Community Dent Oral Epidemiol 2015; 43; 511–520 All rights reserved
Ó 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Minimum intervention dentistry approach to managing early childhood caries: a randomized control trial
Peter Arrow and Elizabeth Klobas Dental Health Services, Bentley Delivery Centre, Perth, WA, Australia
Arrow P, Klobas E. Minimum intervention dentistry approach to managing early childhood caries: a randomized control trial. Community Dent Oral Epidemiol 2015; 43: 511–520. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Abstract – Objectives: A pragmatic randomized control trial was undertaken to compare the minimum intervention dentistry (MID) approach, based on the atraumatic restorative treatment procedures (MID–ART: Test), against the standard care approach (Control) to treat early childhood caries in a primary care setting. Methods: Consenting parent/child dyads were allocated to the Test or Control group using stratified block randomization. Inclusion and exclusion criteria were applied. Participants were examined at baseline and at follow-up by two calibrated examiners blind to group allocation status (j = 0.77), and parents completed a questionnaire at baseline and follow-up. Dental therapists trained in MID–ART provided treatment to the Test group and dentists treated the Control group using standard approaches. The primary outcome of interest was the number of children who were referred for specialist pediatric care. Secondary outcomes were the number of teeth treated, changes in child oral health-related quality of life and dental anxiety and parental perceptions of care received. Data were analyzed on an intention to treat basis; risk ratio for referral for specialist care, test of proportions, Wilcoxon rank test and logistic regression were used. Results: Three hundred and seventy parents/carers were initially screened; 273 children were examined at baseline and 254 were randomized (Test = 127; Control = 127): mean age = 3.8 years, SD 0.90; 59% male, mean dmft = 4.9, SD 4.0. There was no statistically significant difference in age, sex, baseline caries experience or child oral health-related quality of life between the Test and Control group. At follow-up (mean interval 11.4 months, SD 3.1 months), 220 children were examined: Test = 115, Control = 105. Casenotes review of 231 children showed Test = 6 (5%) and Control = 53 (49%) were referred for specialist care, P < 0.0001. More teeth were filled in the Test group (mean = 2.93, SD 2.48) than in the Control group (mean = 1.54, SD 2.20), Wilcoxon’s test, P < 0.0001. Logistic regression, after controlling for age and baseline caries experience, showed a higher risk of referral by allocation to control group, OR 32.6, 95% CI 10.8–98.4, P < 0.0001. Conclusion: The MID–ART approach reduced significantly the likelihood of referral for specialist care, and more children and teeth were provided with treatment.
The minimum intervention dentistry approach to manage dental caries and its potential role in the provision of public dental services have been canvassed in the dental literature (1). The Atraumatic doi: 10.1111/cdoe.12176
Key words: atraumatic restorative treatment; clinical trials; early childhood caries Peter Arrow, Dental Health Services, Bentley Delivery Centre, Locked Bag 15, Perth, WA 6983, Australia Tel.: +61 8 93130600 Fax: +61 8 93134141 e-mail: [email protected], peter. [email protected]
Trials register (ANZCTR: ACTRN12612000474853). Submitted 5 January 2015; accepted 20 May 2015
Restorative Treatment (ART) approach, initially developed to assist dental care delivery in underserved communities, where access to electricity and running water may not be readily available, is
511
Arrow & Klobas
now increasingly seen to have relevant applications in subpopulation groups in developing as well as developed countries (2). The ART approach principally relies on removing caries using hand instruments alone without the administration of a local anesthetic and restoration of the prepared cavity with glass–ionomer cement. Since its introduction in the 1990s, the approach has been examined and tested in different populations and subgroups, such as children, the elderly and people with disabilities, and has been shown to be clinically successful as well as being acceptable to patients (3). Management of early childhood caries (ECC) is demanding of clinician time and skills. In some instances, the treatment of ECC is delivered in a hospital setting under general anesthesia (GA), which is costly and not without risk. Admissions to a hospital for dental care are classified as potentially preventable with timely and adequate nonhospital care (4). However, there is a trend of increasing hospital admissions for dental care among children, especially among 0- to 4-year-olds (5), and in Australia, this is occurring in spite of the apparent low dental caries experience among children (6, 7). Western Australia had the highest, and increasing, rate of hospital admission for dental treatment among all the States and Territories in Australia, and the cost for hospital admission for dental care for children was approximately A $9 million per year (5, 8). The ART approach to restorative care among children has been reported to be well accepted (3), and children reported experiencing less dental anxiety and pain compared with standard care approaches (9, 10). The ART approach has also been shown to be acceptable for preschool children (11). While there have been studies reporting improved changes in child oral health-related quality of life (COHRQoL) after oral rehabilitation under GA (12, 13), there are few reports of the impact on COHRQoL after dental treatment in primary care settings (14). Recent reports also suggest that treatment of caries in children under GA may provoke higher levels of dental anxiety (15), and the approach is not without risks (16). Different approaches to managing caries in the primary dentition including ECC have been, and continue to be, tested (17, 18). There are also reports of variations between general dental practitioners and specialist pediatric dentists in managing primary tooth caries in young children, with general practitioners preferring a less aggressive
512
approach using atraumatic techniques, while specialists preferring a more comprehensive rehabilitation (19, 20). Care for children under GA tends to be more aggressive (21) than in other settings, in particular to avoid repeat GA episodes (22). But, limited research evidence is available on the potential for treatment approaches to reduce the need for specialist care and GA treatment. In Western Australia, public dental services are provided through Dental Health Services of Western Australia (DHS). The School Dental Service (SDS) of the DHS, through employing dental therapists, has been the mainstay of a successful, publicly provided dental care for 5- to 17-year-old schoolchildren since the early 1970s (23, 24). Dental therapists were trained for 2 years (it is now a 3-year training program) and provide basic primary care that includes oral examination and diagnosis, fissure sealing, application of fluoride varnish, scale and clean, oral health counseling, restoration of primary and secondary teeth using plastic materials and administration of a local anesthetic, and extractions of primary teeth. The general dental clinics of the DHS provide dental care to adults and preschool children (0- to 4-year-old) eligible for subsidized care (liable for copayments), with restorative care principally being provided by dentists. Some government general dental clinics do utilize dental therapists, principally to provide dental hygienist services. Standard treatment for caries in children requiring restorative care within the DHS would involve the use of a local anesthetic and rotary instruments. The aim of this study was to test the hypothesis that provision of dental care to children with ECC by dental therapists using a hybrid ART approach would be at least as successful as that of managing ECC by the standard care protocols prevailing within DHS. Successful management of care was defined as management of the necessary dental treatment of the child without referral for specialist pedodontic care.
Materials and methods Ethical approval for the study was provided by the University of Western Australia Human Research Ethics Committee. Participating parents provided signed informed consent. The study was a pragmatic, parallel group, open label, randomized control trial in a primary healthcare setting of a public health dental service in
Minimum intervention dentistry early childhood caries
Perth, Western Australia. Participants were recruited over a 12-month period through invitation poster displays of young children’s teeth at various stages of ECC (precavitated enamel lesions to frank cavitation) at preschool centers, child health clinics and school dental clinics of the DHS with an offer of free primary dental care for their child (not specialist care). Inclusion and exclusion criteria were applied. Children younger than 6 years of age with ECC were eligible. Exclusions were children with acute dental pain and/or infections; physical/mental disability which limited dental treatment in the primary care setting; general or dental developmental conditions, which required specialist care (such as, amelogenesis imperfecta, cleft lip/palate); and children who were caries free. Initial screening was undertaken over the phone and a secondary screening was undertaken at a baseline examination (prior to group allocation) conducted at SDS clinics. Parents completed a baseline questionnaire after the initial screening, but before the baseline examination. The questionnaire sought information about the COHRQoL using the Early Childhood Oral Health Impact Scale (ECOHIS) and collected sociodemographic characteristics (25). The ECOHIS score ranges 0–52. The ECOHIS has been shown to have acceptable reliability and validity among Australian preschool children (26). Parents also completed a dental fear and anxiety schedule (27), and children older than 3 years of age completed a faces child dental fear scale, range 1–5 (dichotomized into afraid, score 1 or 2; or not afraid, score 3, 4, or 5; score 1 was classified as severe fear) at the baseline examination appointment (28). The ECOHIS and faces child dental fear were completed again at follow-up. Socioeconomic level was determined through parent response to a question on eligibility for government concession card. In Western Australia, concession card holders (CCHs) can access a range of services, such as dental and medical services at government agencies at subsidized cost. Children were examined for follow-up oral health status at SDS clinics (6–12 months after group allocation) by the same examiners who conducted the baseline examination and who were blind to group allocation status. The 6- to 12-month follow-up interval was deemed long enough to be able to observe the outcomes. The primary outcome of interest was the number of children referred for specialist pediatric care. The information was abstracted from the clinical records of the
children where a letter of referral for specialist pedodontic care or where clinical notes indicate that a verbal discussion was held with parents recommending specialist consultation for care was taken as being referred for specialist care. Secondary outcome measures were changes in caries experience using the dmft/dmfs index (decayed, missing, and filled primary teeth/tooth surfaces), COHRQoL, and child dental fear over the follow-up interval and between the groups. Teeth treated with restorations were evaluated using the criteria established for evaluating ART restorations (29, 30). Parental perspectives on care delivered through the two arms of the trial were elicited through focus group interviews with a sample of parents from Test and Control groups. Parents were also asked a global question on how their child’s oral health changed through the participation in the study on a five-point scale: 1 = worsened a lot, 2 = worsened a little, 3 = stayed the same, 4 = improved a little, and 5 = improved a lot. The response was dichotomized into improved (score 4 or 5) or no change/worsened (score 1, 2 or 3). Sample size estimation was based on demonstrating equivalence of effect between the two arms of the randomized trial (31). The primary outcome measure was the number (proportion) of children referred for specialist care for ECC. Little information was available on the number of children referred through the standard care approach within DHS, and it was assumed that 10% of children were referred through standard care, and the outcome was deemed equally effective if care provided using the MID–ART approach was at most 10% less effective than standard care, that is, at most 20% referred. The findings will be deemed statistically significant at a = 0.05 with an 80% power to detect a true difference (b = 0.2). A sample size of 145 in each arm of the trial was required. CONSORT participant flow chart is shown in Fig. 1. Consenting parent/child dyads were stratified on caries experience (≤5 dmfs and >5 dmfs) after baseline examination. A central study coordinator allocated the children to Test or Control group using a computer-generated block randomization procedure, using blocks of two, four, and six. Two calibrated dentists undertook the baseline (and follow-up) examinations (interexaminer j = 0.77) at SDS clinics, using established criteria to classify ECC (32) and the presence of enamel defects using the Developmental Defects of Enamel index (DDE) (33). Repeat examinations were not
513
Arrow & Klobas
caries and without the use of a local anesthetic. However, use of rotary instruments was permitted based on the dental therapist’s assessment of the need for, and the child’s capacity to cope with, the procedure. The approach was underpinned by minimum intervention dentistry whereby caries removal was undertaken to the extent of achieving caries-free status at the dentino-enamel junction while leaving softened caries on the pulpal floor if its removal was likely to expose the dental pulp. The prepared cavity was restored with glass–ionomer cement, any remaining at-risk pits and fissures sealed and fluoride varnish applied to all approximal surfaces, and noncavitated enamel lesions. Children were then recalled at 3- to 6-month intervals, at which time repairs to restorations and fluoride varnish application and further counseling on preventive measures were undertaken. Dental radiographs were not routinely taken. Control group children were provided with care by clinicians working in government general dental clinics (with capacity for 8–10 clinicians) in the Perth metropolitan area, and the coordinating senior dentist allocated the child to a clinician within the clinic for dental care (dentists or dental therapists). Dentists principally provided the restorative care in government general dental clinics, and dental therapists, where employed, would provide hygienist services. Clinicians providing
undertaken due to the age of the study participants and logistical difficulties of arranging a second appointment for families with young children. After group allocation, the study coordinator contacted the Test clinicians or Control general dental clinics and provided participant details to enable scheduling of appointments for follow-up care.
