	1.165 (1.002–1.359)	0.047	1.131 (1.042–1.227)	0.004
Weight-for-age (Z-score < −2)	1.784 (1.319–1.854)	0.002	1.824 (1.365–1.938)	0.001	1.791 (1.338–1.951)	0.002
Weight-for-height (Z-score < −2)	0.965 (0.841–1.104)	0.718	1.044 (0.845–1.291)	0.698	0.987 (0.881–1.106)	0.885
BMI-for-age (Z-score < −2)	1.231 (1.091–1.392)	0.001	1.127 (1.001–1.249)	0.047	1.204 (1.086–1.336)	0.003
MUAC-for-age (Z-score < −2)	1.032 (0.946–1.127)	0.551	1.062 (0.927–1.219)	0.412	1.027 (0.954–1.106)	0.421
HC-for-age (Z-score < −2)	1.041 (0.932–1.161)	0.948	1.021 (0.858–1.214)	0.793	1.051 (0.958–1.153)	0.312
TSFT-for-age (Z-score < −2)	0.988 (0.892–1.093)	0.894	0.951 (0.812–1.114)	0.521	0.972 (0.893–1.059)	0.538
ECC status
deft (deft ≥ 6); Ref: deft < 6	1.023 (0.941–1.112)	0.592	1.078 (0.946–1.104)	0.892	1.012 (0.944–1.085)	0.741
defs (defs ≥ 6); Ref: defs < 6	3.328 (2.985–3.711)	0.002	3.224 (2.751–3.778)	0.002	3.281 (2.999–3.589)	0.001
D1 caries	1.012 (1.007–1.017)	0.001	1.004 (0.996–1.013)	0.485	1.009 (1.004–1.013)	0.001
D2 caries	1.009 (1.004–1.015)	0.002	1.007 (0.998–1.016)	0.091	1.007 (1.003–1.012)	0.002
D3 caries	1.015 (1.011–1.020)	0.003	1.013 (1.005–1.020)	0.002	1.017 (1.013–1.021)	0.003
D4 caries	1.005 (1.002–1.008)	0.008	1.002 (0.997–1.007)	0.284	1.005 (1.002–1.007)	0.003
Demographics
Gender (Male); Ref: Female	1.061 (1.009–1.115)	0.021	0.992 (0.919–1.072)	0.849	1.042 (0.998–1.087)	0.063
Siblings	0.985 (0.946–1.026)	0.478	1.019 (0.957–1.085)	0.548	0.997 (0.963–1.031)	0.854
Birth order	1.042 (1.001–1.091)	0.046	1.028 (0.958–1.103)	0.441	1.047 (1.008–1.088)	0.019
The child was looked after by parents (Ref: grandparents/caretaker)	0.836 (0.777–0.899)	0.002	0.857 (0.798–0.999)	0.038	0.858 (0.806–0.913)	0.001
Ethnicity
Sinhalese	0.854 (0.782–0.933)	0.001	0.963 (0.835–1.111)	0.603	0.884 (0.819–0.954)	0.001
Tamil	0.929 (0.801–1.078)	0.331	1.139 (0.903–1.435)	0.279	0.987 (0.871–1.119)	0.835
Muslim (Ref)	1.000		1.000		1.000
Mother’s occupation (Unemployed); Ref: employed	1.137 (1.053–1.227)	0.001	1.044 (0.928–1.173)	0.472	1.109 (1.039–1.182)	0.002
Mother’s education
Up to secondary education (Grade 11)	0.949 (0.852–1.056)	0.334	0.927 (0.786–1.093)	0.369	0.946 (0.864–1.035)	0.224
Up to collegiate education (Grade 13)	0.981 (0.891–1.081)	0.695	1.014 (0.875–1.175)	0.852	0.989 (0.912–1.072)	0.781
Up to tertiary education (University education)	1.000		1.000		1.000
Mother’s age (≤ 30 years); Ref: > 30 years	1.079 (1.016–1.147)	0.014	1.096 (1.002–1.201)	0.045	1.084 (1.030–1.140)	0.002
Father’s occupation (Unemployed/daily paid employee); Ref: employed	1.099 (1.032–1.171)	0.008	1.159 (1.051–1.278)	0.003	1.116 (1.058–1.177)	0.001
Father’s education
Up to secondary education (Grade 11)	1.092 (0.975–1.225)	0.126	1.038 (0.871–1.236)	0.681	1.068 (0.970–1.175)	0.178
Up to collegiate education (Grade 13)	1.018 (0.918–1.129)	0.737	1.022 (0.874–1.195)	0.785	1.019 (0.935–1.111)	0.652
Up to tertiary education (University education)	1.000		1.000		1.000
Monthly household income (≤ 50,000 LKR); Ref: > 50,000 LKR	1.049 (0.989–1.112)	0.112	1.121 (1.023–1.228)	0.014	1.078 (1.025–1.133)	0.003
General health status
Birth weight	0.858 (0.805–0.914)	0.001	0.826 (0.749–0.909)	0.001	0.844 (0.801–0.890)	0.002
Pregnancy week in which the child was born (< 37 weeks); Ref: ≥ 37 weeks	0.903 (0.839–0.971)	0.006	0.883 (0.788–0.991)	0.035	0.899 (0.846–0.957)	0.003
Breast-fed duration (≤ 2 years) Ref: > 2 years)	0.864 (0.812–0.918)	0.001	0.846 (0.769–0.929)	0.001	0.859 (0.816–0.904)	0.001
Bottle-fed (Ref: not feeding)	1.081 (1.018–1.149)	0.012	1.059 (1.001–1.119)	0.047	1.078 (1.026–1.133)	0.003
Was the child given vitamin and mineral supplements? (Ref: No)	1.017 (0.892–1.118)	0.094	1.285 (0.971–1.699)	0.134	1.311 (1.126–1.527)	0.082
Oral health status
Frequency of tooth brushing (Once); Ref: twice	1.068 (1.012–1.127)	0.016	1.062 (1.004–1.114)	0.029	1.063 (1.015–1.112)	0.009
Who brushes your child’s teeth? (The child); Ref: an adult/both	1.114 (1.056–1.175)	0.003	1.079 (1.011–1.147)	0.022	1.104 (1.056–1.155)	0.001
Type of toothbrush (Adult toothbrush); Ref: junior toothbrush	0.939 (0.830–1.062)	0.315	0.903 (0.745–1.095)	0.298	0.941 (0.849–1.042)	0.243
Type of toothpaste (Non-fluoridated); Ref: fluoridated	1.059 (1.001–1.120)	0.045	1.021 (1.003–1.037)	0.025	1.057 (1.007–1.106)	0.024
Drinking sugary liquids (Bottle/sippy cup); Ref: regular cup	1.119 (0.992–1.182)	0.621	0.898 (0.761–1.058)	0.194	0.938 (0.855–1.030)	0.178

