What do systematic reviews say?

Caries assessment at enamel threshold vs None (descriptive prevalence review; subgroup analyses by urban/rural status, country and year)

Enamel-threshold caries prevalence in subgroup 'Region: non-Scandinavian' was 0.74 (95% CI 0.53 to 0.96; 11 studies).

• Dental caries at enamel and dentine level among European adolescents - a systematic review and meta-analysis. 2022 All outcomes

  Europe (community, subnational and national dental surveys) · Region: non-Scandinavian
  Prevalence 0.74 [0.53, 0.96]
  11 studies
  GRADE not reported
• Dental caries at enamel and dentine level among European adolescents - a systematic review and meta-analysis. 2022 All outcomes

  Europe (community, subnational and national dental surveys) · Age group: 12–15 years
  Prevalence 0.73 [0.59, 0.87]
  24 studies
  GRADE not reported
• Dental caries at enamel and dentine level among European adolescents - a systematic review and meta-analysis. 2022 All outcomes

  Europe (community, subnational and national dental surveys) · Examination type: partial-mouth
  Prevalence 0.81 [0.70, 0.91]
  5 studies
  GRADE not reported

Machine-learning algorithms for dental caries detection vs Different machine-learning algorithms for caries prediction

The review reported 92.37% accuracy for automated caries detection in Duong et al. (Ungraded).

• Uses of Different Machine Learning Algorithms for Diagnosis of Dental Caries. 2022 All outcomes

  Taiwan · Duong et al., 2021; support vector machine
  Accuracy 92.37
  1 study · 620 participants
  GRADE not reported

Combined antiretroviral therapy (post-cART) vs Pre-cART (same children, before therapy)

Oral candidiasis prevalence fell significantly after cART compared with before cART in the same children (very low certainty).

• Impact of Antiretroviral Therapy on Oral Health among Children Living with HIV: A Systematic Review and Meta-Analysis. 2022 All outcomes

  After vs before cART (within-person) · Hospitals and AIDS outpatient centres in India, Brazil, and several Asian countries · post-cART vs pre-cART
  OR 0.19 [0.15, 0.25]
  Studies not reported
  Very low

P-hydroxybenzoic acid vs placebo

P-hydroxybenzoic acid showed no clear difference in presence of dry socket compared with placebo; GRADE not reported.

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  RR 0.08 [0.00, 1.37]
  1 study · 45 participants
  GRADE not reported

Heal-all tissue patch vs no treatment

Heal-all tissue patch showed no clear difference in presence of dry socket compared with no treatment; GRADE not reported.

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  RR 0.08 [0.00, 1.32]
  1 study · 80 participants
  GRADE not reported

Caregiver help with child tooth brushing vs Not applicable (single-group prevalence)

About 54% of children with caries in LMICs received caregiver help with tooth brushing (pooled prevalence 0.541, 95% CI 0.266 to 0.803; 3 studies; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia)
  Prevalence 0.5408 [0.2663, 0.8025]
  3 studies · 830 participants
  GRADE not reported

Gaseous ozone vs placebo or control

Gaseous ozone showed no clear difference in presence of dry socket compared with placebo or control; GRADE not reported.

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  RR 0.20 [0.02, 1.61]
  1 study · 60 participants
  GRADE not reported

Female sex vs Not applicable (single-group prevalence)

Among children with caries in LMICs, about half were female (pooled prevalence 0.497, 95% CI 0.475 to 0.519; 6 studies; I2=0%; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia)
  Prevalence 0.4968 [0.4749, 0.5186]
  6 studies · 1999 participants
  GRADE not reported

Sugary drinks/sweets consumption vs Not applicable (single-group prevalence)

About 70% of children with caries in LMICs consumed sugary drinks/sweets (pooled prevalence 0.697, 95% CI 0.478 to 0.877; 7 studies; very high heterogeneity; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia)
  Prevalence 0.6974 [0.4784, 0.8773]
  7 studies
  GRADE not reported

Machine-learning algorithms for dental caries detection vs Different machine-learning algorithms for caries prediction

The review reported 88% accuracy for premolar caries in Lee et al. (Ungraded).

