Utilization of caries risk assessment tools within the underserved population: a scoping review

Abstract

Objectives

Caries Risk Assessment (CRA) tools can be utilised to assess caries risk levels within underserved individuals to provide risk-based caries management. With no previous review mapping the evidence of CRA tools in underserved populations, a scoping review was conducted to provide a comprehensive view of the current literature and the utilisation of CRA tools in underserved populations. The main objectives of this review are as follows: (1) to comprehensively review CRA tools utilised, and (2) to highlight the important findings indicating the oral health status of underserved population subgroups.

Methods

A systematic search was performed using MEDLINE, EMBASE, Scopus, Google Scholar, and Dissertations & Theses Global (ProQuest). All relevant English-language papers published between January 2004 to June 2024 were identified. Retrieved references were imported and underwent 2-stage screening. The type of CRA tool was extracted as the primary outcome and oral health status of underserved subgroups were extracted as the secondary outcome.

Results

A total of 26 studies and nine different CRA tools were identified. Included studies examined caries risk in low-income families, people with disabilities, Indigenous peoples, refugees, veterans, and rural communities. Most studies indicated moderate to high caries risk and significant unmet oral health needs in underserved populations.

Conclusions

The underserved populations experience elevated caries risk and poor oral health status that require the attention of policymakers and practitioners. Significant heterogeneity across the utilised CRA tools was identified. Future research focusing on developing a standardised and appropriately validated CRA tool that can be utilised is necessary.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12903-025-05637-8.

Introduction

Dental caries remains one of the most prevalent chronic diseases worldwide, with approximately two billion people experiencing caries of permanent teeth and 514 million children experiencing caries of primary teeth [1]. The disease can negatively impact one’s quality of life due to pain, problems with eating, and social stigma—restricting activities at school and work for children and adults, respectively [2]. Oral diseases such as dental caries, continue to be a major contributor to the global economic burden [2]. Globally, 3.5 billion people are affected by oral diseases annually, costing approximately $710 billion in treatment costs and productivity losses [3]. In Canada, as per the last national health measures survey from 2007 to 2009, 2.26 million school days are missed annually due to oral diseases and one-third of all pediatric day surgeries performed account for treating caries in children [4]. Extensive evidence describes social determinants of health cause patients to experience unequal access to care, including income, race or ethnicity, education level, employment, and housing [5–9]. This leaves many underserved and equity-seeking individuals prone to experiencing dental caries [5, 10–12]. Underserved populations are defined as communities with limited access to care due to social, economic, cultural, linguistic, and/or physical barriers [13]. This includes older adults, those residing in rural regions, of low-literacy, or with low-income [14]. Ethnic minorities and people with disabilities (i.e., persons with tailored needs) often belong to one or more of the stated categories (i.e., older adults, rural, low-literacy, low-income) [14]. A systematic review by Schwendicke et al. (2015), revealed individuals of low socioeconomic status (SES) had significantly higher caries risk compared to high-SES individuals [6]. Another systematic review by Reda et al. (2018), verified that individuals of ethnic minorities or immigrants, those with low educational or economic background, and with no (or limited) insurance coverage showed lower utilisation of care—putting them at a higher risk of experiencing caries [15]. As underserved populations continue to experience a greater burden of caries, it is important to identify the susceptible subjects and the related risk factors to mitigate the current disparities in oral health care and dental caries. Additionally, the power of technology such as artificial intelligence and genomic data mapping can be harnessed to predict caries development in a susceptible community and target prevention methods for those that are truly at risk. However, such technologies may not be accessible for all communities, especially in resource-poor settings, warranting the need to identify the most appropriate CRA tool for that community to reduce caries incidence and health disparities [16].

Assessing caries risk level can identify high-risk communities and aid in managing the disease overtime according to identified risk factors. The multifactorial nature of caries makes it difficult for one preventative or restorative strategy to work for all patients hence individualised caries risk assessment and management is warranted [9]. Caries Risk Assessment (CRA) is defined as “the clinical process of establishing the probability of an individual patient to develop carious lesions over a certain period of time or the likelihood that there will be a change in size or activity of lesions already present” [17]. In 2003, the Journal of California Dental Association (CDA) proposed and published one of the first CRA tools in North America referred to as the CDA CRA form that classifies patients on a low, moderate, high or extreme risk of caries development [18, 19]. Overtime, the field of CRA quickly evolved as numerous cariologists, dental associations, and organizations developed similar tools to evaluate caries risk [20]. CRA tools aid dental care professionals assess caries risk early on based on different risk factors, protective factors, and disease indicators and help provide personalised care to each individual or population based on the assessment details [9, 21, 22]. Furthermore, CRAs can assist public health officials in the enhancement of oral health programs, support public dental clinics, allocate resources to reduce the redundancy of services, aid in cost effective prevention and implement other measures to maximise the accessibility of care for underserved populations [23]. Despite the numerous advantages of incorporating CRA into clinical practice, clinicians remain hesitant to adopt this approach. A recent study by Hans et al. (2024) reported that only 15% of clinicians integrated CRA tools into their routine caries assessments [24].

CRA tools such as Cariogram, CAMBRA, and others summarised in Additional file 1: Table S1. Typically examine various assessment criteria, such as previous caries experience, saliva flow, buffer capacity, dietary patterns, and/or sociodemographic characteristics [18, 19, 23–33].

Existing CRA tools present both similarities and differences in numbers and combinations of risk factors as predictors, evaluation methods, and interpretation of results—accounting for high heterogeneity in the current literature [34]. However, no standardised guidelines have been outlined despite the vast heterogeneity across the different CRA tools, while many lack appropriate validation and impact studies [35]. There is also a clear gap in the literature due to the lack of systematic or scoping reviews examining the utilisation of CRA tools in underserved populations. Since these populations face several inequalities and are disproportionately affected by a multitude of personal, social and environmental risk factors, they are at a higher risk of experiencing caries [5]. Hence, CRA tools are a pertinent component of delivering effective personalised oral health care, particularly benefiting underserved populations through the use of affordable yet effective preventive measures.