Negative binomial regression analysis examining the association between independent variables and individual sections as well as the total Early Childhood Oral Health Impact Scale (ECOHIS) score. Abbreviations: BMI – body mass index, MUAC - mid-upper arm circumference, HC - head circumference, TSFT - triceps skinfold thickness, deft - sum of decayed, extracted due to caries, and filled teeth index, and defs - sum of decayed, extracted due to caries, and filled surfaces index

The second negative binomial regression analysis revealed that HAZ (RR = 1.165, CI: 1.002–1.359, P = 0.047), WAZ (RR = 1.824, CI: 1.365–1.938, P = 0.001), and BMIAZ (RR = 1.127, CI: 1.001–1.249, P = 0.047) were significantly associated with the FIS scores. In the ECC status-related factors category, defs index (RR = 3.224, CI: 2.751–3.778, P = 0.002) was significantly associated with the FIS scores. Among demographic characteristics, children cared for by parents rather than grandparents or caretakers were less likely to have high FIS scores (RR = 0.857, CI: 0.798–0.999, P = 0.038). Higher FIS scores were also significantly associated with maternal age ≤ 30 years (P = 0.045), paternal unemployment/daily-paid employment (P = 0.003), and household income ≤ 50,000 LKR (P = 0.014) (RR > 1.000). Similar to CIS findings, toothbrushing once daily (P = 0.029), brushing by the child alone (P = 0.022), and using non-fluoridated toothpaste (P = 0.025) were associated with higher FIS scores (RR = 1.062, 1.079, and 1.021, respectively).

The third negative binomial regression analysis identified independent variables significantly associated with overall ECOHIS scores. According to that, HAZ (RR = 1.131, CI: 1.042–1.227, P = 0.004), WAZ (RR = 1.791, CI: 1.338–1.951, P = 0.002), and BMIAZ (RR = 1.204, CI: 1.086–1.336, P = 0.003) Z-scores were significantly and positively associated with the overall ECOHIS scores. Among the ECC status-related factors, the defs index (P = 0.001), D1 (P = 0.001), D2 (P = 0.002), D3 (P = 0.003), and D4 (P = 0.003) caries were positively associated with the overall ECOHIS scores. Among demographics, birth order (P = 0.019), maternal unemployment (P = 0.002), maternal age ≤ 30 years (P = 0.002), paternal unemployment/daily-paid employment (P = 0.001), and household income ≤ 50,000 LKR (P = 0.003) were more likely to have higher overall ECOHIS scores, as evidenced by rate ratios of > 1.000. Under oral health-related factors, toothbrushing only once daily (P = 0.009), toothbrushing by the child alone (P = 0.001), and using non-fluoridated toothpaste (P = 0.024) appeared to result in higher overall ECOHIS scores as demonstrated by RR of 1.063, 1.104, and 1.057, respectively.

