Early Childhood Oral Health Impact Scale: Psychometric Evaluation in Portuguese Preschoolers

Abstract

Objective

to contribute to the validation of the Early Childhood Oral Impact Scale (ECOHIS) by studying its psychometric properties when applied to a Portuguese preschool population.

Methods

Cross-sectional study conducted with children aged between three and five-years-old. The non-probabilistic sample included two preschools in the municipality of Lisbon. Children who agreed to participate and whose guardians signed the informed consent were included. Data collection included a questionnaire, administered to the parents, and an intraoral examination of the children. The questionnaire included the Portuguese version of ECOHIS. The intraoral examination included the caries diagnosis according to the World Health Organization criteria. Discriminant validity compared the ECOHIS score between children with and without caries experience (Mann-Whitney U-test). Cohen's d was calculated to estimate the magnitude of the difference. Reliability analysis included Cronbach's α and test-retest. Construct validity was analyzed by the correlation between the ECOHIS score and dmft (Spearman's correlation). A significance level of 5% was used.

Results

The sample included 104 children (mean age 4.1 years). ECOHIS values were significantly different between children with and without caries (p=0.004). The Cohen's d was 0.84. The Cronbach's was 0.78, with no significant increase in value when eliminating any of the items. The test-retest showed significant correlation (r=0.76; p=0.01). There was a significant correlation between the ECOHIS score and caries experience (r=0.28; p=0.004).

Conclusion

The Portuguese version of the ECOHIS showed good psychometric properties, indicating that it is a reliable and valid tool to measure the impact of oral health in preschool children.

Introduction

Children's oral health is a relevant factor for their healthy development (1). The World Health Organization (WHO) defines oral health as "The state of the mouth, teeth and orofacial structures that enable individuals to perform essential functions such as eating, breathing, and speaking, and encompasses psychosocial dimensions such as self-confidence, well-being, and the ability to socialize and work without pain, discomfort, and embarrassment" (2). This definition incorporates the fact that oral health is an integral part of general health and well-being, and it is not only defined by the absence of caries or periodontal disease (3).

Dental caries continues to be considered a public health problem, especially during childhood, due to its high prevalence and severity. According to the World Health Organization (WHO) the prevalence of Early Childhood Caries is increasing in countries with low and middle income (4). In the most recent national survey in Portugal, it was found that 53% of 6-year-old children had dental caries (5). Some Portuguese studies in preschool (6, 7) found similar results indicating a high prevalence of caries in primary dentition, and unfortunately the World Health Organization's target of achieving 80% of children being caries-free by 2020 was not met. The presence of the disease is linked to social factors (6, 7) and behavioural factors (7), underscoring the importance of early and targeted interventions. Various strategies can be used to prevent and control dental caries in children, such as supervised brushing in kindergartens, oral health education and promotion for children and their caregivers, application of fluorides or sealants (1) or the use of an oral health passport (8). Also, besides fluorides some new agents are being tested with promising results in caries prevention (9, 10).

The study of Oral Health-Related Quality of Life (OHRQoL) is a highly relevant indicator since oral disorders can impair daily activities. It is defined as the impact of oral conditions on an individual's physical functioning as well as psychological and social well-being (11). The association between clinical oral health indicators and OHRQoL measurement provides a broad assessment of the patient's oral health, thus allowing this assessment to be improved globally by considering the individual's perception (11).

The impact of oral health conditions is not restricted to children, but it can also affect their families. The absence of proper oral hygiene practices in children can result in dental problems, generating feelings of guilt and anguish in their parents. In addition, the need to take children to the dentist and take time off work for dental treatment can have a financial impact on the family (12).

The quality of life of children and their families can be affected by oral pathologies, and since dental caries is the most prevalent disease, it is one of the diseases with the greatest impact (4).

