The use of child oral-health-related quality of life measures in a randomised control trial of the Hall crown technique in a primary care setting

Abstract

The first aim of this thesis was to analyse the health-related (HRQoL) and oral-health-related quality of life (OHRQoL) information collected during the clinical trial of the Hall technique, to determine whether there were differences between those receiving conventional dental treatment (restorations and conventional stainless-steel crowns), and those having the Hall technique of stainless-steel crown placement. The second aim was to see whether the current child HRQoL and OHRQoL measures were responsive to routine dental treatment in the primary health setting.

Method The Transform a Tooth (TAT) study was undertaken in the Whanganui District Health Board Community Oral Health Service (COHS) from 2014-2018 by Dr LA Foster Page and Dr DH Boyd, with funding from the Cure Kids Foundation and support from 3MTM. It was a randomised control trial, investigating the use of the Hall technique for the placement of preformed stainless-steel crowns on children’s primary teeth to treat dental caries. The children involved in the study were aged 3-7 years at the time of recruitment. Prior to the initial appointment consent was obtained, and questionnaires were completed. A clinical examination was undertaken where the baseline oral health data was recorded, and posterior bitewing radiographs were taken. Those children who met the selection criteria, were randomly allocated to the test group (Hall) or control group (non-Hall) and proceeded to have their treatment. The children had subsequent dental examinations and completed repeat OHRQoL questionnaires at the 12-month, and 24-month follow-up appointments. Data from the clinical aspects of the trial are presented elsewhere. The parent questionnaire used in the TAT study contained five components. This included a global question on oral health impact, the short-form 16-item Parent/Caregivers Perceptions Questionnaire (P-CPQ) on OHRQoL, a short form version of the Family Impact Scale (FIS) consisting of eight questions on how oral health may have impacted family life, a global question on how the child’s oral health affects the family, and nine questions on the child’s general quality of life, collected using the Child Health Utility 9D (CHU-9D). The initial parent questionnaire also included the Index of Socioeconomic Deprivation (NZiDep) questions. The child questionnaire included the CHU-9D (child version) of nine questions relating to the child’s quality of life, the seven question Scale of Oral Health Outcomes (SOHO), and five general ‘global’ questions. The OHRQoL questionnaires were administered initially prior to selection into the study, and then repeated at the 12-month and 24-month follow-up appointments.

Findings The children included in the study differed from those excluded by: being more likely to be from a high deprivation area (area-deprivation method only, no difference was found using the household deprivation method), having higher caries experience (respective dmft scores of 4.6 and 1.4), and higher mean OHRQoL scale scores on the both the P-CPQ and SOHO scales. The two randomised treatment groups of Hall (51% of included sample) and non-Hall (49%) had no significant differences in their demographic characteristics, caries experience or mean HRQoL/OHRQoL scale scores. The HRQoL/OHRQoL measures were tested for internal consistency using Cronbach’s alpha co-efficient. The P-CPQ had the strongest score (0.82), and the CHU-9D child version the lowest (0.63). At the 12-month follow-up there were 129 (92%) remaining in the Hall group, and 129 (88%) in the non-Hall group. There was a significantly higher proportion of Māori children lost to follow-up. For those followed up at 12 months, all repeat HRQoL/OHRQoL questionnaires showed slightly higher mean scale scores for the Hall group, with statistical significance for the CHU-9D parent measure. Change scores and effect sizes for the P-CPQ and FIS had small improvements for both Hall and non-Hall groups. The CHU-9D parent version Hall had a small worsening and the non-Hall had a small improvement. The CHU-9D child version Hall group had a moderate worsening and the non-Hall a small worsening. For the SOHO scale the Hall group had a small worsening and non-Hall a small improvement. By the 24-month follow-up fewer than half of the participants remained, with 43% and 48% (respectively) retained in the Hall and non-Hall groups. Despite this relatively high attrition rate there were no statistically significant differences in their demographic characteristics, caries experience or baseline mean HRQoL/OHRQoL scores for those assessed and those lost to follow-up. At that 24-month follow-up the Hall group had higher mean HRQoL/OHRQoL scale scores for three of the measures (P-CPQ, CHU-9D parent, and SOHO) than the non-Hall. The change scores and effect sizes at 24 months for both groups had moderate improvement in their mean the P-CPQ scores, and small improvements in their FIS scores. The CHU-9D parent version had no change for the Hall group and small improvement in the non-Hall, and the SOHO score had no change in the Hall group and small improvement in the non-Hall. The CHU-9D child measure had a small worsening for the Hall group, and moderate worsening in the non-Hall group. However, none of the differences were statistically significant.

Conclusions The HRQoL/OHRQoL information is an important adjunct to the clinical data collected during clinical trials. It provides valuable information on the impact that different treatment modalities have on children and their families. This is important when new clinical techniques are being considered, such as the Hall technique, which should not be judged just on its clinical success or failure. It is important to consider the timeliness of the questionnaires with regard to the treatment intervention, and whether we can directly infer the studied intervention technique without allowing for the confounding variables over the period of a year. The children receiving the Hall crown treatment technique after 12 month had higher scale scores, but this effect was not clear at the 24-month review, with mixed results between the two groups. The P-CPQ was the only measure which did not have a large floor effect in the scale scores, and had the best internal reliability. The P-CPQ score showed a small improvement at 12-months and moderate improvement at 24 months in both the Hall and non-Hall groups. The P-CPQ may be the best measure suited to this younger age group (age 3-8) in this scenario type (routine dental treatment in primary care setting). Careful consideration needs to be given to whether these measures have the responsiveness to identify any meaningful change.