Treatment procedures The Test group children were provided with care by school dental therapists (n = 5), who volunteered for the study and were provided with additional training on the MID–ART approach to manage dental caries. Training was undertaken over a half-day session, a PowerPoint presentation covering the ART approach to restorative care and age appropriate behavioral techniques based on the manual by developers of the ART approach (34). The dental therapists then also undertook a restorative procedure on a primary tooth of a preprimary child, scheduled for routine restorative care, using the ART approach. Treatments for the Test children were provided in SDS clinics (with capacity for two clinicians and situated on primary school grounds) by dental therapists. The approach was a hybrid ART in that hand instruments principally were used to remove
Telephone screen = 370 Excluded = 84 Completed Baseline questionnaire = 286 Failed to attend baseline examination = 13 Baseline examination = 273 Excluded = 19
Randomised = 254
dmfs≤5 = 54 dmfs>5 = 73
Control = 127
Test = 127
At least 1 visit for care = 122
Follow-up = 220
Lost to follow-up = 12
Test Follow-up = 115
514
dmfs≤5 = 56 dmfs>5 = 71
At least 1 visit for care = 111 Lost to follow-up = 22
Control Follow-up = 105
Fig. 1. CONSORT participant flowchart.
Minimum intervention dentistry early childhood caries
care to Control children were advised to provide care in the usual manner using the usual standard care approach for a child of this age, and for the children to be reviewed as per the DHS protocol. No other instructions on who should provide the care or how the care was to be delivered were provided. Standard treatment approach usually involves the use of a local anesthetic, rotary instruments and restoration of the prepared cavity with adhesive restorations (glass–ionomer, or resin cements). Patient management strategies used within the primary care setting of the DHS were principally behavioral management techniques. The use of other approaches such as inhalation sedation was not available, and pharmacological approaches were at the discretion and skill level of the treating clinician (for dentists only). Both Test and Control clinicians were advised that they were to provide care to the children to the best of their ability. The outcomes of interest were not communicated to any of the clinicians. All clinicians were further advised that referral for specialist care was an available option if the treating clinician felt that the required care was best managed through referral to a specialist. Referral option for a dental therapist in the first instance was to refer to a supervising dentist, who in turn determined whether referral to a specialist was required, while dentists can refer directly for specialist care. Data were analyzed on an intention to treat basis. Point estimates of effect (cumulative incidence) and the associated 95% confidence intervals for specialist referral were calculated, and the number needed to treat (1/absolute risk difference) to avoid one additional specialist referral was calculated (35). Tests of proportions for categorical variables between two groups (chi-square) and test of means using parametric (t-test or paired t-test) and nonparametric (Kruskal–Wallis test or Wilcoxon signed-rank test) statistical procedures were used. Multivariate analysis using logistic regression was used to control for any potentially confounding factors that were not balanced at baseline or at follow-up. Statistical analyses were undertaken on a personal computer using STATA 12 (36).
Results The distribution of baseline characteristics of the study participants overall and within Test (T) and Control (C) groups suggests that randomization
was fair, Table 1. There were no statistically significant differences between the T and C groups on any of the factors as indicated by overlapping 95% confidence intervals. The age range of study participants was (Test, 1.0–5.5 years; Control, 1.1– 5.2 years). Initially, there were five clinicians (dental therapists) providing care to T children; however, during the study, one clinician was relocated and one clinician commenced maternity leave. The coordinating senior clinician at the metropolitan government general dental clinic (ten metropolitan clinics) had the discretion, based on their judgment, to allocate the C participant child to a clinician at that clinic. In all but one clinic, dentists provided the care to C children (one dental therapist provided care to four children at one government general dental clinic). As shown in Fig. 1, after a mean interval of 11.4 months (range 5.8–22.7 months; 95% CI 10.9, 11.8), there were 220 children examined at followup. Reasons for lost to follow-up were moving away from the area (n = 4) and noncontactable (n = 27) or failed to attend the requested examination appointments (n = 3). The baseline characteristics of those followed up and lost to follow up are shown in Table 2. Those lost to follow up were significantly younger than those followed up, Wilcoxon rank-sum, P < 0.001. CCHs and those who were referred for specialist care were also less likely to have attended for a follow-up examination. Although more Test group participants were followed up, the difference was not statistically significant, v2, P = 0.065, Table 3. Also, those lost to follow up among the C group were significantly younger than those followed up, and among the C group, those referred for care were less likely to have had a follow-up examination, Fisher’s exact test, P < 0.05, while among the Test group the difference was of borderline significance, Fisher’s exact test, P = 0.049. Table 4 shows the primary outcome evaluated in this study. There were more children from the C group that were referred for specialist pedodontic care than in the T group (all children requested for referral to a specialist by the dental therapists in the Test group were referred for specialist care by the supervising dentist), v2, P < 0.0001. The risk ratio for children being referred for specialist care if allocated to the C group was 10.1, 95% CI 4.5, 22.5, and the risk difference was 0.44, 95% CI 0.34, 0.54. The number needed to treat for one additional child to avoid specialist referral using the test
515
Arrow & Klobas Table 1. Baseline characteristics of study participants overall and within groups Factor
Total
Test n = 127 (% or mean and 95% CI)
Control n = 127 (% or mean and 95% CI)
Sex Age (years) dmft dmfs Parent fear Child fear % Afraid ECOHIS CCH
Male = 59% (53, 65) 3.8 (3.7, 3.9) 4.9 (4.4, 5.4) 11.0 (9.5, 12.4) 16.7 (15.5, 17.8) 11.4% (6.9, 15.9) 11.1 (10.0, 12.2) Yes = 41% (36, 47) No = 59% (53, 64)
Male = 55% (46, 63) 3.9 (3.8, 4.0) 4.9 (4.2, 5.5) 10.8 (8.7, 12.9) 17.3 (15.6, 19.1) 12.5% (6.1, 18.9) 10.1 (8.7, 11.4) Yes = 37% (29, 45) No = 63% (55, 71)
Male = 63% (55, 71) 3.7 (3.5, 3.9) 5.0 (4.3, 5.7) 11.2 (9.1, 13.2) 16.0 (14.5, 17.5) 10.1% (3.8, 16.5) 12.1 (10.5, 13.7) Yes = 45% (36, 54) No = 55% (46, 64)
dmft = mean count of decayed, missing filled primary teeth; dmfs = mean count of decayed, missing and filled primary tooth surfaces; ECOHIS = mean Early Childhood Oral Health Impact Scale; CCH = concession card holder.