Although caries affects people of all ages, previous studies have mainly focused on investigating caries in children. Therefore, the aim of this study is to offer a comprehensive review of the utilisation of different CRA tools among both children and adult underserved populations. The main objectives of this review are as follows: (1) to comprehensively review CRA tools utilised, and (2) to highlight the important findings indicating the oral health status of underserved population subgroups.

Methods

Data sources and search strategy

The Joanna Briggs Institute (JBI) Reviewers Manual was followed for conducting this scoping review [36]. For any clarifications on the review process, the authors referred to Chap. 10 titled “Scoping reviews” to ensure our methodological rigour [36]. Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) reporting guidelines was followed [37]; a completed checklist has been included in Additional file 1: Table S1.

The protocol for this scoping review was registered with Open Science Framework (OSF) prior to beginning study screening (ID: https://osf.io/5hx93). The preregistered protocol outlines information regarding the study objectives, inclusion criteria, data extraction and analysis, and the presentation of results. This was done to ensure maximal transparency in the review process and show that our original intention aligns with the methodology.

Search strategy

The search strategy was developed by MP and AJ—in consultation with a university librarian. The following electronic databases were searched for eligible studies in June 2024: Ovid MEDLINE, Ovid Excerpta Medica (EMBASE), Scopus, Google Scholar, and Dissertations & Theses Global (ProQuest). Both peer-reviewed and grey literature sources were identified for a thorough review of the current literature. Grey literature was searched from the Dissertations & Theses Global (ProQuest) database using the same search strategy as other databases. MeSH terms were combined by reviewing previously completed literature and referencing keywords in relevant citations; a list of CRA tools from a review conducted by Featherston et al. (2021) [9] was referenced for search terms included in “caries risk assessment” category (Additional file 1: Table S3.). For a detailed view of our full search strategy, refer to Additional file 1: Table S4.

Reference management & screening

Retrieved references were imported from each database search into Covidence [36]. After the removal of duplicates, remaining references entered a 2-stage screening process: (1) title and abstract screening; (2) full-text screening. Two reviewers (MP and FP) screened articles independently for eligibility. Any disagreements were resolved through discussion.

Inclusion and exclusion criteria

Inclusion criteria

This review included all study designs, including observational (cross-sectional, prospective, and retrospective) and experimental studies (randomised controlled trials) published in the last 20 years. The study population included “underserved populations” (e.g., refugees, low-income, seniors, immigrants, Indigenous peoples, rural communities, people with disabilities, etc.,) of all ages—defined as communities with limited access to care due to social, economic, cultural, linguistic, and/or physical barriers, as well as a lack of familiarity with the healthcare system [13]. English-language papers from any global context were included. The primary outcome of interest was the utilised CRA tool and its respective result indicating risk status; any measurement assessing the oral health status of study participants was collected as the secondary outcome of interest.

Exclusion criteria

Studies were excluded if a non-structured CRA tool (i.e., customised survey questionnaire) was utilised, did not discuss or report on the oral health status, did not include underserved participants and/or was written in a non-English language.

Data charting

Microsoft Excel (2024) was used to create a data extraction table and manage records retrieved from searches. The data extraction form was piloted on 15% of the included studies and refined to ensure that all relevant data were extracted. The following data were extracted: author & year, title, study ID, country, aim, study design, method of recruitment, sample size, age group, underserved group, CRA tool with examined population risk status, and the oral health status of study participants. Data extraction was completed and verified independently by MP and AJ. Extracted data were reviewed for any discrepancies with the information reported in the study manuscripts.

Risk of bias assessment of included studies

Although risk of bias assessment is not required for scoping reviews, the authors believed it to be valuable given the inclusion of studies of various methodology and settings in our review. Understanding the possible sources of bias will be crucial in examining the quality of evidence presented by the included studies. All studies were assessed for their risk of bias using the CLARITY Group’s Risk of Bias Instruments, which included five domains for the cross-sectional studies tool [38], eight domains for the cohort studies tool [39], and six domains for the randomised controlled trials tool [40]. Additional file 1: Figure S1 presents a detailed risk of bias assessment for each study with domain descriptions. These instruments were chosen for their simplicity and ability to provide a comprehensive summary. Assessment results scaled from “definitely yes (low risk of bias),” “probably yes (moderate risk of bias),” “probably no (moderate risk of bias),” and “definitely no (high risk of bias).” The domain was marked as “unclear” if the study did not explicitly provide related information.

Results

Study characteristics

The search strategy yielded 1143 citations: MEDLINE (n = 261), Embase (n = 283), Scopus (n = 446), Google Scholar (n = 7), ProQuest (Dissertations & Theses; n = 146). After the removal of 500 duplicates and 598 irrelevant studies during title and abstract screening, the authors screened a total of 45 full-text articles. During full-text screening, 19 were excluded due to wrong outcomes (n = 5), full-text not found (n = 2), CRA tool not utilised (n = 6), oral health status not reported (n = 3), and wrong study population (n = 3). The remaining 26 studies met the inclusion criteria and underwent data extraction. Our search results and screening process are detailed in Fig. 1.

Fig. 1.

PRISMA diagram of study identification and selection

Description of included studies

Extracted data have been summarised in Table 1. Most studies originated from the United States (n = 13) [20, 23, 33, 41–50]. Others originated from countries in Asia, Europe, the Middle East, and South America [21, 51–62].

Table 1.