OHRQoL is linked to social, psychological, and behavioral variables that affect children's oral health in addition to clinically recognized oral health issues (13), Kumar (14) has shown that certain family characteristics such as the parent's level of education, living in a non-traditional family, the crowdedness of the house, and the presence of siblings, are associated with poorer OHRQoL in children.

Locker et al. (15) highlighted the importance of including a family impact scale as an essential component for measuring OHRQoL in children since guardians play a vital role in children's health and long-term illness can also affect them. In childhood, it is therefore important to assess the impact of oral health on both the child and the family.

Several instruments have been developed and validated to study OHRQoL in children. Zaror et al. (16) identified five instruments available to measure OHRQoL in pre-school children: the Dental Discomfort Questionnaire (DDQ), the Early Childhood Oral Health Impact Scale (ECOHIS), the Michigan Oral Health-Related Quality of Life scale (MichiganOHRQoL), the Oral Health-related Early Childhood Quality of Life tool (OH-ECQOL) and the Scale of Oral Health Outcomes (SOHO-5). In the specific case of pre-school children, because they have lower levels of autonomy and understanding related to oral diseases and their health, the parents/guardians, as children’s caretakers, answered to the questionnaire to measure OHRQoL (17).

The ECOHIS was developed by Pahel et al. (18) for children between the ages of 3 and 5 years old. This scale was constructed from 45 items of the Child Oral Health Quality of Life Instrument (COHQoL) questionnaire. From these 45 items, the scale was reduced to 13 items: 9 aimed at the impact of oral problems on the child (symptoms, function, psychology, and the child's self-image/social interaction), and 4 aimed at the impact on the family (suffering and family function).

The ECOHIS has already been validated in several countries (19-41), showing, generally, good levels of reliability and acceptable construct validity. In Portugal, this scale was translated and validated by Costa in 2013 (Pt-ECOHIS), in a study carried out on a population of adolescents (42). However, no publications were found evaluating the psychometric properties of the Portuguese version in pre-school children, the age group for which the scale was originally developed. Therefore, this study aimed to contribute to the Portuguese validation of the Pt-ECOHIS by analyzing its psychometric properties when applied to a Portuguese pre-school population.

Materials and methods

This cross-sectional study included a non-probabilistic sample, that consisted of two kindergartens in Lisbon, one public and the other private with public support, that were selected for convenience, because of their proximity to the dental school. The population of the two kindergartens was 157 children, aged between 3 and 5 years, who agreed to participate and whose guardians signed the informed consent.

Data were collected between May 2022 and March 2023. They included a questionnaire administered to parents/guardians and an intraoral observation of the children.

The questionnaire collected information on sociodemographic characterization (age and gender of the child and mother's educational level) and on OHRQoL, using the Pt-ECOHIS (42). The questionnaire was distributed to parents with the help of kindergarten teachers. A concise description of the study, outlining its goals and methods, was included with each questionnaire, as well as the informed consent. The documents were placed in an envelope, hence they could be returned confidentially. A retest of the Pt-ECOHIS was made on 10% of the participants to analyze the reliability of the instrument.

An Intra-oral observer used the WHO dental caries criteria to collect information on decayed, missing, and filled teeth in the deciduous dentition (dmft) (43). The study was carried out by a single examiner on the premises of the participating schools, using an intraoral mirror and a CPI probe, in a room with natural light. The observer also used a frontal LED light. All measures to prevent cross-infection were used.

The field team included two investigators: the examiner, who made the intraoral observation, and another examiner who recorded the observation on the respective sheet. The examiner was trained by a researcher who had experience in using the WHO criteria. During the study, double observations were carried out on around 10% of the children, to calculate the intraobserver reproducibility of applying the criteria throughout the study (43). Calculations were made using Cohen's kappa, and perfect agreement was obtained (44).