Table 2. Baseline characteristics of those followed up and those lost to follow up Factor Sex Male Female Age (years) dmft dmfs Parent fear Child fear % afraid ECOHIS CCH† Referred‡
Followed up n = 220 (% or mean and 95% CI)
Lost to follow up n = 34 (% or mean and 95% CI)
N = 130 (59.1; 52.5, 65.6) N = 90 (40.9; 34.4, 47.5) 3.9 (3.8, 4.0)* 4.9 (4.4, 5.5) 11.0 (9.4, 12.7) 16.8 (15.5, 18.0) 11.1 (6.4, 15.9) 11.3 (10.1, 12.4) Yes = 82 (37.3) No = 138 (62.7) Yes = 45 (21.5) No = 164 (78.5)
N = 20 (58.8; 42.0, 75.7) N = 14 (41.2; 24.3, 58.0) 3.2 (2.8, 3.6)* 4.8 (3.7, 6.0) 10.5 (7.1, 13.9) 15.9 (13.0, 18.8) 13.6 ( 1.1, 28.4) 10.0 (7.1, 13.0) Yes = 22 (66.7) No = 11 (33.3) Yes = 14 (63.6) No = 8 (36.4)
dmft = mean count of decayed, missing filled primary teeth; dmfs = mean count of decayed, missing and filled primary tooth surfaces; ECOHIS = mean Early Childhood Oral Health Impact Scale; CCH = concession card holder. *Wilcoxon rank-sum test P < 0.001; †CCH status for one participant missing, and Pearson’s chi-square, P < 0.05; ‡ Pearson’s chi-square, P < 0.001.
intervention was 3 (95% CI 2–3). For 23 children, information regarding referral status was not available either because the child did not attend any of the allocated appointments (n = 20; T = 4, C = 16) or had attended for an examination but then subsequently failed to attend any further appointments (n = 3: C = 3). One child in the Test group failed to attend for any treatment appointments, but was seen for the follow-up examination and provided information about the care received. A sensitivity analysis was undertaken to examine the extent of changes to the principal findings by assuming, at the limit, that all children with missing data in the Test group were referred for specialist care (four children) and all children with missing data in the control group were not referred for specialist care (19 children). The Pearson’s chi-square was P < 0.001, and the risk ratio was reduced from 10.1 to 5.3. Conversely, if the assumptions were reversed,
516
all Test children not referred and all control children referred, the risk ratio increased to 12.0, Pearson’s chi-square P < 0.001. The findings for caries experience between T and C are shown in Fig. 2. The T children had significantly fewer tooth surfaces and teeth that were decayed (T ds = 4.4, SD 9.3; C ds = 8.6, SD 11.9: T dt = 2.0, SD 3.0; C dt = 3.8, SD 4.2) and more tooth surfaces and teeth that had been filled (T fs = 7.1, SD 7.3; C fs = 3.6, SD 6.3: T ft = 2.9, SD 2.5; C ft = 1.5, SD 2.2) compared with children in the C group. There was also a trend of lower caries increment in the T group, although this was not statistically significant, Kruskal–Wallis test, P = 0.29. There was no change in the level of fear reported by the children in either group from baseline to follow-up, Wilcoxon signed-rank test, P = 0.87. However, at follow-up, the proportion of children reporting
Minimum intervention dentistry early childhood caries Table 3. Baseline characteristics of those followed up and lost to follow up among Test and Control participants Test group Followed up n = 115 (mean or %; and 95% CI)
Factor Sex (N) Male Female Age (years) dmft dmfs Parent fear Child fear % afraid ECOHIS CCH Referred
Control group Lost to follow up n = 12 (mean or % and 95% CI)
63 (55) 52 (45) 3.9 (3.8, 4.1) 5.0 (4.2, 5.7) 11.2 (8.9, 13.5) 17.7 (15.9, 19.5) 11.5 (5.0, 17.9) 10.3 (8.9, 11.8) Yes = 39 (33.9) No = 76 (66.1) Yes = 4 (3.5) No = 111 (96.5)
Followed up n = 105 (mean or % and 95% CI)
Lost to follow up n = 22 (mean or % and 95% CI)
67 (64) 38 (36) 3.8 (3.7, 4.0)* 4.9 (4.1, 5.7) 10.9 (8.5, 13.2) 15.8 (14.1, 17.5) 10.7 (3.6, 17.8)
13 (59) 9 (41) 3.1 (2.7, 3.5)* 5.4 (3.8, 7.0) 12.7 (7.8, 17.6) 17.1 (13.6, 20.6) 7.1 ( 7.7, 22.0)
7 (58) 5 (42) 3.4 (2.5, 4.3) 3.8 (2.2, 5.3) 6.6 (3.3, 9.8) 13.3 (7.5, 19.2) 25.0 (-9.0, 59.0) Yes No Yes No
7.6 (3.6, 11.6) = 8 (66.7) = 4 (33.3) = 2 (25.0) = 6 (75.0)
Yes No Yes No
12.3 (10.5, 14.1) = 43 (41.0) = 62 (59.0) = 41 (43.6)† = 53 (56.4)
11.4 (7.2, 15.5) Yes = 14 (66.7) No = 7 (33.3) Yes = 12 (85.7)† No = 2 (14.3)
dmft = mean count of decayed, missing filled primary teeth; dmfs = mean count of decayed, missing and filled primary tooth surfaces; ECOHIS = mean Early Childhood Oral Health Impact Score; CCH = concession card holder. *Wilcoxon rank-sum test P < 0.001; †Fisher’s exact test, P < 0.05.
Table 4. Children referred for specialist pedodontic care among groups
Referred Yes No Total
Test, n (%)a
Control, n (%)b
Total
6 (5) 117 (95) 123 (100)
53 (49) 55 (51) 108 (100)
59 (26) 172 (74) 231 (100)
v2 = 59.1, P < 0.0001. a One Test child who failed to attend for any treatment appointments presented for follow-up examination and provided information regarding referral status. b Three Control children failed to attend for treatment appointments and had no information regarding referral status in the case notes.
more severe dental fear was significantly higher in the C group, v2, P < 0.05. Parent reported COHRQoL was significantly better at follow-up compared with baseline for both groups, paired t-test, P < 0.05, but there were no statistically significant differences between the groups, P > 0.05. Two hundred and two parents responded to the global question of the change in their child’s oral health (T = 107, C = 95) of which 78 T (73%) and 40 C (42%) reported improved oral health, v2, P < 0.001. Multivariate analysis using logistic regression to model the risk of referral for specialist care found a significant effect of age and caries experience as well as group allocation. Older children were less likely to be referred, OR 0.5, P = 0.003, while an increase of one dmfs increased the odds of referral
by 10%, OR 1.1, P < 0.001. The highest odds of referral were for those children allocated to the C group, OR 32.6, P < 0.001.
Discussion The study’s hypothesis that the MID–ART approach was as effective as the standard care approach in managing ECC in a primary care setting, without the need for referral for specialist care, was rejected. The hybrid MID–ART intervention approach as undertaken reduced the rate of referral for specialist pedodontic care by 45% (Test 5%, Control 49%). In Australia, a child being referred to a specialist for management of dental caries is likely to be treated under GA, and WA also has the highest, and increasing, rate of hospital admission for dental treatment among all the States and Territories (5). A limitation of the study was that the study was a pragmatic trial, undertaken within a functioning public dental service delivering primary dental care using standard care procedures. The control was the standard care approach, which was delivered at government general dental clinics where dentists and dental therapists are employed, and dentists provided the majority of restorative care and dental therapists principally provided dental hygienist services. The pragmatic design allowed standard care to be delivered by clinicians within the general dental clinics; no restrictions were
517
Arrow & Klobas
* * *
*
Fig. 2. Comparison of caries experience outcome measures between Test and Control groups. ds = decayed primary tooth surface; ms = missing primary tooth surface; fs = filled primary tooth surface; dmfs = decayed, missing, filled primary tooth surfaces; dt = decayed primary tooth; mt = missing primary tooth due to caries; ft = filled primary tooth; dmft = decayed missing, filled primary teeth; dmfs incr = dmfs follow-up – dmfs baseline; dmft incr = dmft follow-up – dmft baseline. Error bars indicate 95% confidence limits; *95% CI do not overlap, significant at P < 0.05.
placed on the type of clinician providing the care, because the trial was testing the MID–ART approach rather than the type of clinician (dentist/ therapist) delivering the care. In one instance, four Control children were treated for ECC by a dental therapist. This was unlikely to have biased the study’s findings. Also, the strength of pragmatic trials was that interventions were tested within a real-world situation and were felt to be better at evaluating the effectiveness of an intervention (37). Little prior information was available on the likely referral patterns of the standard care clinicians. Information available on treatment decisionmaking between dentists and dental therapists within the DHS suggests that the approach to caries management was similar and it was unlikely that they will differ significantly when caring for preschool children under standard care conditions (38). It was possible and more likely that dentists who usually provide care to preschool children in public dental services were less likely to refer a child for specialist care because of greater experience in caring for children of such an age than dental therapists, who usually do not care for preschool children, and the sample size estimation was based on this premise. Participants were screened to exclude children with acute conditions requiring urgent care, and the study’s findings may not apply to those children. Participants were not randomly selected, but were those who responded to a poster display or recruited through child health nurses who may have noticed an oral health problem which required professional care. Assessment of baseline variables indicated that the stratified randomization process was fair. Although more children from
518
the T group were examined, there was no statistically significant difference between the T and C groups. Control group children that had a followup examination were significantly older and less likely to have been referred for specialist care. All children in the study were classified as having severe ECC (32). Referral status was associated with a child attending for a follow-up examination, fewer children who were referred for specialist care attended for a follow-up examination. Anecdotal information from the study trial coordinator was that the parents responded to the request for a follow-up attendance with comments such as they see little value in attending for a follow-up when their child was not provided with the needed care in the primary care setting and they are now confronted with costly specialist care. An observation, not reported in this manuscript, was that the majority of children who were referred for care had not followed through with the recommended care, principally due to cost. Sensitivity analysis to account for those lost to follow up suggests that the principal findings would not have altered significantly. Secondary outcome measures also indicated greater effectiveness of the MID–ART approach in the provision of primary care. More children in the T group were provided with restorative care (T fs = 7.1, SE 0.7; C fs = 3.6, SE 0.6) and had nearly half the level of untreated decay (T ds = 4.4, SE 0.9; C ds = 8.6, SE 1.2). There was also a lower level of caries increment among the T children. T children who attended for care were provided with comprehensive preventive care, at least one application of fluoride varnish, sealing of at-risk pits and fissures, and oral health counseling to prevent dental caries,
Minimum intervention dentistry early childhood caries
and this may explain the observed difference in caries increment. Additionally, more parents in the T group also perceived their child’s oral health had improved as a result of the care provided. Most of the studies that have examined the changes in COHRQoL have been undertaken in relation to oral rehabilitation under GA and have reported improvements after oral rehabilitation (12, 13). Delivery of dental care to children in a primary healthcare setting can improve their quality of life (14). This study found that dental care delivered in a primary care setting improved children’s oral health-related quality of life (both the T and C groups reported improved ECOHIS). A full report of the findings in relation to early childhood oral health-related quality of life is in a companion manuscript, which is under review. Although oral rehabilitation for ECC under GA has reported improvements in COHRQoL, oral rehabilitation under GA may increase child dental anxiety, which may be related to a more ‘aggressive’ approach undertaken, with more teeth extracted to avoid repeat GA episodes (15, 21, 22). The MID–ART approach on the other hand has been reported to provoke less stressful behavior in children undergoing restorative treatment (39). Although this study did not compare the MID– ART approach against the treatment under GA or that of specialist provided care, the finding of lower levels of dental fear in the T children suggests that the MID–ART approach may mitigate the development of severe dental fear. The MID–ART approach has been suggested as the most appropriate measure for managing ECC because of its ‘atraumatic’ approach (3). Other studies have reported lower levels of pain and anxiety experienced by children when treated using the ART approach compared with the standard approach. Although this study did not measure perceived pain by children, the overall level of anxiety and dental fear was assessed and indicated that children in the MID–ART group experienced lower levels of severe dental fear.