Characteristics of the included studies and findings

Author(s), Year	Country	Type of study	Sample size (n)	Age group (mean ± SD or mean or range)	Underserved subgroup	Objective 1: CRA tool		Objective 2: oral health status
Alzahrani & Bhat, 2024	Saudi Arabia	Cross-sectional	89	Adults (Mean ± SD = 30.11 ± 4.39 years)	People with Disabilities	Cariogram (validated)	CAC (Mean ± SD) = 37.9 ± 8.6	DMFT (Mean ± SD) = 10.9 ± 4.6
Berger, 2012	United States	Cross-sectional	36	Children (Range = 2 to 6 years)	Rural	AAPD-CAT (non-validated)	Low Risk = 2.8% Moderate Risk = 22.2% High Risk = 75%	Clinical Conditions (%): White Spot Lesions (3.8%), Rampant Decay (22%), Untreated Decay (25%), Evidence of ECC (27.8%)
Canares et al., 2019	United States	Cross-sectional	32	Children (Range = 0 to 17 years)	Low SES	ADA-CRA (non-validated)	Low Risk = 0% Moderate Risk = 15.6% High Risk = 84.3%	Clinical Conditions (%): No new carious lesions or restorations in last 24 months for age < 6, in last 36 months age > 6 (62.5%); 1 or 2 new carious lesions or restorations in last 36 months for age > 6 (9.4%); Carious lesions or restorations in last 24 months, age < 6, 3 or more in 36 months age > 6 (28.1%); Teeth missing due to caries in past 36 months (3.1%); Visible plaque (68.8%); Dental/orthodontic appliances present (0%); Salivary flow problems (0%)
Chaffee et al., 2017	United States	Cohort (Retrospective)	2188	Children (Mean = 40.2 months)	Low-income	CAMBRA (validated)	Low Risk = 11.3% Moderate Risk = 13.4% High Risk = 72.6% Extreme Risk = 2.7%	Clinical Conditions (%): Evident tooth decay or white spots (44.2%), Heavy dental plaque (28.6%), Recently placed restorations (9.1%)
Chang & Kim, 2014	South Korea	Cohort (Prospective)	64	Children & Adults (Mean ± SD = 23.7 ± 9.3 years)	People with Disabilities	Cariogram (validated)	Low Risk = 14.1% Moderate Risk = 28.1% High Risk = 57.8%	DMFT at baseline: (Mean ± SD) = 9.2 ± 6.4 iDMFT at follow-up: (Mean ± SD) = 2.1 ± 4.2
Chang et al., 2014	South Korea	Cross-sectional	102	Children & Adults (Mean ± SD = 23.8 ± 9.3 years)	People with Disabilities	Cariogram (validated)	CAC (Mean ± SD) = 28.1 ± 20.4	DMFT (Mean ± SD) = 8.8 ± 5.7
Curtis et al., 2018	Ecuador	Cross-sectional	131	Children (Range = 6 to 12 years)	Rural	CAMBRA (validated)	Not reported	Clinical Conditions (%): Obvious white spot lesions, cavitated carious lesions, or obvious decay (89.3%), Restorations present (63.6%), Visible plaque accumulation (70.5%), Gingivitis (38.9%), Obvious fluorosis (29.2%)
Custodio-Lumsden et al., 2016	United States	Cross-sectional	108	Children (Mean ± SD = 8.5 ± 1.8 years)	Low-income	MSB (non-validated)	Low Risk = 56.5% Moderate Risk = 41.6% High Risk = 1.8%	Clinical Conditions (%): Oral Mutans Low (32.4%), Moderate (60.2%), High (4.6%), Very High (2.8%) Visible Plaque None (25%), Mild (46.3%), Moderate (24.1%), Severe (4.6%) ‘Decalcifications’ Yes (34.3%), No (65.7%) ECC None (49%), ECC (13%), S-ECC (38%)
Francis, 2018	India	Cross-sectional	400	Children (Range = 6 to 18 years)	People with Disabilities	ADA-CRA (non-validated)	Low Risk = 0% Moderate Risk = 28% High Risk = 72%	DMFT (Mean ± SD): Mild ID (1.4 ± 2.2), Moderate ID (1.7 ± 2.2), Severe ID (1.1 ± 1.6) deft (Mean ± SD): Mild ID (0.77 ± 1.7), Moderate ID (0.6 ± 1.7), Severe ID (0.8 ± 1.91) Clinical Conditions (%): Cavitated/Incipient lesion (68.2%), Visible plaque (33%), Unusual tooth morphology (0%), Exposed root surface (0%), Dental Appliance (0%), Xerostomia (6.8%)
						AAPD-CAT (non-validated)	Low Risk = 0% Moderate Risk = 31.2% High Risk = 68.8%	DMFT (Mean ± SD): Mild ID (1.4 ± 2.2), Moderate ID (1.7 ± 2.2), Severe ID (1.1 ± 1.6) deft (Mean ± SD): Mild ID (0.7 ± 1.7), Moderate ID (0.61 ± 1.7), Severe ID (0.84 ± 1.9) Clinical Conditions (%): Dental Appliance (0%), Xerostomia (6.8%)
						CDA CRA (validated)	Low Risk = 23.8% Moderate Risk = 22% High Risk = 54.2%	DMFT (Mean ± SD): Mild ID (1.4 ± 2.2), Moderate ID (1.7 ± 2.2), Severe ID (1.1 ± 1.6) deft (Mean ± SD): Mild ID (0.77 ± 1.7), Moderate ID (0.61 ± 1.7), Severe ID (0.84 ± 1.9) Clinical Conditions (%): Cavitated/Incipient lesion (68.2%), Visible plaque (33.0%), Exposed root surface (0%), Dental Appliance (0%), Xerostomia (6.8%)
Halasa-Rappel et al., 2019	United States	Cross-sectional	1520	Children & Adults (Range = 1 to 20 years)	Low-income	CAMBRA (validated)	CAMBRA-C: Low Risk = 5% Moderate Risk = 40% High Risk = 55% CAMBRA-A: Low Risk = 33% Moderate Risk = 5% High Risk = 62%	DMFT/dmft-all: DMFT of 0 (46%), DMFT of 1 (8.0%), DMFT of greater than 1 (47%)
						Cariogram (validated)	Low Risk = 32% Moderate Risk = 5% High Risk = 62%
						AAPD-CAT (non-validated)	AAPD nondental-I: Low Risk = 12% Moderate Risk = 0% High Risk = 88% AAPD-C: Low Risk = 12% Moderate Risk = 5% High Risk = 83% AAPD-A: Low Risk = 12% Moderate Risk = 3% High Risk = 85%
						ADA-CRA (non-validated)	ADA-C: Low = 0% Moderate = 0% High = 100% ADA-A: Low Risk = 30% Moderate Risk = 50% High Risk = 19%