Discussion

ECC status, poor nutritional status, and OHRQoL

Few studies have examined the prevalence of ECC across different districts in Sri Lanka [40–42]; however, research utilizing the detailed D1–D4 caries classification is particularly limited in Sri Lanka. The mean deft score in the present sample was higher, likely due to the hospital-based nature of the study population [40–42]. For instance, a study in the Kegalle District reported that 71.9% of 4–6-year-old children had dental caries, with a mean dmfs score of 6.4 [22]. The mean defs score in the current clinic-based study was considerably higher (15.76 ± 15.61), reflecting that most participating children were seeking treatment at the dental hospital. Similarly, a study conducted in a dental hospital in the Colombo District found that 66.53% of children aged 0–3 years had ≥ 6 carious teeth [43]. The present study showed a comparable defs proportion (65.5%) among the assessed 545 children aged 4–5 years.

Dietary patterns in the present study revealed high consumption frequencies of sugary foods: 70.5% consumed sweets, 61.1% chocolate, and 47.2% ice cream more than three days per week. According to WHO guidelines, free sugar intake should be < 10% of total energy intake; it is likely that most children in this sample exceeded this threshold (10). Although a substantial proportion consumed pulses (63.3%), green vegetables (94.3%), fruits (54.3%), and cheese (15.4%) frequently, the intake of cariogenic foods was also high. For instance, 90.8% consumed biscuits, 47.3% bakery items, 61.1% chocolates, and 70.5% other sweets more than three days per week.

Factors associated with the nutritional status of children

The null hypothesis, which stated that there is no significant association between higher ECC status and poor nutritional status among Sri Lankan children aged 4–5 years, was rejected based on the observed strong positive association between the defs index and the occurrence of poor nutritional status. It was further confirmed by a study that mentioned each tooth of children with deep decay increased an additional 10% odds of having undernutrition, measured by HA, WA, and BMIA [44]. In addition, a few studies have reported that the higher incidence of dental caries increased malnutrition status among children [45, 46]. This nutritional deficit can be driven by caries-related pain and discomfort, which impairs masticatory function (chewing difficulty) and subsequently reduces overall dietary intake [47].

Maternal education showed a negative correlation with poor nutritional status, consistent with findings from numerous international studies [48–50]. The observed positive associations between the intake of ice cream and tea with sugar and poor nutritional status align with published evidence indicating that frequent consumption of sweet snacks and tea with sugar is a significant predictor of undernutrition [51]. High intake of low-nutrient carbohydrates may reduce dietary diversity; children consuming fewer than four food varieties have been shown to have a higher risk of being underweight [52]. The negative association between birth weight and poor nutritional status was also in line with earlier findings [48, 53, 54].

Factors associated with the ECC status of children

Although many reports would call for references in declining ECC rates in other countries, high prevalence persists in parts of Asia, including Sri Lanka [55]. In this study, S-ECC was significantly associated with WAZ and BMIA, supporting evidence that children with borderline or poor nutritional status are more prone to ECC [24, 25, 56]. These results further support the rejection of the null hypothesis that there is no association between higher ECC and poor nutritional status among children aged 4–5 years in Sri Lanka. Maternal educational level was marginally associated with higher untreated caries, consistent with previous findings [43], although one study in Kegalle reported that lower maternal education was linked to better OHRQoL [22]. Globally, lower maternal education is strongly associated with higher ECC occurrence [57–59].

Frequent consumption of sugary snacks, sweets, and sweetened beverages was associated with S-ECC, similar to findings from Sri Lankan studies [60–62] and international studies [19, 63–65]. Frequent consumption of free sugars creates a sustained acidic microenvironment within dental plaque, favoring the proliferation of aciduric bacteria over beneficial oral microbiota. This continuous acid exposure exceeds the natural remineralizing capacity of saliva, leading to rapid, unchecked destruction of primary tooth structure [66]. Long-term breastfeeding [63, 64, 67, 68] and bottle-feeding [20, 69, 70] were very strongly and significantly associated with ECC, with ORs of 6.258 and 2.051, respectively, which is consistent with several previous reports. Toothbrushing less than twice daily [71, 72], brushing without supervision [73, 74], and the use of non-fluoridated toothpaste [75, 76] significantly increased ECC abundance, supporting evidence from numerous studies.