Conclusions The findings from the study suggest significant potential to reduce the rate of hospital admissions for dental treatment among preschool children by reducing the rate of referral for specialist care, which was strengthened by the sensitivity analyses. Delivery of care by dental therapists using the
MID–ART approach was well accepted by parents and children. Significantly improved COHRQoL was reported by the parents of both groups after treatment, but more parents of the T group rated their child’s oral health as improved. The T children reported lower levels of dental anxiety and had a higher level of dental caries treated. The delivery of care for ECC in a primary care setting appears to have the potential to significantly reduce the cost of care. The approach as tested is also likely to be cost-effective, given that a senior dental therapist’s salary is approximately 60% of that of a senior dentist within DHS.
Acknowledgements The authors would like to thank the study coordinators Rowena McInnes and Di Winston for their untiring efforts in participant recruitment and follow-up and the clinicians who participated in the study, especially the MID–ART clinicians; Deborah Arrow, Jodie Burton, Lesley Dwyer, Helen Forrest, and Chris Smith; and the DHS in supporting the study. The study was partly funded by the Targeted Research Fund of the WA Health Department. Lastly, the study would not have been possible without the willing participation of parents and children in this study. The authors declare no potential conflict of interests with respect to authorship or publication of this manuscript.
References 1. Calache H, Hopcraft MS, Martin JM. Minimum intervention dentistry – a new horizon in public oral health care. Aust Dent J 2013;58(1 Suppl):17–25. 2. Holmgren CJ, Roux D, Domejean S. Minimal intervention dentistry: part 5. Atraumatic restorative treatment (ART)–a minimum intervention and minimally invasive approach for the management of dental caries. Br Dent J 2013;214:11–8. 3. Frencken JE, Leal SC, Navarro MF. Twenty-five-year atraumatic restorative treatment (ART) approach: a comprehensive overview. Clin Oral Investig 2012;16:1337–46. 4. Katterl R, Anikeeva O, Butler C, Brown L, Smith B, Bywood P. Potentially avoidable hospitalisations in Australia: causes for hospitalisations and primary health care interventions. Adelaide: Primary Health Care Research & Information Service; 2012. 5. Australian Institute of Health and Welfare. Oral health and dental care in Australia: key facts and figures trends 2014. Canberra: AIHW; 2014. 6. Ha D, Amarasena N, Crocombe L. The dental health of Australia’s children by remoteness: Child Dental Health Survey Australia 2009. Canberra: Australian Institute of Health and Welfare; 2013. 7. Jamieson L, Roberts-Thomson K. Dental general anaesthetic trends among Australian children. BMC Oral Health 2006;6:16.
519
Arrow & Klobas 8. Alsharif AT, Kruger E, Tennant M. Dental hospitalization trends in Western Australian children under the age of 15 years: a decade of population-based study. Int J Paediatr Dent 2015;25:35–42. 9. de Menezes Abreu DM, Leal SC, Mulder J, Frencken JE. Pain experience after conventional, atraumatic, and ultraconservative restorative treatments in 6- to 7-yr-old children. Eur J Oral Sci 2011;119:163–8. 10. Leal SC, de Menezes Abreu DM, Frencken JE. Dental anxiety and pain related to ART. J Appl Oral Sci 2009;17(Spec Iss):84–8. 11. Lo EC, Holmgren CJ. Provision of Atraumatic Restorative Treatment (ART) restorations to Chinese pre-school children–a 30-month evaluation. Int J Paediatr Dent 2001;11:3–10. 12. Gaynor WN, Thomson WM. Changes in young children’s OHRQoL after dental treatment under general anaesthesia. Int J Paediatr Dent 2012;22:258–64. 13. Jankauskiene B, Virtanen JI, Kubilius R, Narbutaite J. Oral health-related quality of life after dental general anaesthesia treatment among children: a follow-up study. BMC Oral Health 2014;14:18. 14. Turton BJ, Thomson WM, Foster Page LA, Saub R, Ishak AR. Responsiveness of the Child Perception Questionnaire 11–14 for Cambodian children undergoing basic dental care. Int J Paediatr Dent 2015; 25:2–8. 15. Cantekin K, Yildirim MD, Cantekin I. Assessing change in quality of life and dental anxiety in young children following dental rehabilitation under general anesthesia. Pediatr Dent 2014;36:12E–7E. 16. Lee HH, Milgrom P, Starks H, Burke W. Trends in death associated with pediatric dental sedation and general anesthesia. Paediatr Anaesth 2013;23:741–6. 17. Innes NPT, Clarkson JE, Speed C, Douglas GVA, Maguire A. The FiCTION dental trial protocol– filling children’s teeth: indicated or not? BMC Oral Health 2013;13:25. 18. Santamaria RN, Innes NPT, Machiulskiene V, Evans DJP, Splieth CH. Caries management strategies for primary molars: 1-yr randomized control trial results. J Dent Res 2014;93:1062–9. 19. Lee GHM, McGrath C, Yiu CKY. The care of the primary dentition by general dental practitioners and paediatric dentists. Int Dent J 2013;63:273–80. 20. Tickle M, Threlfall AG, Pilkington L, Milsom KM, Duggal MS, Blinkhorn AS. Approaches taken to the treatment of young children with carious primary teeth: a national cross-sectional survey of general dental practitioners and paediatric specialists in England. Br Dent J 2007;203:E4. 21. Graves CE, Berkowitz RJ, Proskin HM, Chase I, Weinstein P, Billings R. Clinical outcomes for early childhood caries: influence of aggressive dental surgery. ASDC J Dent Child 2004;71:114–7. 22. Harrison M, Nutting L. Repeat general anaesthesia for paediatric dentistry. Br Dent J 2000;189:35–7.
520
23. Riordan PJ. Secular changes in treatment in a school dental service. Community Dent Health 1995;12: 221–5. 24. Riordan PJ. Can organised dental care for children be both good and cheap? Community Dent Oral Epidemiol 1997;25:119–25. 25. Pahel BT, Rozier RG, Slade GD. Parental perceptions of children’s oral health: the Early Childhood Oral Health Impact Scale (ECOHIS). Health Qual Life Outcomes 2007;5:6. 26. Arrow P, Klobas E. Evaluation of the Early Childhood Oral Health Impact Scale in an Australian preschool child population. Aust Dent J 2014. doi: 10. 1111/adj.12236. 27. Armfield J. Development and psychometric evaluation of the Index of Dental Anxiety and Fear (IDAF4C). Psychol Assess 2010;22:279–87. 28. Buchanan H, Niven N. Validation of a Facial Image Scale to assess child dental anxiety. Int J Paediatr Dent 2002;12:47–52. 29. Frencken JE, Pilot T, Songpaisan Y, Phantumvanit P. Atraumatic restorative treatment (ART): rationale, technique, and development. J Public Health Dent 1996;56:135–40; discussion 61–3. 30. Honkala E, Behbehani J, Ibricevic H, Kerosuo E, Al-Jame G. The atraumatic restorative treatment (ART) approach to restoring primary teeth in a standard dental clinic. Int J Paediatr Dent 2003;13:172–9. 31. Pocock S. Clinical trials: a practical approach. Chichester: John Wiley & Sons; 1983. 32. Drury TF, Horowitz AM, Ismail AI, Maertens MP, Rozier RG, Selwitz RH. Diagnosing and reporting early childhood caries for research purposes. J Public Health Dent 1999;59:192–7. 33. FDI Commission on Oral Health Research & Epidemiology. A review of the developmental defects of enamel index (DDE Index). Int Dent J 1992;42:411–26. 34. Frencken J, Van Amerongen E, Phantumvanit P, Songpaisan Y, Pilot T. Manual for the atraumatic restorative treatment approach to control dental caries. Groningen, the Netherlands: WHO Collaborating Centre for Oral Health Services Research; 1997. 35. Altman DG. Confidence intervals for the number needed to treat. BMJ 1998;317:1309–12. 36. StataCorp. Stata: release 12. College Station, TX: StataCorp LP; 2011. 37. Schwartz D, Lellouch J. Explanatory and pragmatic attitudes in therapeutic trials. J Chronic Dis 1967;20: 637–48. 38. Riordan PJ, Espelid I, Tveit AB. Radiographic interpretation and treatment decisions among dental therapists and dentists in Western Australia. Community Dent Oral Epidemiol 1991;19:268–71. 39. Schriks M, Van Amerongen W. Atraumatic perspectives of ART: psychological and physiological aspects of treatment with and without rotary instruments. Community Dent Oral Epidemiol 2003;31:15–20.