						BCH-CRA (non-validated)	BCH-C: Low Risk = 66% Moderate Risk = 23% High risk = 11%
Iype et al., 2021	India	Cross-sectional	150	Children (Range = 6 to 18 years)	People with Disabilities	Cariogram (validated)	CAC (Mean %): Mild ID = 64% Moderate ID = 58% Severe ID = 47%	DMFT (Mean): Mild ID (1.5), Moderate ID (1.8), Severe ID (3.1)
Jurasic et al., 2021	United States	Cohort (Retrospective)	57,675	Adults (Mean ± SD = 60.1 ± 12.9 years)	Veterans	ADA-CRA (non-validated)	Low Risk = 50.1% Moderate Risk = 33.2% High Risk = 16.8%	Prior treated caries (before follow-up): 0 (67.9%), 1 (12.0%), 2+ (20.1%) Current treated caries (carious lesions detected at first follow-up exam, 12 months prior): 0 (56.5%), 1 (17.1%), 2 (9.7%), 3+ (16.7%)
Laniado et al., 2019	United States	Cross-sectional	543	Children (Mean = 5.4 years)	Low-income	AAP-OHRAT (non-validated)	Not reported	Clinical Conditions (%): White spots (26.5%), Obvious decay (23.8%), Restorations present (16.4%), Visible plaque (41.1%), Gingivitis (24.7%), dental caries experience (39.6%)
Makan et al., 2019	Jordan	Cross-sectional	125	Children (Mean = 8.3 years)	Refugees	CAMBRA (validated)	Low Risk = 13.6% Moderate Risk = 36.0% High Risk = 50.4%	DMFT (Mean ± SD) = 3.6 ± 9.8 dmft (mean ± SD) = .9 ± 4.7 Clinical Conditions (%): missing teeth (20.8%), more than two filled teeth (6.4%), dental fluorosis (44.8%), mild gingivitis (24.8%), moderately severe gingivitis (72.8%), severe gingival inflammation (2.4%)
Malhotra et al., 2019	India	Cross-sectional	50	Children (Mean = 12 years)	Rural	Cariogram (validated)	CAC (Mean ± SD) = 58.0 ± 18.9	Caries prevalence = 74%
Murphy & Larsson, 2017	United States	Cross-sectional	47	Children (Mean ± SD = 27.81 ± 20.62 months)	Indigenous	AAP-OHRAT (non-validated)	Low Risk = 7.8% High Risk = 91.1%	Clinical Conditions (%): White spots and cavitated carious lesions past 12 months (63.8%), Obvious decay (25%), Restorations present (41.6%), Visible plaque accumulation (55.5%), Gingivitis (2.7%), Teeth present (76.6%), Healthy teeth (27.7%)
Nishi et al., 2019	Republic of Ireland	Randomized Controlled Trial	111	Adults (Mean ± SD = 38.9 ± 12.8 years)	Low-income	Cariogram (validated)	CAC (Mean ± SD): Personalised = 39.3 ± 20.2 Non-personalised = 36.5 ± 23.4	DMFS at baseline (Mean ± SD): Personalised (31.0 ± 19.4), non-personalised (31.7 ± 18.6) DMFS at follow-up (Mean ± SD): Personalised (46.2 ± 19.6), non-personalised (42.8 ± 22.0)
Ogawa et al., 2018	United States	Cross-sectional	228	Children (Range = 0 to 17 years)	Refugees	AAPD-CAT (non-validated)	Low Risk = 46.1% Moderate Risk = 11.0% High Risk = 43.0%	Treatment urgency (%): Routine (50.9%), Urgent (38.2%), Emergency (11%)
Patil et al., 2011	India	Cohort (Prospective)	54	Children (Mean ± SD = 11.9 ± 2.71 years)	People with Disabilities	Cariogram (validated)	CAC at baseline = 44% CAC at follow-up = 87%	DMFT (%) at baseline: DMFT of 0 (31%), DMFT of 1 (17%), DMFT of 2 (17%), DMFT of ≥ 3 = 35% DMFT (%) at 10 month follow-up: DMFT of 0 (30%), DMFT of 1 (15%), DMFT of 2 (10%), DMFT of ≥ 3 (37%)
Prokshi et al., 2022	Republic of Kosovo	Cohort (Prospective)	100	Children (Mean ± SD = 6.0 ± 1.1 years)	Rural	Cariogram (validated)	CAC (Mean ± SD) = 27.7 ± 11.8	dmft (mean ± SD) = 8.0 ± 2.3
Rajih, 2017	United States	Cross-sectional	110	Children (Mean ± SD = 3.6 ± 1.2 years)	Low-income	AAPD-CAT (non-validated)	Low Risk = 30% Moderate Risk = 18% High Risk = 52%	Caries Prevalence (%): ECC (49%), S-ECC (30%) Plaque on teeth (%): No (38%), Yes (61%)
Shah et al., 2020	Saudi Arabia	Cohort (Prospective)	163	Children (Range = 6 to 15 years	People with Disabilities	Cariogram (validated)	CAC at baseline = 56% CAC at final = 73%	DMFT/dmft (Mean ± SD) = 3.2 ± 3.3 DMFS/dmfs (Mean ± SD) = 6.4 ± 9.4
Southward et al., 2008	United States	Cross-sectional	422	Children (Mean = 38.5 months)	Low SES	AAPD-CAT (non-validated)	Not reported	Clinical Conditions (%): Evidence of oral diseases (33%), Plaque (62.3%), Difficulty chewing (1.6%), Difficulty biting hard (2.9%), Sensitive to temperature (6.3%), Sensitive to sweets (3.9%)
Watanabe et al., 2016	United States	Cohort (Prospective)	315	Children (Mean ± SD = 8.4 ± 4.0 years)	Low-income	WesternU CDM (non-validated)	Low Risk = 2.2% Moderate Risk = 15.5% High Risk = 82.2%	Cavitated Lesions (%): Yes (72.1%), No (27.9%)
Yoon et al., 2012	United States	Cross-sectional	471	Children (Mean ± SD = 24.7 ± 6.6 months)	Low SES	AAPD-CAT (non-validated)	Not reported	S-ECC (%): Yes (48.6%) Clinical Conditions (S-ECC): > 1 area of demineralization (57.2%), gingivitis or visible plaque (94.8%), enamel hypoplasia (5.2%) Clinical Conditions (No S-ECC): > 1 area of demineralization (0%), gingivitis or visible plaque (33.1%), enamel hypoplasia (2.9%)
Zukanović et al., 2007	Bosnia and Herzegovina	Cross-sectional	38	Children (Mean = 12 years)	Low SES	Cariogram (validated)	CAC (Mean ± SD) = 55.1 ± 3.7	DMFT (Mean ± SD) = 5.4± 2.5

