Objectives: To determine predictive indicators for dental health management of patients with different abilities, according to a new screening protocol. Materials and methods: This was an observational, analytical, retrospective, cross-sectional study. The sample consisted of all patients with different abilities, diagnosed with autism, Down syndrome, cognitive deficit, cerebral palsy and others, who received care between 1999 and 2019 at the Oral Health Service of the Centro Ann Sullivan del Peru and met the inclusion criteria. Data were obtained from the clinical histories to determine the predictive indicators for managing dental care in these patients. A multivariate statistical analysis was performed using binary logistic regression test. The research project was evaluated by the Research Project Reviewing Committee and then referred to the Human Ethics Committee from a university, securing the relevant approvals. Results: Of the 589 cases evaluated, mean age was 14.5 years, 67.9% were male; systemic diagnosis of different abilities was autism in 62%, the most frequent treatment was caries (52.1%); and final management of these patients was conscious in 96.4% of the cases. With regard to type of classification and horizon, 67.1% of the cases corresponded to classification A, with 35.9% of these in horizon I. With regard to diagnosis of different abilities, autism presented 40.2% of the cases in classification A and horizon I. A predictive equation of 97.1% probability of whether or not sedation was used was obtained. Regression analysis showed that this model has high sensitivity (100%) and low specificity (15%). Conclusions: Findings show that the type of management for dental care was conscious in nearly all the cases dealt with. This was significantly influenced by the Classification and Horizon Screening Protocol indicator, which established the best predictive model for whether or not to use sedation. The model adequately classifies patients who do not require sedation.