Ó 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Minimum intervention dentistry approach to managing early childhood caries: a randomized control trial
Peter Arrow and Elizabeth Klobas Dental Health Services, Bentley Delivery Centre, Perth, WA, Australia
Arrow P, Klobas E. Minimum intervention dentistry approach to managing early childhood caries: a randomized control trial. Community Dent Oral Epidemiol 2015; 43: 511–520. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd Abstract – Objectives: A pragmatic randomized control trial was undertaken to compare the minimum intervention dentistry (MID) approach, based on the atraumatic restorative treatment procedures (MID–ART: Test), against the standard care approach (Control) to treat early childhood caries in a primary care setting. Methods: Consenting parent/child dyads were allocated to the Test or Control group using stratified block randomization. Inclusion and exclusion criteria were applied. Participants were examined at baseline and at follow-up by two calibrated examiners blind to group allocation status (j = 0.77), and parents completed a questionnaire at baseline and follow-up. Dental therapists trained in MID–ART provided treatment to the Test group and dentists treated the Control group using standard approaches. The primary outcome of interest was the number of children who were referred for specialist pediatric care. Secondary outcomes were the number of teeth treated, changes in child oral health-related quality of life and dental anxiety and parental perceptions of care received. Data were analyzed on an intention to treat basis; risk ratio for referral for specialist care, test of proportions, Wilcoxon rank test and logistic regression were used. Results: Three hundred and seventy parents/carers were initially screened; 273 children were examined at baseline and 254 were randomized (Test = 127; Control = 127): mean age = 3.8 years, SD 0.90; 59% male, mean dmft = 4.9, SD 4.0. There was no statistically significant difference in age, sex, baseline caries experience or child oral health-related quality of life between the Test and Control group. At follow-up (mean interval 11.4 months, SD 3.1 months), 220 children were examined: Test = 115, Control = 105. Casenotes review of 231 children showed Test = 6 (5%) and Control = 53 (49%) were referred for specialist care, P < 0.0001. More teeth were filled in the Test group (mean = 2.93, SD 2.48) than in the Control group (mean = 1.54, SD 2.20), Wilcoxon’s test, P < 0.0001. Logistic regression, after controlling for age and baseline caries experience, showed a higher risk of referral by allocation to control group, OR 32.6, 95% CI 10.8–98.4, P < 0.0001. Conclusion: The MID–ART approach reduced significantly the likelihood of referral for specialist care, and more children and teeth were provided with treatment.
The minimum intervention dentistry approach to manage dental caries and its potential role in the provision of public dental services have been canvassed in the dental literature (1). The Atraumatic doi: 10.1111/cdoe.12176
Key words: atraumatic restorative treatment; clinical trials; early childhood caries Peter Arrow, Dental Health Services, Bentley Delivery Centre, Locked Bag 15, Perth, WA 6983, Australia Tel.: +61 8 93130600 Fax: +61 8 93134141 e-mail: [email protected], peter. [email protected]
Trials register (ANZCTR: ACTRN12612000474853). Submitted 5 January 2015; accepted 20 May 2015
Restorative Treatment (ART) approach, initially developed to assist dental care delivery in underserved communities, where access to electricity and running water may not be readily available, is
511
Arrow & Klobas
now increasingly seen to have relevant applications in subpopulation groups in developing as well as developed countries (2). The ART approach principally relies on removing caries using hand instruments alone without the administration of a local anesthetic and restoration of the prepared cavity with glass–ionomer cement. Since its introduction in the 1990s, the approach has been examined and tested in different populations and subgroups, such as children, the elderly and people with disabilities, and has been shown to be clinically successful as well as being acceptable to patients (3). Management of early childhood caries (ECC) is demanding of clinician time and skills. In some instances, the treatment of ECC is delivered in a hospital setting under general anesthesia (GA), which is costly and not without risk. Admissions to a hospital for dental care are classified as potentially preventable with timely and adequate nonhospital care (4). However, there is a trend of increasing hospital admissions for dental care among children, especially among 0- to 4-year-olds (5), and in Australia, this is occurring in spite of the apparent low dental caries experience among children (6, 7). Western Australia had the highest, and increasing, rate of hospital admission for dental treatment among all the States and Territories in Australia, and the cost for hospital admission for dental care for children was approximately A $9 million per year (5, 8). The ART approach to restorative care among children has been reported to be well accepted (3), and children reported experiencing less dental anxiety and pain compared with standard care approaches (9, 10). The ART approach has also been shown to be acceptable for preschool children (11). While there have been studies reporting improved changes in child oral health-related quality of life (COHRQoL) after oral rehabilitation under GA (12, 13), there are few reports of the impact on COHRQoL after dental treatment in primary care settings (14). Recent reports also suggest that treatment of caries in children under GA may provoke higher levels of dental anxiety (15), and the approach is not without risks (16). Different approaches to managing caries in the primary dentition including ECC have been, and continue to be, tested (17, 18). There are also reports of variations between general dental practitioners and specialist pediatric dentists in managing primary tooth caries in young children, with general practitioners preferring a less aggressive
512
approach using atraumatic techniques, while specialists preferring a more comprehensive rehabilitation (19, 20). Care for children under GA tends to be more aggressive (21) than in other settings, in particular to avoid repeat GA episodes (22). But, limited research evidence is available on the potential for treatment approaches to reduce the need for specialist care and GA treatment. In Western Australia, public dental services are provided through Dental Health Services of Western Australia (DHS). The School Dental Service (SDS) of the DHS, through employing dental therapists, has been the mainstay of a successful, publicly provided dental care for 5- to 17-year-old schoolchildren since the early 1970s (23, 24). Dental therapists were trained for 2 years (it is now a 3-year training program) and provide basic primary care that includes oral examination and diagnosis, fissure sealing, application of fluoride varnish, scale and clean, oral health counseling, restoration of primary and secondary teeth using plastic materials and administration of a local anesthetic, and extractions of primary teeth. The general dental clinics of the DHS provide dental care to adults and preschool children (0- to 4-year-old) eligible for subsidized care (liable for copayments), with restorative care principally being provided by dentists. Some government general dental clinics do utilize dental therapists, principally to provide dental hygienist services. Standard treatment for caries in children requiring restorative care within the DHS would involve the use of a local anesthetic and rotary instruments. The aim of this study was to test the hypothesis that provision of dental care to children with ECC by dental therapists using a hybrid ART approach would be at least as successful as that of managing ECC by the standard care protocols prevailing within DHS. Successful management of care was defined as management of the necessary dental treatment of the child without referral for specialist pedodontic care.
Materials and methods Ethical approval for the study was provided by the University of Western Australia Human Research Ethics Committee. Participating parents provided signed informed consent. The study was a pragmatic, parallel group, open label, randomized control trial in a primary healthcare setting of a public health dental service in
Minimum intervention dentistry early childhood caries
Perth, Western Australia. Participants were recruited over a 12-month period through invitation poster displays of young children’s teeth at various stages of ECC (precavitated enamel lesions to frank cavitation) at preschool centers, child health clinics and school dental clinics of the DHS with an offer of free primary dental care for their child (not specialist care). Inclusion and exclusion criteria were applied. Children younger than 6 years of age with ECC were eligible. Exclusions were children with acute dental pain and/or infections; physical/mental disability which limited dental treatment in the primary care setting; general or dental developmental conditions, which required specialist care (such as, amelogenesis imperfecta, cleft lip/palate); and children who were caries free. Initial screening was undertaken over the phone and a secondary screening was undertaken at a baseline examination (prior to group allocation) conducted at SDS clinics. Parents completed a baseline questionnaire after the initial screening, but before the baseline examination. The questionnaire sought information about the COHRQoL using the Early Childhood Oral Health Impact Scale (ECOHIS) and collected sociodemographic characteristics (25). The ECOHIS score ranges 0–52. The ECOHIS has been shown to have acceptable reliability and validity among Australian preschool children (26). Parents also completed a dental fear and anxiety schedule (27), and children older than 3 years of age completed a faces child dental fear scale, range 1–5 (dichotomized into afraid, score 1 or 2; or not afraid, score 3, 4, or 5; score 1 was classified as severe fear) at the baseline examination appointment (28). The ECOHIS and faces child dental fear were completed again at follow-up. Socioeconomic level was determined through parent response to a question on eligibility for government concession card. In Western Australia, concession card holders (CCHs) can access a range of services, such as dental and medical services at government agencies at subsidized cost. Children were examined for follow-up oral health status at SDS clinics (6–12 months after group allocation) by the same examiners who conducted the baseline examination and who were blind to group allocation status. The 6- to 12-month follow-up interval was deemed long enough to be able to observe the outcomes. The primary outcome of interest was the number of children referred for specialist pediatric care. The information was abstracted from the clinical records of the
children where a letter of referral for specialist pedodontic care or where clinical notes indicate that a verbal discussion was held with parents recommending specialist consultation for care was taken as being referred for specialist care. Secondary outcome measures were changes in caries experience using the dmft/dmfs index (decayed, missing, and filled primary teeth/tooth surfaces), COHRQoL, and child dental fear over the follow-up interval and between the groups. Teeth treated with restorations were evaluated using the criteria established for evaluating ART restorations (29, 30). Parental perspectives on care delivered through the two arms of the trial were elicited through focus group interviews with a sample of parents from Test and Control groups. Parents were also asked a global question on how their child’s oral health changed through the participation in the study on a five-point scale: 1 = worsened a lot, 2 = worsened a little, 3 = stayed the same, 4 = improved a little, and 5 = improved a lot. The response was dichotomized into improved (score 4 or 5) or no change/worsened (score 1, 2 or 3). Sample size estimation was based on demonstrating equivalence of effect between the two arms of the randomized trial (31). The primary outcome measure was the number (proportion) of children referred for specialist care for ECC. Little information was available on the number of children referred through the standard care approach within DHS, and it was assumed that 10% of children were referred through standard care, and the outcome was deemed equally effective if care provided using the MID–ART approach was at most 10% less effective than standard care, that is, at most 20% referred. The findings will be deemed statistically significant at a = 0.05 with an 80% power to detect a true difference (b = 0.2). A sample size of 145 in each arm of the trial was required. CONSORT participant flow chart is shown in Fig. 1. Consenting parent/child dyads were stratified on caries experience (≤5 dmfs and >5 dmfs) after baseline examination. A central study coordinator allocated the children to Test or Control group using a computer-generated block randomization procedure, using blocks of two, four, and six. Two calibrated dentists undertook the baseline (and follow-up) examinations (interexaminer j = 0.77) at SDS clinics, using established criteria to classify ECC (32) and the presence of enamel defects using the Developmental Defects of Enamel index (DDE) (33). Repeat examinations were not
513
Arrow & Klobas
caries and without the use of a local anesthetic. However, use of rotary instruments was permitted based on the dental therapist’s assessment of the need for, and the child’s capacity to cope with, the procedure. The approach was underpinned by minimum intervention dentistry whereby caries removal was undertaken to the extent of achieving caries-free status at the dentino-enamel junction while leaving softened caries on the pulpal floor if its removal was likely to expose the dental pulp. The prepared cavity was restored with glass–ionomer cement, any remaining at-risk pits and fissures sealed and fluoride varnish applied to all approximal surfaces, and noncavitated enamel lesions. Children were then recalled at 3- to 6-month intervals, at which time repairs to restorations and fluoride varnish application and further counseling on preventive measures were undertaken. Dental radiographs were not routinely taken. Control group children were provided with care by clinicians working in government general dental clinics (with capacity for 8–10 clinicians) in the Perth metropolitan area, and the coordinating senior dentist allocated the child to a clinician within the clinic for dental care (dentists or dental therapists). Dentists principally provided the restorative care in government general dental clinics, and dental therapists, where employed, would provide hygienist services. Clinicians providing
undertaken due to the age of the study participants and logistical difficulties of arranging a second appointment for families with young children. After group allocation, the study coordinator contacted the Test clinicians or Control general dental clinics and provided participant details to enable scheduling of appointments for follow-up care.