The Pt-ECOHIS had 13 items and all the items have a response scale with 5 categories: 0 = never; 1 = almost never; 2 = occasionally; 3 = frequently; 4 = very frequently; 5 = I don't know. Three ECOHIS scores were calculated: the ECOHIS total score included the sum of the response values of all 13 response items on the scale, and could vary between 0 and 52; the ECOHIS child score refers to the sum of the 9 items on the child impact subscale and could vary between 0 and 36; and lastly, the ECOHIS family score refers to the sum of the values of the 4 items on the family impact subscale, with values between 0 and 16. The higher the score, the greater the impact. Therefore, it results in the worse OHRQoL. If the answer was "I don't know" the value was recoded, and the value “5” was replaced by the mean value of the other items in the subscale. Individuals with more than two "I don't knows" answers in the child subscale and one in the family section were eliminated from the study as described in the original study scale (18).

Descriptive statistical analysis included the calculi of absolute and relative frequencies of all the variables. The mean and standard deviation were also calculated for numerical variables. Inferential analysis used a statistical significance level of 5%. The Kolmogorov-Smirnov tested the normal distribution. The χ2 test was used to compare children with and without caries experience. The study of psychometric properties included the item frequency analysis, total inter-item correlation, internal consistency analysis, and test-retest. The internal consistency analysis (Cronbach's α) and the test-retest (Spearman's correlation) study the reliability of the scale. To assess discriminant validity, the differences in ECOHIS total between children with and without caries experience were compared using the Mann-Whitney U-test. To estimate the magnitude of the difference between the two ECOHIS means, Cohen's d was calculated, which in the case of ECOHIS is calculated by dividing the mean scores by the standard deviation (SD). A value of 0.2 indicates a small magnitude of difference; a value of 0.5 indicates a moderate difference; and the value greater than 0.8 indicates a large difference (45). Construct validity was studied with the correlation between the ECOHIS Total score and the dmft score, using Spearman's correlation. The Kappa statistics was used to verify the application of the caries diagnostic criteria throughout the study, demonstrating a perfect level of agreement (44).

This study was approved by the Health Ethics Committee of the Faculdade de Medicina Dentária da Universidade de Lisboa (registration number 202107).

Results

The sample included 104 children with a mean age of 4.1 (SD = 0.8). No significant differences were found in caries experience related to the child's sex or the mother's educational level (p>0.05). Caries experience was found to be higher in older children (p=0.014), with a difference quite evident between the 4- and 5-years-old children (Table 1).

Table 1. Sociodemographic characterization of the sample and comparison of children with and without caries experience.

Table 1 Sociodemographic characterization of the sample and comparison of children with and without caries experience.
	Total sample	Caries Experience	No Caries Experience	p value*
	% (n)	% (n)	% (n)
Age
3 years	27.89 (29)	17.24 (5)	82.76 (24)	0.014
4 years	32.69 (34)	14.71 (5)	85.29 (29)
5 years	39.42 (41)	41.46 (17)	58.54 (24)
Sex
Male	45.19 (47)	29.79 (14)	70.21 (33)	0.4
Female	54.81 (57)	22.81 (13)	77.19 (44)
Mother´s education level
Less than 9 years	5.10 (5)	60.00 (3)	40.00 (2)	0.4
9 years completed	18.37 (18)	22.22 (4)	77.78 (14)
12 years completed	24.49 (24)	29.17 (7)	70.83 (17)
High education completed	52.04 (51)	25.49 (13)	74.51 (38)
* χ2 independence test

All the items on the ECOHIS had a higher percentage of "never" responses. In the ECOHIS child subscale, the item with the greatest impact on OHRQoL was "tooth/mouth pain", where 21.15% of parents/guardians reported that their child had tooth or mouth pain, and 11.54% reported that their child had tooth or mouth pain occasionally or frequently. The items "difficulty drinking" (10.78%), "difficulty eating" (10.58%), and "problems falling asleep" (9.62%) also had some impact. Regarding the ECOHIS family subscale, it was found that 18.27% of parents had been upset or annoyed because of their child's dental problems, which was the item that revealed the greatest impact (Table 2).

