Quality of Information Regarding Repair Restorations on Dentist Websites: Systematic Search and Analysis

Background
Repairing instead of replacing partially defective dental restorations represents a minimally invasive treatment concept, and repairs are associated with advantages over complete restoration replacement. To participate in the shared decision-making process when facing partially defective restorations, patients need to be aware of the indications, limitations, and advantages or disadvantages of repairs. Patients are increasingly using the internet to gain health information like this online.


Objective
We aimed to assess the quality of German-speaking dentist websites on repairs of partially defective restorations.


Methods
Three electronic search engines were used to identify German-speaking websites of dental practices mentioning repairs. Regarding information on repairs, websites were assessed for (1) technical and functional aspects, (2) comprehensiveness of information, and (3) generic quality and risk of bias. Domains 1 and 3 were scored using validated tools (LIDA and DISCERN). Comprehensiveness was assessed using a criterion checklist related to evidence, advantages and disadvantages, restorations and defects suitable for repairs, and information regarding technical implementation. Generalized linear modeling was used to assess the impact of practice-specific parameters (practice location, practice setting, dental society membership, and year of examination or license to practice dentistry) on the quality of information. An overall quality score was calculated by averaging the quality scores of all three domains and used as primary outcome parameter. Quality scores of all three domains were also assessed individually and used as secondary outcomes.


Results
Fifty websites were included. The median score of quality of information was 23.2% (interquartile range [IQR] 21.7%-26.2%). Technical and functional aspects (55.2% [IQR 51.7%-58.6%]) showed significantly higher quality than comprehensiveness of information (8.3% [IQR 8.3%-16.7%]) and generic quality and risk of bias (3.6% [IQR 0.0%-7.1%]; P<.001/Wilcoxon). Quality scores were not related to practice-specific parameters (P>.05/generalized linear modeling).


Conclusions
The quality of German-speaking dentist websites on repairs was limited. Despite sufficient technical and functional quality, the provided information was neither comprehensive nor trustworthy. There is great need to improve the quality of information to fully and reliably inform patients, thereby allowing shared decision making.