Note CAC = Chance of Avoiding Caries; DMFT = Decayed, Mean, or Filled Teeth; DMFS = Decayed Mean, or Filled Surfaces; deft = decayed extracted or filled teeth; ECC = Early Childhood Caries; S-ECC = Severe Early Childhood Caries; SES = Socioeconomic Status; ID = Intellectual Disabilities; CAMBRA = Caries Management by Risk Assessment; ADA-CRA = American Dental Association Caries Risk Assessment; AAPD-CAT = American Academy of Pediatric Dentistry Caries Assessment Tool; MSB = My Smile Buddy; CDA-CRA = California Dental Association Caries Risk Assessment; BCH-CRA = Boston Children’s Hospital Caries Risk Assessment

*CAMBRA-C = CAMBRA for children 0–5 years of age; CAMBRA-A = CAMBRA for people 6 years and older; AAPD-nondental-I = AAPD-CAT for nondental providers for children 0–3 years of age; AAPD-C = AAPD-CAT for children 0–5 years of age; ADA-C = ADA-CRA for children 0–6 years of age; BCH-C = BCH-CRA for children 0–5 years of age; Cariogram-all = Cariogram for all ages; AAPD-A = AAPD-CAT for people 6 years and older; ADA-A = ADA-CRA for people 7 years and older

Majority of the studies were observational in design (n = 25), including cohort and cross-sectional studies [20, 21, 23, 33, 41–60, 62]. One study conducted by Nishi et al. (2019) [61], was a randomised controlled trial. Eight studies included fewer than 100 participants [41, 42, 47, 51, 53, 54, 58, 59], while 14 studies included between 100 and 500 participants [20, 21, 33, 44, 48–50, 52, 55–57, 60–62]. Four studies included more than 500 participants [23, 43, 45, 46] while most of the studies had fewer than 500 participants, ranging from 32 to 471.

CRA were conducted exclusively in children by most authors (n = 20) [20, 21, 33, 41–44, 46–50, 52, 53, 56–60, 62]. A few studies examined caries risk in adults only (n = 3) [44, 50, 60] and others included both children and adult participants (n = 3) [23, 54, 55].

Seven different underserved subgroups were identified in this review, including people with disabilities (n = 7) [51, 52, 54–57, 59], refugees (n = 2) [48, 60], Indigenous peoples (n = 1) [42], rural communities (n = 4) [21, 41, 58, 62], individuals with low SES (n = 4) [20, 42, 50, 53], individuals with low-income (n = 7) [23, 33, 43, 44, 46, 49, 61], and Veterans (n = 1) [45].

Twelve studies utilized CRA tools assessing social determinants of health (n = 12) [20, 21, 23, 37–39, 42, 43, 45, 46, 52, 57]. Resources required for each tool were variable, where most were administered by dental providers (n = 7) [21, 23, 38, 39, 49, 50, 63] while some were administered by non-dental providers (n = 8) [20, 23, 41, 43–45, 51, 56]. Other studies did not specify the type of personnel used for CRA assessment with their respective tool (n = 11) [37, 41, 43, 47–49, 53–56, 58]. Lastly, three studies implemented preventive measures to detect oral health improvements based on CRA results (n = 3) [20, 47, 54]. However, none mentioned whether the utilised CRA tool was sensitive or appropriate for detecting oral health improvements following the interventions.

Risk of bias assessment

The risk of bias assessment conducted using the CLARITY Group’s Risk of Bias Instruments [37, 38, 63] highlighted differences between studies in terms of the reliability of the results reported. Of the 18 cross-sectional studies, nine studies exhibited high risk of bias in at least one of the domains, most frequently observed in the lack of clinical sensibility and validation of the CRA tools utilized [18, 39, 40, 42, 44–48]. Of the seven cohort studies, five studies exhibited high risk of bias in at least one of the domains, most frequently observed in the lack of adjustment for prognostic factors [31, 41, 50, 57, 60]. Finally, the randomised controlled trial indicated high risk of bias for potential selective reporting and other issues related to protocol violation [59]. The risk of bias assessment results has been summarised in Additional file 1: Figure S1.—colour-coded with green representing low risk of bias, yellow representing moderate risk of bias, red representing high risk of bias, and blue representing unclear information.

Objective 1: CRA tool utilisation & risk status

A total of nine different CRA tools were utilised across the studies: (1) Cariogram; (2) CAMBRA; (3) American Dental Association Caries Risk Assessment (ADA-CRA); (4) American Academy of Pediatric Dentistry Caries Assessment Tool (AAPD-CAT); (5) American Academy of Pediatrics’ Oral Health Risk Assessment Tool (AAP-OHRAT); (6) MySmileBuddy (MSB); (7) California Dental Association Caries Risk Assessment (CDA-CRA); (8) Boston Children’s Hospital Caries Risk Assessment (BCH-CRA); (9) and WesternU CDM.