Table 2. Distribution of ECOHIS scale items and scores.

Table 2 Distribution of ECOHIS scale items and scores.
Items	Never	Hardly ever	Occasional ally	Often	Very often	Mean (SD)
	% (n)	% (n)	% (n)	% (n)	% (n)
Child subscale
Pain in the teeth, mouth or jaws	78.85 (82)	9.61 (10)	7.69 (8)	3.85 (4)	0 (0)	0.37 (0.79)
Difficulty drinking hot or cold beverages	89.22 (91)	8.82 (9)	1.96 (2)	0 (0)	0 (0)	0.13 (0.39)
Difficulty eating some foods	89.42 (93)	4.81 (5)	4.81 (5)	0.96 (1)	0 (0)	0.17 (0.55)
Difficulty pronouncing words	92.00 (92)	1.00 (1)	4.00 (4)	2.00 (2)	0 (0)	0.16 (0.57)
Missed preschool, daycare, or school	93.28 (97)	2.88 (3)	2.88 (3)	0.96 (1)	0 (0)	0.11 (0.47)
Trouble sleeping	90.38 (94)	4.81 (5)	4.81 (5)	0 (0)	0 (0)	0.14 (0.47)
Irritable or frustrated	92.30 (96)	3.85 (4)	3.85 (4)	0 (0)	0 (0)	0.11 (0.42)
Avoided smiling or laughing	97.12 (101)	0.96 (1)	0.96 (1)	0.96 (1)	0 (0)	0.06 (0.36)
Avoided talking	99.03 (102)	0 (0)	0.97 (1)	0 (0)	0 (0)	0.04 (0.27)
Family subscale
Been upset	81.73 (85)	11.54 (12)	4.81 (5)	0.96 (1)	0.96 (1)	0.28 (0.69)
Felt guilty	84.62 (88)	7.69 (8)	4.81 (5)	2.88 (3)	0 (0)	0.26 (0.68)
Taken time off from work	93.27 (97)	3.85 (4)	2.88 (3)	0 (0)	0 (0)	0.10 (0.38)
Financial Impact	85.44 (88)	6.80 (7)	4.85 (5)	0.97 (1)	1.94 (2)	0.27 (0.76)

SD: standard deviation.

The mean ECOHIS total score was 2.2 (SD=4.1), with a minimum of 0 and a maximum of 25.9 (Table 3).

Table 3. Mean, median, maximum and minimum values of ECOHIS scores.

Table 3 Mean, median, maximum and minimum values of ECOHIS scores.
ECOHIS sum	Minimum	Maximum	Mean (SD)	Median
Child subscale	0	16.9	1.3 (2.7)	0
Family subscale	0	9.0	0.9 (1.8)	0
Total score	0	25.9	2.2 (4.1)	0

SD: standard deviation.

Reliability Analysis and Inter-Item Correlation

The internal consistency of the ECOHIS scores revealed Cronbach's α values of 0.78 for the child subscale, 0.66 for the family subscale, and 0.83 for the total value (Table 4).

Table 4. Reliability analysis – Internal consistency and test-retest reliability.

Table 4 Reliability analysis – Internal consistency and test-retest reliability.
ECOHIS (number of items)	Internal consistency reliability (Cronbach´s Alpha)	Test-retest reliability Spearman´s rank correlation
Child subscale (9)	0.78	r= 0.85; p=0.002
Family subscale (4)	0.66	r= 0.26; p=0.47
Total score (13)	0.83	r= 0.76; p=0.011

The correlation between the test and retest showed a significant and strong correlation for the child subscale (r=0.85; p=0.002) and for the total scale (r=0.76; p=0.011). However, in the family subscale, the correlation was lower, and it was not significant (r=0.26; p=0.47) (Table 4).