• Uses of Different Machine Learning Algorithms for Diagnosis of Dental Caries. 2022 All outcomes

  Korea · Lee et al., 2018; CNN
  Accuracy 88
  1 study · 3000 participants
  GRADE not reported

Povidone iodine 1% vs control

Povidone iodine 1% led to lower presence of dry socket compared with control; GRADE not reported.

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  RR 0.36 [0.14, 0.98]
  1 study · 189 participants
  GRADE not reported

Recombinant bovine fibroblast growth factor gel vs control

Recombinant bovine fibroblast growth factor gel led to lower presence of dry socket compared with control; GRADE not reported.

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  RR 0.10 [0.01, 0.76]
  1 study · 160 participants
  GRADE not reported

Machine-learning algorithms for dental caries detection vs Different machine-learning algorithms for caries prediction

The review reported 82% accuracy for combined molar and premolar caries in Lee et al. (Ungraded).

• Uses of Different Machine Learning Algorithms for Diagnosis of Dental Caries. 2022 All outcomes

  Korea · Lee et al., 2018; CNN
  Accuracy 82
  1 study · 3000 participants
  GRADE not reported

Platelet-rich fibrin (PRF) graft vs control

Platelet rich fibrin showed no clear difference in presence of dry socket compared with control; GRADE not reported.

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  RR 0.44 [0.15, 1.35]
  1 study · 100 participants
  GRADE not reported

Iodine tampon vs Saline irrigation (monoject syringe)

• Local interventions for the management of alveolar osteitis (dry socket) 2022 · Cochrane All outcomes

  Follow-up: median 7 days
  Iodine tampon versus rinsing with saline in a syringe (Tuk 2019): Iodine tampon 0/54 Monoject with saline 0/54; no events (effect not estimable)
  1 study · 108 participants
  GRADE not reported

Socioeconomic status category vs Not applicable (single-group prevalence)

About 36% of children with caries in LMICs were of low SES (pooled prevalence 0.359, 95% CI 0.167 to 0.578; 4 studies; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia) · Low SES
  Prevalence 0.359 [0.1673, 0.5779]
  4 studies
  GRADE not reported

Machine-learning algorithms for dental caries detection vs Different machine-learning algorithms for caries prediction

The review reported 97.1% accuracy for root caries prediction in Hung et al. (Ungraded).

• Uses of Different Machine Learning Algorithms for Diagnosis of Dental Caries. 2022 All outcomes

  Jordan · Hung et al., 2019; support vector machine
  Accuracy 97.1
  1 study · 5135 participants
  GRADE not reported

Dental caries assessment (WHO/clinical diagnostic criteria; caries = DMFT>0) vs None (descriptive prevalence; random-effects pooled proportion)

About 39% of children in LMICs had dental caries (pooled prevalence 0.387, 95% CI 0.284 to 0.495; 11 studies; very high heterogeneity I2=99%; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia)
  Prevalence 0.387 [0.284, 0.495]
  11 studies · 11115 participants
  GRADE not reported

Maternal education attainment vs Not applicable (single-group prevalence)

About 10% of children with caries in LMICs had mothers with no formal education (pooled prevalence 0.101, 95% CI 0.033 to 0.201; 3 studies; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia) · No formal education
  Prevalence 0.1014 [0.0334, 0.2007]
  3 studies · 1499 participants
  GRADE not reported

Maternal education attainment vs Not applicable (single-group prevalence)

About 42% of children with caries in LMICs had mothers with primary education (pooled prevalence 0.425, 95% CI 0.280 to 0.576; 7 studies; not GRADE-rated).

• Individual, Family, and Socioeconomic Contributors to Dental Caries in Children from Low- and Middle-Income Countries. 2022 All outcomes

  Low- and middle-income countries (Brazil, Nigeria, Cambodia, Tanzania, Mexico, Ethiopia) · Primary education
  Prevalence 0.4246 [0.2799, 0.5763]
  7 studies · 3240 participants
  GRADE not reported