Cariogram, a validated computer-based CRA tool, was the most frequently utilised among the included studies (n = 11) [23, 51–55, 57–59, 61, 62]. Alzahrani & Bhat (2024) characterised caries risk in disabled adults and found a low chance of avoiding caries (CAC) of less than 50% [51]. Zukanović et al. (2007), revealed that children with low SES had a moderate mean CAC [53]. Chang et al. (2014), and Chang & Kim (2014) assessed caries risk in both children and adults with intellectual disabilities (ID), with both studies reporting low CAC [54, 55]. Halasa-Rappel et al. (2019), used the National Health and Nutrition Examination Survey (NHANES) data, focusing on patients enrolled in Medicaid or CHIP—indicating low-income; assessment results revealed that majority of the participants were at high-risk [23]. Iype et al. (2021), built caries risk profiles in children with ID and found that children with severe ID had low mean CAC [57]. Malhotra et al. (2019), compared mean CAC of children in rural and urban areas, discovering lower mean CAC in rural children [58]. Nishi et al. (2019), compared the impact of personalised versus non-personalised text messaging on CAC in low-income adults; both treatment groups (personalised and non-personalised) showed low mean CAC at baseline and follow-up [61]. Patil et al. (2011), tested the effect of a preventive program on caries risk in children with ID, resulting in the improvement of a low mean CAC at baseline to a high mean CAC at follow-up [59]. Prokshi et al. (2023), examined caries risk of children in rural areas, where most were found to be at high caries risk [62]. Shah et al. (2020), determined caries risk of Special Care School Children (SCSC) with disabilities and analyzed the effectiveness of preventive oral health measures; results at baseline and at six months showed a dramatic improvement of mean CAC [52].

Of the 26 included studies, four utilised CAMBRA [21, 23, 43, 60]. Chaffee et al. (2017), provided CRA results of low-income patients, where 72.6% were at high-risk [43]. Halasa-Rappel et al. (2019), reported two versions of CAMBRA: CAMBRA-C for children of 0–5 years of age and CAMBRA-A for people 6 years and older [23]. CAMBRA-C results reported that 55% were at high-risk, while CAMBRA-A reported that 62% were at high-risk [23]. Makan et al. (2019), generated caries risk scores of Syrian refugee children, where half of the participants were at high-risk [60]. Curtis et al. (2018), utilised CAMBRA to conduct CRA in children from rural areas but did not report on the risk status [21].

Four studies provided data on the utilisation of ADA-CRA [23, 42, 45, 56]. Canares et al. (2019), reported that 84.3% of children with low SES were at high-risk [42]. Results reported from ADA-CRA by Francis et al. (2017), found that the majority of children with ID were at high-risk [56]. Halasa-Rappel et al. (2019), reported two versions of ADA-CRA: ADA-C for children of 0–6 years of age and ADA-A for people 7 years and older [23]. ADA-C results showed that 100% were at high-risk, while ADA-A results showed that 19% were at high-risk [23]. Jurasic et al. (2021), evaluated the validity of ADA-CRA within Veterans, where the results indicated that 16.8% were at high-risk [45].

Seven studies utilised AAPD-CAT to characterise caries risk in underserved children [20, 23, 41, 48–50, 56]. Berger (2021) conducted CRA in children from rural areas and found the majority to be at high-risk [41]. Results reported from AAPD-CAT by Francis found the majority of children with ID were classified as high-risk [56]. Halasa-Rappel et al. (2019), reported three versions of AAPD-CAT: AAPD-nondental-I administered by non-dental providers for children of 0–3 years of age, AAPD-C for children of 0–5 years of age, and AAPD-A for people 6 years and older [23]. All three versions (AAPD-nondental-I, AAPD-C, and AAPD-A) reported that most were at high-risk [23]. Ogawa et al. (2018), graded caries risk levels in refugee children and found 43% at high-risk [48]. Rajih (2017) showed approximately half of children from low-income families were at high-risk [49]. Southward et al. (2008), and Yoon et al. (2012), utilised AAPD-CAT to assess CRA in children with low SES but did not report on the risk status [20, 50].

Murphy & Larsson (2017) utilised AAP-OHRAT to assess caries risk in American Indigenous children, who were previously found to experience a higher rate of untreated caries than the general population; results indicated that majority of children were categorised as high-risk [47]. Laniado et al. (2019), conducted CRA in children of low-income families by utilizing AAP-OHRAT but did not report on the risk status [46].

One study utilised MSB to conduct CRA in a low-income Hispanic population, where only 2% were classified as high-risk [44]. Francis (2017) reported caries risk levels in children with ID utilizing CDA-CRA, discovering approximately half of the participants at high-risk [56]. Halasa-Rappel et al. (2019), utilised BCH-CRA to classify caries risk levels in low-income children and adults, where majority were found to be at low-risk [23]. Watanabe et al. (2016), utilised WesternU CDM tool and discovered that most children from low-income families were at high-risk [33].

Objective 2: oral health status of underserved subgroups

Seven studies reported on the oral health status of people with disabilities [51, 52, 54–57, 59]. Alzahrani & Bhat (2024) reported a mean Decayed Missing Filled Teeth (DMFT) score of 10.97 in adults with ID (mean age = 30.11 years) [51]. Francis et al. (2017), and Iype et al. (2021), reported mean DMFT scores of children (age of 6–18 years) based on the severity of their ID (mild, moderate, severe), ranging from 1.18 to 3.14 [56, 57]. Shah et al. (2020), reported mean DMFT/dmft score of 3.2 in children with ID (age of 6 to 15 years) [52]. Patil et al. (2011), found that 35% of children with ID (mean age = 11.9 years) had a DMFT score ≥ 3 at baseline and 37% at follow-up after 10 months [59]. Chang & Kim (2014) found a mean DMFT score of 9.2 baseline and increased DMFT (iDMFT) score of 2.1 at follow-up (mean follow-up = 16.3 months) in children and adults (mean age = 23.7 years) [54]. Similarly, Chang et al. (2014), grouped children and adults (mean age = 23.8 years) and found a mean DMFT score of 8.8 [55].