The item-total correlation analysis verified the non-redundancy of the items, with all the items being moderately correlated with each other. The item with the lowest value was the item 4, which corresponds to difficulty pronouncing words. No item led to a significant increase in Cronbach's α when eliminated (Table 5).

Table 5. Item-total correlation and Cronbach's α without item.

Table 5 Item-total correlation and Cronbach's α without item.
ECOHIS Items	r item-total	Cronbach´s alpha Without item
1. Pain in the teeth, mouth or jaws	0.58	0.81
2. Difficulty drinking hot or cold beverages	0.62	0.81
3. Difficulty eating some foods	0.38	0.83
4. Difficulty pronouncing words	0.12	0.85
5. Missed preschool, daycare or school	0.65	0.81
6. Trouble sleeping	0.54	0.82
7. Irritable or frustrated	0.67	0.81
8. Avoided smiling or laughing	0.5	0.82
9. Avoided talking	0.52	0.82
10. Been upset	0.40	0.83
11. Felt guilty	0.60	0.81
12. Taken time off from work	0.55	0.82
13. Financial Impact	0.49	0.82

Discriminant validity: ECOHIS in children with and without caries experience

When the ECOHIS scores of children with and without caries experience were compared, there were significant differences in the ECOHIS Child (p=0.004), ECOHIS Family (p=0.04), and ECOHIS Total (p=0.004). The mean values were higher in the group of children with caries experience, thus demonstrating a significant impact of this disease on OHRQoL (Table 6).

Table 6. Comparison of mean ECOHIS scores between the groups of children with and without caries experience.

Table 6 Comparison of mean ECOHIS scores between the groups of children with and without caries experience.
ECOHIS	With caries experience Mean (SD)	Without caries experience Mean (SD)	p-value*	Cohen´s d
Child subscale	2.9 (4.3)	0.7 (1.5)	0.004	0.68
Family subscale	1.6 (2.7)	0.7 (1.4)	0.04	0.42
Total score	5.5 (6.4)	1.4 (2.5)	0.004	0.84
* U-Mann-Whitney Test

Construct validity: correlation between ECOHIS and dmft scores

There was a positive and direct correlation between the scores of the ECOHIS and the dmft in the deciduous dentition. These results show that caries has an important impact on daily lives of children and their families. Spearman's correlation revealed a weak but highly significant correlation between the scores of the ECOHIS Child (r=0.28; p=0.004), the ECOHIS Family (r=0.20; p=0.04) and the ECOHIS Total (r=0.28; p=0.004) and the caries severity (dmft) in the deciduous dentition.

Discussion

The cross-cultural adaptation of an instrument requires some important steps such as translation and analysis of psychometric properties. The availability of culturally valid versions of instruments in different languages is important to obtain reliable and comparable data, considering social, cultural, and economic differences. The Pt-ECOHIS had already been developed by Costa in 2013 (42), but the author validated the scale in a population of adolescents encompassing the age of 12 years. Therefore, this study aimed to contribute to the validation of this instrument by studying the psychometric properties of the Pt-ECOHIS. However, it was applied to a pre-school population since that was the age group for which the scale was originally developed.

The questionnaire was self-administered to the child's legal guardians, as in the case of the original version of the ECOHIS (18). This method has advantages over interviewing, namely lower cost, and a reduction in interviewer bias (45). In addition, studies that have assessed quality of life on other scales, such as Malter (46), Sousa (47) and Puhan (48) found that the method of administration did not influence the results of the instruments.

The sample consisted of a total of 104 children, a size similar to several validation studies of the same scale (20, 24, 28-31, 35). The size of the sample for scale validation studies is a discussed topic, and this study followed the general rule adopted by many researchers, which states that the number of participants should be at least equal to the number of response options for each statement for each of the scale's items (49). With the ECOHIS having 13 items and 5 response hypotheses in all items, the sample should have at least 65 individuals. Furthermore, it is recommended that this minimum should include additional 20% of participants (49), which was also met in the study sample.