Treatment procedures The Test group children were provided with care by school dental therapists (n = 5), who volunteered for the study and were provided with additional training on the MID–ART approach to manage dental caries. Training was undertaken over a half-day session, a PowerPoint presentation covering the ART approach to restorative care and age appropriate behavioral techniques based on the manual by developers of the ART approach (34). The dental therapists then also undertook a restorative procedure on a primary tooth of a preprimary child, scheduled for routine restorative care, using the ART approach. Treatments for the Test children were provided in SDS clinics (with capacity for two clinicians and situated on primary school grounds) by dental therapists. The approach was a hybrid ART in that hand instruments principally were used to remove
Telephone screen = 370 Excluded = 84 Completed Baseline questionnaire = 286 Failed to attend baseline examination = 13 Baseline examination = 273 Excluded = 19
Randomised = 254
dmfs≤5 = 54 dmfs>5 = 73
Control = 127
Test = 127
At least 1 visit for care = 122
Follow-up = 220
Lost to follow-up = 12
Test Follow-up = 115
514
dmfs≤5 = 56 dmfs>5 = 71
At least 1 visit for care = 111 Lost to follow-up = 22
Control Follow-up = 105
Fig. 1. CONSORT participant flowchart.
Minimum intervention dentistry early childhood caries
care to Control children were advised to provide care in the usual manner using the usual standard care approach for a child of this age, and for the children to be reviewed as per the DHS protocol. No other instructions on who should provide the care or how the care was to be delivered were provided. Standard treatment approach usually involves the use of a local anesthetic, rotary instruments and restoration of the prepared cavity with adhesive restorations (glass–ionomer, or resin cements). Patient management strategies used within the primary care setting of the DHS were principally behavioral management techniques. The use of other approaches such as inhalation sedation was not available, and pharmacological approaches were at the discretion and skill level of the treating clinician (for dentists only). Both Test and Control clinicians were advised that they were to provide care to the children to the best of their ability. The outcomes of interest were not communicated to any of the clinicians. All clinicians were further advised that referral for specialist care was an available option if the treating clinician felt that the required care was best managed through referral to a specialist. Referral option for a dental therapist in the first instance was to refer to a supervising dentist, who in turn determined whether referral to a specialist was required, while dentists can refer directly for specialist care. Data were analyzed on an intention to treat basis. Point estimates of effect (cumulative incidence) and the associated 95% confidence intervals for specialist referral were calculated, and the number needed to treat (1/absolute risk difference) to avoid one additional specialist referral was calculated (35). Tests of proportions for categorical variables between two groups (chi-square) and test of means using parametric (t-test or paired t-test) and nonparametric (Kruskal–Wallis test or Wilcoxon signed-rank test) statistical procedures were used. Multivariate analysis using logistic regression was used to control for any potentially confounding factors that were not balanced at baseline or at follow-up. Statistical analyses were undertaken on a personal computer using STATA 12 (36).
Results The distribution of baseline characteristics of the study participants overall and within Test (T) and Control (C) groups suggests that randomization
was fair, Table 1. There were no statistically significant differences between the T and C groups on any of the factors as indicated by overlapping 95% confidence intervals. The age range of study participants was (Test, 1.0–5.5 years; Control, 1.1– 5.2 years). Initially, there were five clinicians (dental therapists) providing care to T children; however, during the study, one clinician was relocated and one clinician commenced maternity leave. The coordinating senior clinician at the metropolitan government general dental clinic (ten metropolitan clinics) had the discretion, based on their judgment, to allocate the C participant child to a clinician at that clinic. In all but one clinic, dentists provided the care to C children (one dental therapist provided care to four children at one government general dental clinic). As shown in Fig. 1, after a mean interval of 11.4 months (range 5.8–22.7 months; 95% CI 10.9, 11.8), there were 220 children examined at followup. Reasons for lost to follow-up were moving away from the area (n = 4) and noncontactable (n = 27) or failed to attend the requested examination appointments (n = 3). The baseline characteristics of those followed up and lost to follow up are shown in Table 2. Those lost to follow up were significantly younger than those followed up, Wilcoxon rank-sum, P < 0.001. CCHs and those who were referred for specialist care were also less likely to have attended for a follow-up examination. Although more Test group participants were followed up, the difference was not statistically significant, v2, P = 0.065, Table 3. Also, those lost to follow up among the C group were significantly younger than those followed up, and among the C group, those referred for care were less likely to have had a follow-up examination, Fisher’s exact test, P < 0.05, while among the Test group the difference was of borderline significance, Fisher’s exact test, P = 0.049. Table 4 shows the primary outcome evaluated in this study. There were more children from the C group that were referred for specialist pedodontic care than in the T group (all children requested for referral to a specialist by the dental therapists in the Test group were referred for specialist care by the supervising dentist), v2, P < 0.0001. The risk ratio for children being referred for specialist care if allocated to the C group was 10.1, 95% CI 4.5, 22.5, and the risk difference was 0.44, 95% CI 0.34, 0.54. The number needed to treat for one additional child to avoid specialist referral using the test
515
Arrow & Klobas Table 1. Baseline characteristics of study participants overall and within groups Factor
Total
Test n = 127 (% or mean and 95% CI)
Control n = 127 (% or mean and 95% CI)
Sex Age (years) dmft dmfs Parent fear Child fear % Afraid ECOHIS CCH
Male = 59% (53, 65) 3.8 (3.7, 3.9) 4.9 (4.4, 5.4) 11.0 (9.5, 12.4) 16.7 (15.5, 17.8) 11.4% (6.9, 15.9) 11.1 (10.0, 12.2) Yes = 41% (36, 47) No = 59% (53, 64)
Male = 55% (46, 63) 3.9 (3.8, 4.0) 4.9 (4.2, 5.5) 10.8 (8.7, 12.9) 17.3 (15.6, 19.1) 12.5% (6.1, 18.9) 10.1 (8.7, 11.4) Yes = 37% (29, 45) No = 63% (55, 71)
Male = 63% (55, 71) 3.7 (3.5, 3.9) 5.0 (4.3, 5.7) 11.2 (9.1, 13.2) 16.0 (14.5, 17.5) 10.1% (3.8, 16.5) 12.1 (10.5, 13.7) Yes = 45% (36, 54) No = 55% (46, 64)
dmft = mean count of decayed, missing filled primary teeth; dmfs = mean count of decayed, missing and filled primary tooth surfaces; ECOHIS = mean Early Childhood Oral Health Impact Scale; CCH = concession card holder.