Zukanović et al. (2007), reported a mean DMFT score of 5.40 in 12-year-old children with low SES [53]. Canares et al. (2019), reported the majority of children with low SES (age of 0–17 years) had visible plaque [42]. Southward et al. (2008), reported more than half of low-SES children (mean age = 38.5 months) had visible plaque and 33% showed evidence of oral diseases [50]. Yoon et al. (2012), found that approximately half of the low-SES children (mean age = 24.7 months) experienced severe early childhood caries (S-ECC) [20].

Four studies reported on the oral health status of rural communities [21, 41, 58, 62]. Berger (2012) reported evidence of early childhood caries (ECC), rampant decay and untreated decay in rural children (age of 2 to 6 years) [41]. Curtis et al. (2018), reported evidence of white spots and cavitated carious lesions, decay, and visible plaque accumulation in majority of the rural children (age of 6 to 12 years) [21]. Malhotra et al. (2019), reported a caries prevalence rate of 74% in 12-year-old rural children [58], while Prokshi et al. (2023), reported a mean dmft score of 8.01 in rural children with a mean age of 6 years [62].

Seven studies reported on the oral health status of low-income individuals [23, 33, 43, 44, 46, 49, 61]. Five studies reported prevalence rates of caries in low-income children, ranging between 39.6 and 79% [33, 43, 44, 46, 49]. Halasa-Rappel et al. (2019), found that 55% of low-income children and adults (age of 1 to 20 years) had a DMFT/dmft score of 1 or greater [23]. Nishi et al. (2019), reported mean Decayed, Missing, and Filled Surfaces (DMFS) baseline scores of 31.0 and 31.7 in personalised group and non-personalised group, respectively in low-income adults (mean age = 38.9 years) [61].

Makan et al. (2019), reported a mean DMFT score of 3.6 and dmft score of 2.9 in Syrian refugee children (mean age = 8.3 years) [60]. Ogawa et al. (2018), categorised refugee children (age of 0 to 17 years) according to treatment urgency: 50.9% were classified as routine (no decay visible, oral hygiene appointment recommended), 38.2% were classified as urgent (evidence of clinical decay, restorative treatment recommended), and 11.0% were classified as emergency (infection, abscess, orofacial pain, acute trauma, and/or multiple gross tooth decay) [48].

Jurasic et al. (2021), reported the percentage of Veterans (mean age = 38.9 years) with prior treated caries (backlog of caries developed prior to becoming a Veterans Affairs dental patient) and current treated caries (carious lesions detected at the first follow-up exam linked to the CRA): 32.1% had one or more prior treated caries, while 43.5% had 1 or more current treated caries [45].

Finally, Murphy and Larsson (2017) reported evidence of white spots and cavitated carious lesions in the past 12 months and visible plaque accumulation in more than half of American Indigenous children (mean age = 27.81 months) [42].

Discussion

The primary aim of this review was to identify the utilisation of CRA tools in underserved populations. A total of 26 studies were retrieved worldwide including North and South America, Europe, Asia, and the Middle East. From the extracted studies, nine distinct CRA tools were identified. Nearly all studies revealed significant portions of underserved subgroups to be at moderate- to high-risk of experiencing caries, with a few exceptions justified by methodological considerations and specific CRA tool utilised.

Custodio-Lumsden et al. (2016), found that majority of the low-income children were classified as having low-risk, which may be due to the inclusion criteria of children who presented for routine dental examinations only while excluding those who visited for emergency purposes [44]. Halasa-Rappel et al. (2019), highlighted a lack of consistency in weighing risk factors across the different CRA tools; for instance, low-income subjects were mostly classified as low-risk with BCH-CRA and high-risk with other tools [23]. Lastly, Jurasic et al. (2021), discovered more than half of the Veterans were classified as low-risk and justified it may be due to ADA-CRA’s poor sensitivity and better specificity [45]. Remaining studies highlighted elevated caries risk and provided evidence for a greater burden of caries in underserved populations.

The second aim of this review was to highlight key findings indicating the oral health status of our target population. Our findings raised a critical public health concern, as the results indicated high prevalence rates of caries and clinical indicators. Caries was highly prevalent in low-SES, low-income and rural children, ranging from 39.6 to 79% [33, 43, 44, 46, 49, 58]. The same phenomenon was observed in other studies, where underserved children experienced elevated prevalence of caries [13, 64–66]. Clinical indicators commonly observed across the subgroups included obvious signs of decay including white spot lesions, cavitated lesions, visible plaque accumulation, presence of restorations, and gingivitis. In addition, most studies showed high caries experience scores (i.e., DMFT/dmft) in their respective underserved participants. Overall, the findings of this review can be due to a complex interaction of social, economic, and systemic factors that affect access to care for underserved subgroups and their oral health status as a result [5–9, 68]. The identified oral health challenges in these underserved communities call for the attention of policymakers and practitioners to implement new strategies for improving oral health equity. One effective approach involves incorporating validated CRA assessment into broader public health settings, which can further identify high-risk population groups and appropriately reallocate resources among low-, moderate-, and high-risk groups with more intense support among the former, thereby improving the overall efficiency of the healthcare system [23]. Furthermore, assessment criteria such as diet, plaque accumulation and fluoride use can be addressed through non-surgical interventions that are feasible in both clinical and community settings, offering policymakers and practitioners insight into possible preventive strategies for promoting equitable oral health.

Identified CRA tools shared some level of similarity, as all aimed to assess various caries-related factors in their study participants. Despite the similarities, there is considerable heterogeneity among the tools. Differences existed in the specific components assessed, scoring systems, predictive validity, and methodology. For instance, some tools collected data through technology, while others used traditional interviews and/or questionnaires. Cariogram provided a mean CAC which required further interpretation to finalise the caries risk level, whereas other tools immediately categorised assessment results as low, moderate, or high. Scoring scales differed significantly primarily due to the underlying differences in components that were assessed in each tool. The existing heterogeneity within the current literature shows the need for standardised guidelines on the core components of CRA tools to enhance their reliability and comparability across the studies.

