{
  "abstract": "The use of radiographs (RAD) associated with visual inspection (VI) in the diagnostic strategy for caries detection in children brings more harm due to insignificant therapeutic impact, false-positive results, overdiagnosis and lead-time bias compared to the use of visual inspection (VI) alone (CARDEC-01- NCT02078453; Pontes et al., 2021-doi:10.1186/s12903-021-01528-w). The present study aimed to investigate the professional’s preference for diagnostic strategy (VI+RAD vs. VI alone) and the existence of a possible framing bias in such a choice. Finally, in a secondary analysis, a discrete choice experiment (DCE), we explored how the option for a strategy VI that minimizes overdiagnosis may be potentialized by other attributes related to diagnostic methods for caries detection. This study is part of a preference study designed to assess dental professionals’ preferences (Machado et al., 2024 -doi: 10.1186/s12903-024-04090-3). For this study, we first conducted a DCE focused on a scenario for clinical decision-making when detecting caries lesions. Attributes related to caries detection methods were defined, and their levels (characteristics which define them) were set. One of these attributes was the overdiagnosis. Brazilian dentists who worked in public and private practices were supposed to make a forced choice of one combination of levels within a sequential 18 pairs of alternatives, which defined hypothetical methods. An additional set of pairs of alternatives was also presented to dentists, including labelled and non-labelled methods related to the real methods compared in the CARDEC-01 trial. The comparison of labelled and non-labelled choices might suggest potential framing bias. Panel and simple mixed logit models were used to explore the choice probabilities, dentists’ preferences regarding different attributes, and the choice for labelled vs. non-labelled alternatives. Subgroup analysis was considered for dentists from different types of practices (public/private). Dentists preferred to use the VI for caries detection (64.7%;95%CI:55.5%-72.9%) than the association VI+RAD (35.3%;95%CI:27.1%-44.5%). Convergently, these respondents had a slightly higher chance of preferring hypothetical methods that eventually could not treat a caries lesion (52.0%; 95%CI:49.7- 54.43) instead of treating it unnecessarily (48%;95%CI:45.6–50.3/OR=1.10;95%CI:1.00–1.22, p=0.05). The preference for avoiding overdiagnosis could suffer a 6% reduction if the method leads to unnecessary treatments (OR=0.77;95%CI:0.70–0.85,p<0.001). On the other hand, such preference could be increased for simpler methods, demanding less time or steps (OR:1.20;95%CI:1.11–1.36, p<0.001), such as the VI. Graduate students presented a 2.65 (95%CI:1.05–6.68) higher chance of preferring VI than undergraduate students. Conversely, dentists who were not open to implementing new ideas had approximately 80% less chance of choosing the VI method isolate (OR=0.17; 95%CI: 0.04–0.72 – p=0.01). Labelling the methods did not seem to influence preferences independently of the type of practice and the availability of X-ray devices in the work unit. In conclusion, VI alone is preferred by a slight majority of dentists compared to VI+RAD, reflecting professional preferences for avoiding overdiagnosis but strongly reinforced by preferences for simpler methods and avoidance of unnecessary treatments. These preferences were not influenced by labelling (non-recognition of framing bias), but they are influenced by the professional level of education, signalizing the need for professional education in this sense.",
  "authors": [
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Ana Clara Falabello de Luca"
    },
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Gabriela Manco Machado"
    },
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Renata Paz Leal Pereira"
    },
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Daniela Prócida Raggio"
    },
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Fausto Medeiros Mendes"
    },
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Mariana Minatel Braga"
    },
    {
      "affiliations": [
        "School of Dentistry, University of São Paulo, São Paulo, Brazil"
      ],
      "name": "Laura Regina Antunes Pontes"
    }
  ],
  "title": "059 Radiographs to detect caries lesions in children: are professionals choosing to use them associated with clinical examination despite their negative therapeutic impact and overdiagnosis?",
  "uid": "37e23c08-fa95-5a71-9022-c2be1a15efa9"
}