Factors associated with the OHRQoL

The null hypotheses of no association between poor OHRQoL and poor nutritional status and no association between poor OHRQoL and higher ECC were both rejected by the observed strong positive associations. The observed strong associations are in agreement with published literature, as given in Table 4 [6, 77]. Many authors have observed an association between ECC and OHRQoL assessed by ECOHIS [78]. This association exists because the symptoms of ECC, such as severe dental pain, functional chewing difficulties, and altered sleep patterns, directly compromise daily well-being, functional performance, and psychological development [79]. Mean ECOHIS scores were found statistically significantly higher in children with caries compared with caries-free children below 5 years [80–85].

Table 4.

Comparison of present study findings with published cohorts on factors associated with children’s oral health-related quality of life

Study	Country	Sample size	Age range (years)	Key findings
Current study [2026]	Sri Lanka	545	4–5	Overall ECOHIS scores were significantly associated with poor nutritional status (HAZ, WAZ, BMIAZ), ECC status (defs index), toothbrushing frequency, maternal age, employment, and household income (P = 0.001–0.004)
Bagis et al. [6]	Turkey	95	3–5	Upon the treatment of ECC, the total ECOHIS score significantly decreased (P < 0.001)
Gunay et al. [77]	Turkey	24	3–6	ECOHIS scores were significantly reduced after the elimination of caries pain and inflammation (P < 0.05)
Permatasari et al. [78]	Indonesia	165	3–5	All domains except the self-image/social interaction domain and the overall ECOHIS score were significantly associated with ECC status measured by deft (P < 0.05)
Naidu et al. [81]	Trinidad	251	3–5	The odds of having negative OHRQoL impacts were significantly associated with having higher dmft
Chaffee et al. [82]	Brazil	456	3–5	ECOHIS scores were 3.0 times greater for children with dmft ≥ 5 vs. dmft = 0 (P > 0.3). Adjusted for dental status and socio-demographics, mean ECOHIS scores were lower when reported by mothers with less education, lower social class, or lower income
Bekes et al. [83]	Austria	336	0–5	Mean ECOHIS scores were significantly higher in children with caries compared with caries-free children (P < 0.001)
Sharna et al. [84]	India	238	0.5–6	Children with higher ECC (higher pufa score) had significantly lower quality of life than children with lower ECC (P < 0.001)
Mtalsi et al. [85]	Morocco	546	4–6	The S-ECC was negatively associated with the quality of life of childre
Wong et al. [86]	China	1261	3–5	Higher ECOHIS scores were found with children who had higher dmft and lower parental education and income (P < 0.001)
Scarpelli et al. [87]	Brazil	1632	5–6	ECOHIS scores were negatively associated with dmft scores of children, age of parents/caregivers, and monthly household income (P < 0.001)
Abanto et al. [88]	Brazil	260	2–5	The severity of ECC was significantly correlated with OHRQoL (P < 0.001). The increase in the child’s age, higher household crowding, lower family income, and mother working out of the home were significantly associated with OHRQoL (P < 0.05)
Carrillo-Díaz et al. [88]	Spain	273	2–4	The children’s OHRQoL is positively correlated with the number of night-time breastfeeding sessions, dmft index, and the duration of co-sleeping
Pereira et al. [17]	Brazil	163	3–4	Mean ECOHIS and parental anxiety domain scores were significantly higher in children with dental caries (P < 0.05). Household income, maternal education, mother’s job status, and breastfeeding time were significantly associated with ECOHIS scores

ECOHIS- Early Childhood Oral Health Impact Scale; OHRQoL- Oral Health-Related Quality of Life; dmft- decayed, missing, and filled teeth; defs- decayed, extracted due to caries, and filled teeth, and filled surfaces; HAZ- height-for-age Z-scores; WAZ- weight-for-age Z-scores; BMIAZ- BMI-for-age Z-scores

Some studies in the published literature further confirmed the observed association between ECOHIS scores and demographic and socioeconomic factors, as illustrated in Table 4 [82]. Age, ethnicity, and parents’ education of some Sri Lankan children living in the Kegalle district were significantly associated with ECOHIS scores/OHRQoL [22]. Higher ECOHIS scores were found in parents with lower education or income levels [86]. Among other factors, ECC status assessed by defs index was very strongly associated with OHRQoL, while many other factors showed modest associations. The present finding was aligned with some studies that demonstrated OHRQoL was positively associated with dental caries experience, age of parents/caregivers, and monthly household income [87, 88].