There was also a lack of sociocultural factors in the assessment criteria of the included CRA tools, considering that demographic characteristics such as education level and cultural background can influence access to care and caries risk profiles [5–9]. As shown in Additional file 1: Table S1, most CRA tools focus on assessing biological risk factors, disease indicators, and protective factors, failing to incorporate sociocultural factors in their assessment criteria. Furthermore, most of the identified CRA tools were lacking formal evaluation and validation, apart from Cariogram, CAMBRA, and CDA CRA [9, 35, 69]. The lack of validation raises methodological flaws in many CRA tools, possibly causing overestimation or underestimation of risk levels that can lead to inadequate allocation of resources and implementation of preventive measures [35]. Thus, it is suggested that more CRA tools be properly assessed for their validity, followed by impact studies in randomised controlled trials or machine learning algorithms [35, 70]. In addition, the fundamental properties of CRA tools such as ‘discrimination’ (differentiation between individuals of high-risk and low-risk) and ‘calibration’ (goodness of fit) should be assessed in parallel with internal and external validation–ensuring its utility and allowing dental practitioners to make evidence-based decisions for their patients [35].

The findings of our study can be utilised to enhance dental education programs. Incorporating CRA tools into dental education, particularly for underserved populations, enhances future practitioners’ ability to address oral health disparities through prevention-focused care [67]. The lack of adequate research on the use of CRA tools in underserved populations and poor oral health outcomes in this demographic, as concluded by our findings can help promote awareness of social determinants of health, foster cultural sensitivity, and encourage community-oriented approaches in addressing this gap [70, 71]. Additionally, efforts should be made to incorporate CRA in dental curriculum and interprofessional collaboration in resource-limited settings, to improve access to dental care in marginalised communities [72].

Overall, CRA tools particularly can be valuable in addressing oral health inequities in underserved populations [23]. These tools allow dental care providers to systematically identify individuals at higher risk for dental caries by evaluating factors such as SES, limited access to care, diet, and oral hygiene behaviours. In underserved communities, where barriers to accessing routine dental services are prevalent, CRA tools enable the prioritization of preventive measures, such as fluoride treatments and patient education, tailored to specific risk profiles [23]. By addressing the unique risk factors present in these populations, CRA tools can help clinicians deliver targeted, cost-effective interventions, ultimately reducing the burden of untreated caries and improving oral health outcomes [23].

Limitations

There are some limitations involved with this review. First, the results of this review may be highly heterogeneous, possibly due to large variations in sample sizes, methodological differences, and missing information from the included studies. Some studies included minor modifications in their assessment criteria tailored towards their needs, resulting in heterogenous results across authors who utilised identical CRA tools. In addition, the inclusion of both children and adults from any racial background may account for a lack of consistency in assessment results. Second, there was a lack of available evidence in adults and in African regions due to the gap in the literature. This resulted in an imbalance, with fewer studies examining caries risk in adults compared to children, and none from African regions. Third, our review was restricted to English-language papers published within the last 20 years, which may have limited the inclusion of studies from more diverse settings and timeframe. Lastly, several studies utilised convenience sampling from hospitals and dental clinics, which raises concerns with the restriction of underserved populations in our inclusion criteria. The authors recognise that there remains a wide body of underserved individuals who are not identified in this review and the current literature, as many are unable to access care completely.

Despite these limitations, this review is the first to provide a comprehensive scope of the utilisation of CRA tools and oral health status of underserved populations, including both children and adults. Among the existing studies that focused on examining caries risk in children of general populations, the authors provide insight into the oral health status of underserved populations of all ages globally.

Conclusion

To conclude, this review revealed that underserved populations experience elevated caries risk and significant unmet oral health needs. However, due to the gap in literature regarding caries risk in adult underserved populations and African regions, further representative data are required to fully understand the utilisation of CRA tools from a global and comprehensive perspective. Our findings of underserved populations may be limited due to convenient sampling from hospitals and dental clinics by some authors. Therefore, the authors suggest future research should aim to identify additional underserved subgroups in natural settings to minimise bias and better understand which communities are at risk due to limited access to care. In addition, the findings of this review highlight high levels of heterogeneity across the available CRA tools. Thus, there is a need for the development of a standardised and appropriately validated CRA tool with more integration of sociocultural factors/indicators which can be used in underserved as well as in general populations. Most importantly, potential strategies to increase accessibility and provide personalised care to underserved populations should be explored to mitigate the oral health disparities. The authors encourage policymakers and practitioners to implement diverse perspectives in broader public health policies to provide equitable care to patients from all backgrounds.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

AAP-OHRAT

American Academy of Pediatrics’ Oral Health Risk Assessment Tool

AAPD-CAT

American Academy of Pediatric Dentistry Caries Assessment Tool

ADA-CRA

American Dental Association Caries Risk Assessment

BCH-CRA

Boston Children’s Hospital Caries Risk Assessment

CAC

Chance of avoiding caries

CDA-CRA

California Dental Association Caries Risk Assessment

CRA

Caries Risk Assessment

DMFT

Decayed Missing Filled Teeth

DMFS

Decayed Missing Filled Surfaces

ECC

Early childhood caries

ID

Intellectual disabilities

JBI

Joanna Briggs Institute

MSB

MySmileBuddy

SCSC

Special Care School Children

S-ECC

Severe early childhood caries

SES

Socioeconomic status

Author contributions

M.P. and A.J. conceived the ideas of the scoping review and the methodological design; M.P. and F.P. contributed to data collection screening; M.P. and A.J. performed the data extraction, analysis, and interpretation of results; M.P., F.P., S.T., J.S., A.A., and A.J. led the original draft preparation and critically reviewed the manuscript. All authors read and approved the final manuscript.

Funding

This study was not supported by any sponsor or funder.

Data availability

This scoping review is based on published literature. All data generated are available in the main text or supplementary material.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.