No significant differences were found when comparing the caries experience with the sex of the child or mother's education. The same pattern was seen in the validation study performed in China (24), unlike the study performed in Brazil (19), which found statistically significant differences between the mother's level of education and caries experience. This result can be attributed to the differences in sample sizes, which were smaller in this study (n=104) and in the Chinese study (n=111), compared to the Brazilian study (n=247) since larger sample is required to identify differences to similar extent. There may also be differences in the type of population included, and the population of Lisbon, as a capital city, may include a more differentiated population.

On the other hand, statistically significant differences were found between age and caries experience, which is in line with the results obtained in the validation studies carried out on the Chinese (24) and Australian populations (21). Since dental caries is characterized as a progressive and cumulative disease, this would be the expected result.

In all the items, there was a high percentage of answers in the "never" option, demonstrating the presence of a "floor effect”. Most parents reported that their child's quality of life was not affected by oral health, which can be partly attributed to the fact that the study sample was community-based and made up of children who did not seek medical treatment. As such, the pattern of responses was lower when compared to other studies that included samples of children who were seeking dental treatmen,t and could therefore have more oral health problems (22, 24, 28, 30, 32, 38). In studies that included community samples, the distribution of the items was more focused on the answers with the lowest values (18, 19, 23, 26, 34, 39, 40), as seen in the sample studied.

In the child impact subscale, the item with the greatest impact on OHRQoL was "tooth/mouth pain" (21.15%), followed by "difficulty drinking" (10.78%), "difficulty eating" (10.58%), and "trouble falling asleep" (9.62%). These results were similar to several studies that revealed that toothache was the item with the greatest impact on OHRQoL (18, 19, 21-24, 29-33). However, the studies carried out in Italy (37), Lithuania (28), and Turkey (35) revealed that the items with the greatest impact were "missing school", "difficulty eating" and "being frustrated", respectively.

In the family impact subscale, the items with the greatest impact were "being upset" (18.27%) and "feeling guilty" (15.38%), which is in agreement with several other studies validating the scale (19, 21-24, 26, 28, 29, 39). However, in the original version of ECOHIS (18), in Nigeria (30) and Taiwan (36), the item with the greatest impact was "missing work".

In this study, the mean ECOHIS Total score was 2.21 (SD=4.09), which can be considered low. Validation studies carried out in Colombia (25) and India (26) showed similar results, which is in line with other studies carried out in Portuguese populations (50, 51). Other ECOHIS validation studies showed higher means (22-24, 27, 31, 35), with a wide range of values between 3.7 (22) and 25.7 (27)_.

The values of Cronbach's α revealed good internal consistency in the case of family subscale and very good for the child subscale and the ECOHIS Total (52). The lower value of the ECOHIS family can be explained by its small number of items. There is a wide variation in the internal consistency of the ECOHIS items in various validation studies, with the value for the Total ECOHIS varying between 0.76 (36) and 0.95 (30). The value found in this study was similar to that in some studies (29, 32, 36-40) and lower than others (22, 36). On the other hand, there are also several validation studies with higher values (18, 19, 21, 23-28). One similarity found in most validation studies is that the Family subscale showed lower internal consistency values. However, it is very interesting to see that this subscale improves the internal consistency of the ECOHIS Total, which is why it is important to keep the family subscale as instrument.

The other parameter that studied the scale's reliability was the test-retest, which was applied to 10 children in the sample. One difficulty encountered in repeating the ECOHIS was the time interval between administering the two questionnaires, which was set to be between two and three weeks, but in some cases it varied up to two months due to the lack of cooperation of some parents in returning the questionnaire. Analysis of the correlation between the test and retest showed a significant and strong correlation (53) in the child subscale (r=0.85; p=0.002) and the total scale (r=0.76; p=0.011). However, in the family subscale, the correlation found was weaker and not significant (r=0.26; p=0.47). This lower value can be attributed to several reasons, namely small sample size, the number of items in the subscale, and the difficulty which was already reported, and related to the time between the test and retest.