Long-term breastfeeding was also modestly associated with an increased impact on OHRQoL, as illustrated in the present study, and this finding was further confirmed by several published studies [17, 89]. The finding of the association between bottle-feeding and increased scores of ECOHIS was also in accordance with one study [8]. Poor oral hygiene practices, such as brushing less than twice daily, using non-fluoridated toothpaste, and unsupervised brushing, were significantly associated with poorer OHRQoL, reaffirming published evidence [6, 77].

Limitations and strengths

This study has certain limitations that should be acknowledged. Information on dietary habits, oral hygiene practices, and general health status was obtained through parental interviews, which are subject to recall bias and social desirability bias. In addition, some potential confounding factors, such as broader dietary diversity, were not captured. Detailed data on certain behaviors, snacking habits, including the flossing assistance, frequency of bedtime feeding, and cariogenic microbial flora, were also not collected, despite being reported as associated with ECC. Moreover, nutritional status was assessed using anthropometric measures only, which may not fully reflect biochemical or clinical indicators of malnutrition.

Furthermore, as a hospital-based study, these findings cannot be generalized to the entire child population in Sri Lanka or the region. However, the primary purpose of this study was to investigate the associations between ECC status, nutritional status, and their associations with OHRQoL, rather than to produce population-wide estimates. These associations are relevant beyond the local context for several reasons: first, the mechanisms linking ECC, nutritional status, and OHRQoL are universal and applicable across populations. Second, evidence from regional or local settings fills critical gaps in global knowledge, especially from underrepresented countries where such data are scarce. Third, the findings may help inform public health strategies and interventions in similar socio-economic settings worldwide. Finally, our study is limited by the lack of data on several unmeasured confounding variables that heavily influence cariogenic risk. Specifically, we did not account for baseline plaque index, oral microbial load, or protective exposures such as water fluoride levels. Additionally, nuanced dietary behaviors, including the exact quantity of sugar intake and bedtime feeding habits, were not captured, which may introduce residual confounding to our observed associations.

Despite these limitations, the study has several strengths. It used standardized, pre-validated tools and clinical assessments conducted by calibrated examiners from the Dental Teaching Hospital, University of Peradeniya, ensuring data quality and reliability. Furthermore, the novelty of this study lies in exploring the associations between ECC and nutritional status in Sri Lankan children, an area that has been scarcely investigated as a comprehensive study, accounting for other confounding independent variables such as OHRQoL, oral hygiene practices, general health status, eating habits, and demographic factors.

Conclusions

This hospital-based study demonstrates a significant, interconnected relationship between ECC, poor nutritional status, and diminished OHRQoL among Sri Lankan children aged 4–5 years. The findings emphasize that nutritional status, ECC, and OHRQoL are interconnected and influenced by a combination of socio-economic, dietary, and oral hygiene-related factors. Given the clear association found among ECC, poor nutritional status, and OHRQoL, as public health interventions, authorities should integrate oral health screenings and nutritional counseling into routine maternal and child health clinic visits. Additionally, sugar consumption reduction targets and parental education should be incorporated into sustainable preventive strategies.

Author contributions

HN: Writing – original draft; MN: Formal analysis, Writing – original draft (statistical analysis section); SG: Investigation, Data curation, Visualization; SDA: Investigation, Data curation, Visualization; CD: Visualization; RL: Conceptualization, Writing – review & editing; RJ: Conceptualization, Writing – review & editing; CH: Conceptualization, Writing – review & editing; STK: Conceptualization, Writing – review & editing.

Funding

This study was conducted when Sri Lanka was severely affected by inflation and bankruptcy. Therefore, this research received no specific grant from any funding agency. Printouts of questionnaires were taken from the National Institute of Fundamental Studies, Kandy, Sri Lanka, to conduct interviews.

Data availability

The raw data supporting the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The ethical clearance certificate (ERC/FDS/UOP/I/2021/45) was obtained from the Ethical Review Committee, Faculty of Dental Sciences, University of Peradeniya, on 12th October 2021. According to that, Information regarding the research was conveyed to parents or guardians by a typed information sheet in a manner and a language suitable to the individual’s level of understanding. The participants were given adequate opportunity to ask questions, and they were answered honestly and completely. Parents or guardians were made aware that they were free to withdraw their consent at any time without trepidation of consequences. Finally, the consent was obtained as a signature on a consent form that explicitly stated what was being consented to. Participants were not subjected to coercion, inducement, undue influence, or deception at any cost.

Consent for publication

Parents or guardians were informed that these data would be published in journal articles, and consent was obtained as a signature on a consent form.

Competing interests

The authors declare no competing interests.