Table 2. Baseline characteristics of those followed up and those lost to follow up Factor Sex Male Female Age (years) dmft dmfs Parent fear Child fear % afraid ECOHIS CCH† Referred‡
Followed up n = 220 (% or mean and 95% CI)
Lost to follow up n = 34 (% or mean and 95% CI)
N = 130 (59.1; 52.5, 65.6) N = 90 (40.9; 34.4, 47.5) 3.9 (3.8, 4.0)* 4.9 (4.4, 5.5) 11.0 (9.4, 12.7) 16.8 (15.5, 18.0) 11.1 (6.4, 15.9) 11.3 (10.1, 12.4) Yes = 82 (37.3) No = 138 (62.7) Yes = 45 (21.5) No = 164 (78.5)
N = 20 (58.8; 42.0, 75.7) N = 14 (41.2; 24.3, 58.0) 3.2 (2.8, 3.6)* 4.8 (3.7, 6.0) 10.5 (7.1, 13.9) 15.9 (13.0, 18.8) 13.6 ( 1.1, 28.4) 10.0 (7.1, 13.0) Yes = 22 (66.7) No = 11 (33.3) Yes = 14 (63.6) No = 8 (36.4)
dmft = mean count of decayed, missing filled primary teeth; dmfs = mean count of decayed, missing and filled primary tooth surfaces; ECOHIS = mean Early Childhood Oral Health Impact Scale; CCH = concession card holder. *Wilcoxon rank-sum test P < 0.001; †CCH status for one participant missing, and Pearson’s chi-square, P < 0.05; ‡ Pearson’s chi-square, P < 0.001.
intervention was 3 (95% CI 2–3). For 23 children, information regarding referral status was not available either because the child did not attend any of the allocated appointments (n = 20; T = 4, C = 16) or had attended for an examination but then subsequently failed to attend any further appointments (n = 3: C = 3). One child in the Test group failed to attend for any treatment appointments, but was seen for the follow-up examination and provided information about the care received. A sensitivity analysis was undertaken to examine the extent of changes to the principal findings by assuming, at the limit, that all children with missing data in the Test group were referred for specialist care (four children) and all children with missing data in the control group were not referred for specialist care (19 children). The Pearson’s chi-square was P < 0.001, and the risk ratio was reduced from 10.1 to 5.3. Conversely, if the assumptions were reversed,
516
all Test children not referred and all control children referred, the risk ratio increased to 12.0, Pearson’s chi-square P < 0.001. The findings for caries experience between T and C are shown in Fig. 2. The T children had significantly fewer tooth surfaces and teeth that were decayed (T ds = 4.4, SD 9.3; C ds = 8.6, SD 11.9: T dt = 2.0, SD 3.0; C dt = 3.8, SD 4.2) and more tooth surfaces and teeth that had been filled (T fs = 7.1, SD 7.3; C fs = 3.6, SD 6.3: T ft = 2.9, SD 2.5; C ft = 1.5, SD 2.2) compared with children in the C group. There was also a trend of lower caries increment in the T group, although this was not statistically significant, Kruskal–Wallis test, P = 0.29. There was no change in the level of fear reported by the children in either group from baseline to follow-up, Wilcoxon signed-rank test, P = 0.87. However, at follow-up, the proportion of children reporting
Minimum intervention dentistry early childhood caries Table 3. Baseline characteristics of those followed up and lost to follow up among Test and Control participants Test group Followed up n = 115 (mean or %; and 95% CI)
Factor Sex (N) Male Female Age (years) dmft dmfs Parent fear Child fear % afraid ECOHIS CCH Referred
Control group Lost to follow up n = 12 (mean or % and 95% CI)
63 (55) 52 (45) 3.9 (3.8, 4.1) 5.0 (4.2, 5.7) 11.2 (8.9, 13.5) 17.7 (15.9, 19.5) 11.5 (5.0, 17.9) 10.3 (8.9, 11.8) Yes = 39 (33.9) No = 76 (66.1) Yes = 4 (3.5) No = 111 (96.5)
Followed up n = 105 (mean or % and 95% CI)
Lost to follow up n = 22 (mean or % and 95% CI)
67 (64) 38 (36) 3.8 (3.7, 4.0)* 4.9 (4.1, 5.7) 10.9 (8.5, 13.2) 15.8 (14.1, 17.5) 10.7 (3.6, 17.8)
13 (59) 9 (41) 3.1 (2.7, 3.5)* 5.4 (3.8, 7.0) 12.7 (7.8, 17.6) 17.1 (13.6, 20.6) 7.1 ( 7.7, 22.0)
7 (58) 5 (42) 3.4 (2.5, 4.3) 3.8 (2.2, 5.3) 6.6 (3.3, 9.8) 13.3 (7.5, 19.2) 25.0 (-9.0, 59.0) Yes No Yes No
7.6 (3.6, 11.6) = 8 (66.7) = 4 (33.3) = 2 (25.0) = 6 (75.0)
Yes No Yes No
12.3 (10.5, 14.1) = 43 (41.0) = 62 (59.0) = 41 (43.6)† = 53 (56.4)
11.4 (7.2, 15.5) Yes = 14 (66.7) No = 7 (33.3) Yes = 12 (85.7)† No = 2 (14.3)
dmft = mean count of decayed, missing filled primary teeth; dmfs = mean count of decayed, missing and filled primary tooth surfaces; ECOHIS = mean Early Childhood Oral Health Impact Score; CCH = concession card holder. *Wilcoxon rank-sum test P < 0.001; †Fisher’s exact test, P < 0.05.
Table 4. Children referred for specialist pedodontic care among groups
Referred Yes No Total
Test, n (%)a
Control, n (%)b
Total
6 (5) 117 (95) 123 (100)
53 (49) 55 (51) 108 (100)
59 (26) 172 (74) 231 (100)
v2 = 59.1, P < 0.0001. a One Test child who failed to attend for any treatment appointments presented for follow-up examination and provided information regarding referral status. b Three Control children failed to attend for treatment appointments and had no information regarding referral status in the case notes.
more severe dental fear was significantly higher in the C group, v2, P < 0.05. Parent reported COHRQoL was significantly better at follow-up compared with baseline for both groups, paired t-test, P < 0.05, but there were no statistically significant differences between the groups, P > 0.05. Two hundred and two parents responded to the global question of the change in their child’s oral health (T = 107, C = 95) of which 78 T (73%) and 40 C (42%) reported improved oral health, v2, P < 0.001. Multivariate analysis using logistic regression to model the risk of referral for specialist care found a significant effect of age and caries experience as well as group allocation. Older children were less likely to be referred, OR 0.5, P = 0.003, while an increase of one dmfs increased the odds of referral
by 10%, OR 1.1, P < 0.001. The highest odds of referral were for those children allocated to the C group, OR 32.6, P < 0.001.
Discussion The study’s hypothesis that the MID–ART approach was as effective as the standard care approach in managing ECC in a primary care setting, without the need for referral for specialist care, was rejected. The hybrid MID–ART intervention approach as undertaken reduced the rate of referral for specialist pedodontic care by 45% (Test 5%, Control 49%). In Australia, a child being referred to a specialist for management of dental caries is likely to be treated under GA, and WA also has the highest, and increasing, rate of hospital admission for dental treatment among all the States and Territories (5). A limitation of the study was that the study was a pragmatic trial, undertaken within a functioning public dental service delivering primary dental care using standard care procedures. The control was the standard care approach, which was delivered at government general dental clinics where dentists and dental therapists are employed, and dentists provided the majority of restorative care and dental therapists principally provided dental hygienist services. The pragmatic design allowed standard care to be delivered by clinicians within the general dental clinics; no restrictions were
517
Arrow & Klobas
* * *
*
Fig. 2. Comparison of caries experience outcome measures between Test and Control groups. ds = decayed primary tooth surface; ms = missing primary tooth surface; fs = filled primary tooth surface; dmfs = decayed, missing, filled primary tooth surfaces; dt = decayed primary tooth; mt = missing primary tooth due to caries; ft = filled primary tooth; dmft = decayed missing, filled primary teeth; dmfs incr = dmfs follow-up – dmfs baseline; dmft incr = dmft follow-up – dmft baseline. Error bars indicate 95% confidence limits; *95% CI do not overlap, significant at P < 0.05.
placed on the type of clinician providing the care, because the trial was testing the MID–ART approach rather than the type of clinician (dentist/ therapist) delivering the care. In one instance, four Control children were treated for ECC by a dental therapist. This was unlikely to have biased the study’s findings. Also, the strength of pragmatic trials was that interventions were tested within a real-world situation and were felt to be better at evaluating the effectiveness of an intervention (37). Little prior information was available on the likely referral patterns of the standard care clinicians. Information available on treatment decisionmaking between dentists and dental therapists within the DHS suggests that the approach to caries management was similar and it was unlikely that they will differ significantly when caring for preschool children under standard care conditions (38). It was possible and more likely that dentists who usually provide care to preschool children in public dental services were less likely to refer a child for specialist care because of greater experience in caring for children of such an age than dental therapists, who usually do not care for preschool children, and the sample size estimation was based on this premise. Participants were screened to exclude children with acute conditions requiring urgent care, and the study’s findings may not apply to those children. Participants were not randomly selected, but were those who responded to a poster display or recruited through child health nurses who may have noticed an oral health problem which required professional care. Assessment of baseline variables indicated that the stratified randomization process was fair. Although more children from
518
the T group were examined, there was no statistically significant difference between the T and C groups. Control group children that had a followup examination were significantly older and less likely to have been referred for specialist care. All children in the study were classified as having severe ECC (32). Referral status was associated with a child attending for a follow-up examination, fewer children who were referred for specialist care attended for a follow-up examination. Anecdotal information from the study trial coordinator was that the parents responded to the request for a follow-up attendance with comments such as they see little value in attending for a follow-up when their child was not provided with the needed care in the primary care setting and they are now confronted with costly specialist care. An observation, not reported in this manuscript, was that the majority of children who were referred for care had not followed through with the recommended care, principally due to cost. Sensitivity analysis to account for those lost to follow up suggests that the principal findings would not have altered significantly. Secondary outcome measures also indicated greater effectiveness of the MID–ART approach in the provision of primary care. More children in the T group were provided with restorative care (T fs = 7.1, SE 0.7; C fs = 3.6, SE 0.6) and had nearly half the level of untreated decay (T ds = 4.4, SE 0.9; C ds = 8.6, SE 1.2). There was also a lower level of caries increment among the T children. T children who attended for care were provided with comprehensive preventive care, at least one application of fluoride varnish, sealing of at-risk pits and fissures, and oral health counseling to prevent dental caries,
Minimum intervention dentistry early childhood caries
and this may explain the observed difference in caries increment. Additionally, more parents in the T group also perceived their child’s oral health had improved as a result of the care provided. Most of the studies that have examined the changes in COHRQoL have been undertaken in relation to oral rehabilitation under GA and have reported improvements after oral rehabilitation (12, 13). Delivery of dental care to children in a primary healthcare setting can improve their quality of life (14). This study found that dental care delivered in a primary care setting improved children’s oral health-related quality of life (both the T and C groups reported improved ECOHIS). A full report of the findings in relation to early childhood oral health-related quality of life is in a companion manuscript, which is under review. Although oral rehabilitation for ECC under GA has reported improvements in COHRQoL, oral rehabilitation under GA may increase child dental anxiety, which may be related to a more ‘aggressive’ approach undertaken, with more teeth extracted to avoid repeat GA episodes (15, 21, 22). The MID–ART approach on the other hand has been reported to provoke less stressful behavior in children undergoing restorative treatment (39). Although this study did not compare the MID– ART approach against the treatment under GA or that of specialist provided care, the finding of lower levels of dental fear in the T children suggests that the MID–ART approach may mitigate the development of severe dental fear. The MID–ART approach has been suggested as the most appropriate measure for managing ECC because of its ‘atraumatic’ approach (3). Other studies have reported lower levels of pain and anxiety experienced by children when treated using the ART approach compared with the standard approach. Although this study did not measure perceived pain by children, the overall level of anxiety and dental fear was assessed and indicated that children in the MID–ART group experienced lower levels of severe dental fear.