The item-total correlation analysis found that the items were not redundant, with all items being moderately correlated with each other. Item 4, which corresponds to difficulty pronouncing words, obtained a value of 0.12, which was the only item below the recommended value of 0.20 (26). In the validation study carried out in Malaysia (29), there were also two items below the recommended item-total correlation, the item "missing school" and the item "financial impact," with values of 0.15 and 0.04, respectively. However, most studies (26-28, 36) obtained item-total correlations that were always above 0.20.

When the Cronbach's α value was analyzed when each of the items was removed, it was found that excluding the item "difficulty pronouncing words" would slightly increase the internal consistency value, but not considerably (it would increase 0.02). A similar result was also obtained in the validation study carried out in India (26).

The discriminant validity analysis confirmed that the ECOHIS allows us to differentiate between children with and without caries experience. The mean values of the scores were higher in the group of children with caries experience, thus indicating a significant impact of the disease on OHRQoL. This behavior of the ECOHIS has been demonstrated in several validation studies (18, 19, 21, 23, 24, 26, 29, 31, 32, 35-40).

Construct validity further supported the impact of dental caries by demonstrating a substantial correlation between the mean dmft values. Although significant, the correlation values showed a slightly weak correlation (53). The study carried out in Brazil (19) also found a slightly weak but significant correlation (r=0.40; p<0.01). Validation studies made in China (24), Peru (31), and Turkey (35) found significant medium or strong correlations.

Clinical measures, such as the dmft, focus on objective and observable aspects and do not always fully reflect the psychosocial impact that oral health can have on people's lives. For this reason, other studies use other variables to analyze the construct validity of the ECOHIS, namely measures that allow for the collection of perceptions and experiences reported by caregivers, such as parents' oral health perceptions of their child's. For this reason, numerous studies that validated the ECOHIS used this assessment of the child's oral or general health to analyze the construct validity of the scale (21-23, 26, 28, 30, 32, 38-40). Barbosa pointed out that valid and reliable information can be obtained from parents and children using appropriate questionnaire techniques (54). Although the parents' reports may be incomplete due to a lack of knowledge about some experiences, the information can still be useful (54). This assessment of the parents' perception and self-reporting of the child's oral health could have been an interesting variable to be included in this study, particularly because there may be situations in which dental caries lesions remain asymptomatic and, for this reason, can have less impact on OHRQoL.

Oral health-related quality of life is a relevant element for assessing the impact that oral health has on an individual's well-being. Considering patient's or parent's self-perceived oral health and needs in terms of social or psychological repercussions, complements conventional diagnostic criteria in clinical practice (55). The earlier the impact of oral disorders is assessed the higher the chance to intervene with educational and preventive strategies. Pre-schoolers benefit much from such interventions because this is the time when children develop physically and cognitively as well as form several habits and attitudes, including those pertaining to health and self-care (56).

The present study offers information relevant to health professionals by providing a contribution to the validation of the ECOHIS and by exploring the impact of oral problems on preschool children and their families. This information can be used to create public health initiatives targeted at populations at higher risk and the implementation of more adequate approaches to prevent, reduce, and control oral diseases in preschool children (55).

It would be useful to study the scale behavior in larger and probabilistic samples to allow true extrapolation of the results. However, despite the convenience sample, this study is an important contribution to validating the ECOHIS in the Portuguese preschool population.

Conclusion

The Portuguese version of the ECOHIS showed, generally, good psychometric properties when applied to the study population. The scale showed good internal consistency, acceptable test-retest reliability, good discriminant validity, and acceptable construct validity. Considering the behavior of the Portuguese version of the scale in this preschool population, it can be concluded that it is a reliable and valid tool for measuring the impact of oral health on preschool children and their families.