Conclusions The findings from the study suggest significant potential to reduce the rate of hospital admissions for dental treatment among preschool children by reducing the rate of referral for specialist care, which was strengthened by the sensitivity analyses. Delivery of care by dental therapists using the
MID–ART approach was well accepted by parents and children. Significantly improved COHRQoL was reported by the parents of both groups after treatment, but more parents of the T group rated their child’s oral health as improved. The T children reported lower levels of dental anxiety and had a higher level of dental caries treated. The delivery of care for ECC in a primary care setting appears to have the potential to significantly reduce the cost of care. The approach as tested is also likely to be cost-effective, given that a senior dental therapist’s salary is approximately 60% of that of a senior dentist within DHS.
Acknowledgements The authors would like to thank the study coordinators Rowena McInnes and Di Winston for their untiring efforts in participant recruitment and follow-up and the clinicians who participated in the study, especially the MID–ART clinicians; Deborah Arrow, Jodie Burton, Lesley Dwyer, Helen Forrest, and Chris Smith; and the DHS in supporting the study. The study was partly funded by the Targeted Research Fund of the WA Health Department. Lastly, the study would not have been possible without the willing participation of parents and children in this study. The authors declare no potential conflict of interests with respect to authorship or publication of this manuscript.
References 1. Calache H, Hopcraft MS, Martin JM. Minimum intervention dentistry – a new horizon in public oral health care. Aust Dent J 2013;58(1 Suppl):17–25. 2. Holmgren CJ, Roux D, Domejean S. Minimal intervention dentistry: part 5. Atraumatic restorative treatment (ART)–a minimum intervention and minimally invasive approach for the management of dental caries. Br Dent J 2013;214:11–8. 3. Frencken JE, Leal SC, Navarro MF. Twenty-five-year atraumatic restorative treatment (ART) approach: a comprehensive overview. Clin Oral Investig 2012;16:1337–46. 4. Katterl R, Anikeeva O, Butler C, Brown L, Smith B, Bywood P. Potentially avoidable hospitalisations in Australia: causes for hospitalisations and primary health care interventions. Adelaide: Primary Health Care Research & Information Service; 2012. 5. Australian Institute of Health and Welfare. Oral health and dental care in Australia: key facts and figures trends 2014. Canberra: AIHW; 2014. 6. Ha D, Amarasena N, Crocombe L. The dental health of Australia’s children by remoteness: Child Dental Health Survey Australia 2009. Canberra: Australian Institute of Health and Welfare; 2013. 7. Jamieson L, Roberts-Thomson K. Dental general anaesthetic trends among Australian children. BMC Oral Health 2006;6:16.
519
Arrow & Klobas 8. Alsharif AT, Kruger E, Tennant M. Dental hospitalization trends in Western Australian children under the age of 15 years: a decade of population-based study. Int J Paediatr Dent 2015;25:35–42. 9. de Menezes Abreu DM, Leal SC, Mulder J, Frencken JE. Pain experience after conventional, atraumatic, and ultraconservative restorative treatments in 6- to 7-yr-old children. Eur J Oral Sci 2011;119:163–8. 10. Leal SC, de Menezes Abreu DM, Frencken JE. Dental anxiety and pain related to ART. J Appl Oral Sci 2009;17(Spec Iss):84–8. 11. Lo EC, Holmgren CJ. Provision of Atraumatic Restorative Treatment (ART) restorations to Chinese pre-school children–a 30-month evaluation. Int J Paediatr Dent 2001;11:3–10. 12. Gaynor WN, Thomson WM. Changes in young children’s OHRQoL after dental treatment under general anaesthesia. Int J Paediatr Dent 2012;22:258–64. 13. Jankauskiene B, Virtanen JI, Kubilius R, Narbutaite J. Oral health-related quality of life after dental general anaesthesia treatment among children: a follow-up study. BMC Oral Health 2014;14:18. 14. Turton BJ, Thomson WM, Foster Page LA, Saub R, Ishak AR. Responsiveness of the Child Perception Questionnaire 11–14 for Cambodian children undergoing basic dental care. Int J Paediatr Dent 2015; 25:2–8. 15. Cantekin K, Yildirim MD, Cantekin I. Assessing change in quality of life and dental anxiety in young children following dental rehabilitation under general anesthesia. Pediatr Dent 2014;36:12E–7E. 16. Lee HH, Milgrom P, Starks H, Burke W. Trends in death associated with pediatric dental sedation and general anesthesia. Paediatr Anaesth 2013;23:741–6. 17. Innes NPT, Clarkson JE, Speed C, Douglas GVA, Maguire A. The FiCTION dental trial protocol– filling children’s teeth: indicated or not? BMC Oral Health 2013;13:25. 18. Santamaria RN, Innes NPT, Machiulskiene V, Evans DJP, Splieth CH. Caries management strategies for primary molars: 1-yr randomized control trial results. J Dent Res 2014;93:1062–9. 19. Lee GHM, McGrath C, Yiu CKY. The care of the primary dentition by general dental practitioners and paediatric dentists. Int Dent J 2013;63:273–80. 20. Tickle M, Threlfall AG, Pilkington L, Milsom KM, Duggal MS, Blinkhorn AS. Approaches taken to the treatment of young children with carious primary teeth: a national cross-sectional survey of general dental practitioners and paediatric specialists in England. Br Dent J 2007;203:E4. 21. Graves CE, Berkowitz RJ, Proskin HM, Chase I, Weinstein P, Billings R. Clinical outcomes for early childhood caries: influence of aggressive dental surgery. ASDC J Dent Child 2004;71:114–7. 22. Harrison M, Nutting L. Repeat general anaesthesia for paediatric dentistry. Br Dent J 2000;189:35–7.
520
23. Riordan PJ. Secular changes in treatment in a school dental service. Community Dent Health 1995;12: 221–5. 24. Riordan PJ. Can organised dental care for children be both good and cheap? Community Dent Oral Epidemiol 1997;25:119–25. 25. Pahel BT, Rozier RG, Slade GD. Parental perceptions of children’s oral health: the Early Childhood Oral Health Impact Scale (ECOHIS). Health Qual Life Outcomes 2007;5:6. 26. Arrow P, Klobas E. Evaluation of the Early Childhood Oral Health Impact Scale in an Australian preschool child population. Aust Dent J 2014. doi: 10. 1111/adj.12236. 27. Armfield J. Development and psychometric evaluation of the Index of Dental Anxiety and Fear (IDAF4C). Psychol Assess 2010;22:279–87. 28. Buchanan H, Niven N. Validation of a Facial Image Scale to assess child dental anxiety. Int J Paediatr Dent 2002;12:47–52. 29. Frencken JE, Pilot T, Songpaisan Y, Phantumvanit P. Atraumatic restorative treatment (ART): rationale, technique, and development. J Public Health Dent 1996;56:135–40; discussion 61–3. 30. Honkala E, Behbehani J, Ibricevic H, Kerosuo E, Al-Jame G. The atraumatic restorative treatment (ART) approach to restoring primary teeth in a standard dental clinic. Int J Paediatr Dent 2003;13:172–9. 31. Pocock S. Clinical trials: a practical approach. Chichester: John Wiley & Sons; 1983. 32. Drury TF, Horowitz AM, Ismail AI, Maertens MP, Rozier RG, Selwitz RH. Diagnosing and reporting early childhood caries for research purposes. J Public Health Dent 1999;59:192–7. 33. FDI Commission on Oral Health Research & Epidemiology. A review of the developmental defects of enamel index (DDE Index). Int Dent J 1992;42:411–26. 34. Frencken J, Van Amerongen E, Phantumvanit P, Songpaisan Y, Pilot T. Manual for the atraumatic restorative treatment approach to control dental caries. Groningen, the Netherlands: WHO Collaborating Centre for Oral Health Services Research; 1997. 35. Altman DG. Confidence intervals for the number needed to treat. BMJ 1998;317:1309–12. 36. StataCorp. Stata: release 12. College Station, TX: StataCorp LP; 2011. 37. Schwartz D, Lellouch J. Explanatory and pragmatic attitudes in therapeutic trials. J Chronic Dis 1967;20: 637–48. 38. Riordan PJ, Espelid I, Tveit AB. Radiographic interpretation and treatment decisions among dental therapists and dentists in Western Australia. Community Dent Oral Epidemiol 1991;19:268–71. 39. Schriks M, Van Amerongen W. Atraumatic perspectives of ART: psychological and physiological aspects of treatment with and without rotary instruments. Community Dent Oral Epidemiol 2003;31:15–20.
