Describing and understanding patient safety incidents in primary care dentistry and building consensus on never events
Ensaldo Carrasco, Eduardo
MetadataShow full item record
Introduction: In recent decades, there has been considerable international attention directed towards minimising healthcare-associated harm and improving the safety of hospital care. More recently, this attention has broadened to include primary medical care. In 2002, the World Health Assembly recognised the issue of inadequate levels of patient safety as a major threat to global public health. In the following years, many countries have developed national strategies for the measurement, monitoring and prevention of patient safety incidents (PSIs) and their outcomes. Experience accumulated from secondary care has shown that the initial steps for understanding patient safety include the systematic identification of the most frequent and most harmful threats. However, the safety profile of primary care dentistry remains poorly investigated. As a result, current evidence cannot provide reliable estimates of the types of PSIs in primary care dentistry, the causes of these incidents, or the associated disease burden caused by such incidents. In medicine, improvements in patient safety were achieved at a national level by developing a shared conceptual understanding, the standardisation of terminology and through preventive initiatives such as the introduction of a national incident reporting and learning system. In the United Kingdom (UK), the England and Wales’ National Reporting Learning System (NRLS) has been an important source of insight, from the perspectives of the reporter, into understanding why PSIs occur. This initiative has led to the implementation of patient safety oriented policies to monitor and reduce cases of healthcare-associated harm. Examples of such policy initiatives include national guidelines and national safety recommendations to encourage the reporting of serious reportable events called ‘never events’ (NEs). These are defined as serious, preventable PSIs that should not occur if the available preventive measures are implemented. At a national level, serious incidents and NEs must be reported to the NRLS and/or other reporting systems. However, little is known about NEs in dentistry as wrong-tooth extractions are the only currently defined NE that has a clear application in dentistry. Although surgical NEs, such as wrong-site surgery and wrong implants may be related to dental procedures, these overlap with procedures conducted in secondary care. As a result, there is no agreed list of NEs for primary care dentistry. The overall aim of my PhD was to explore patient safety, its concepts, including error and harm, and how these can help to create an understanding of the types of PSIs that occur in primary care dentistry, their contributory factors and their consequences. In addition, I also aimed to identify NEs with the greatest need and opportunity for future intervention strategies, in order to improve patient safety in primary care dentistry. Methodology and methods: My PhD was conducted in three phases. For the first phase, I conducted a systematic scoping review of the empirical evidence published over a 20-year period (1994-2014). To achieve this, I searched MEDLINE and EMBASE for articles reporting incidents that could have or did result in unnecessary harm from primary dental care. I also extracted and synthesised data on the types and frequencies of PSIs (including NEs) and adverse outcomes. Then, for the second phase, I undertook an exploratory sequential mixed-methods evaluation, which involved the qualitative exploration and analysis of a weighted-by-year randomised sample (n=2,000) of the most severe incident reports from primary care dentistry submitted to the England and Wales’ NRLS. This approach generated three coding frameworks, aligned to the International Classification for Patient Safety developed by the World Health Organization, for i) the classification of incidents, ii) contributor y factors and iii) incident outcomes. These coding frameworks informed the quantitative analysis, during which myself together with a trained second coder, applied codes to deconstruct the narrative of these patient safety incident reports whilst retaining the meaning of the report. To assess inter-rater reliability, Cohen’s Kappa statistic was calculated for the primary incident type which was defined as “the incident that resulted in the outcome experienced by the patient.” Finally, for the third phase, I undertook an electronic Delphi exercise to achieve international agreement on NEs for primary care dentistry. The results obtained from Phases 1 and 2 were used to identify candidate NEs. I then invited an international panel of 41 experts to complete two rounds of questionnaires; 32 (78%) agreed to participate and completed the first round, and 29 (91%) completed the second round. I provided anonymised controlled feedback between rounds and used a cut-off of 80% agreement to define consensus. The results from the first stage built the evidence base for the second and third phases. Likewise, the results from the second phase further informed the third and final stage of my PhD. Results: I undertook a systematic scoping review which demonstrated: a) there were considerable differences in definitions for terms used to describe patient safety, b) that a range of populations had been studied, and c) that major differences in sampling strategies exist between studies. The main five PSIs I identified were errors in i) diagnosis/examination, ii) treatment planning, iii) communication, iv) procedural errors and v) the accidental ingestion or inhalation of foreign objects. However, little attention has been paid to wider organisational factors such as problems within the physical environment, scheduling (e.g. errors in managing appointments) and patient access, management and lines of responsibility. Also there is very little evidence of interest in researching into the influence of policies for either quality or patient safety assurance. The retrieved evidence was used to build a conceptual literature-derived model of patient safety risks in primary care dentistry. This model helped to bring structure to the analysis of the 1,456 patient incident reports that were eligible for analysis out of a total of 2,000. These reports described incidents across the preoperative (40.3%; n=587), intra-operative (56.1%; n=817) and post-operative (3.6%; n=52) clinical stages of care delivery. Further analysis showed the more frequently reported incidents were related to a) delays in treatment (333/1,456; 22.9%), b) procedural errors (220/11,456; 15.1%), c) medication-related adverse incidents (160/1,456; 11.0%), d) equipment failure (90/1,456; 6.2%) and e) errors in obtaining or processing x-rays (87/1,1456; 6.0%). Only 5.3% (77/1,456) of the incidents resulted in harmful outcomes. Of the 77 incidents that resulted in a harmful outcomes (n=77; 5.3%), around half were due to wrong tooth extractions (37/77; 48.1%) and resulted in unnecessary procedures. Three out of the 1,456 incidents (0.2%) resulted in death. Data from the scoping review and the mixed-method analysis informed a list of 42 candidate NEs. I further sought and achieved international consensus for 23 of these NEs. These were related to routine assessment, and pre-operative, intra-operative and post-operative stages of dental procedures. Conclusions: The findings from my PhD have revealed that patient safety research in dentistry is mostly descriptive and poorly organised with various approaches to defining and measuring PSIs and their outcomes. This poor organisation of patient safety research also includes differing study designs and patient populations studied. The evidence-based conceptual framework from the systematic scoping review, and coding frameworks from analysis of PSI reports selected from a national database, can bring structure to future work by providing a robust approach to classifying PSIs, their contributory factors and outcomes. My research findings also show that PSI reports are an important source of information that can generate important insights about patient safety in primary care dentistry. The mixed-method analysis of PSI reports showed that most incidents in primary dental care do not result in harm. PSIs that resulted in harmful outcomes more frequently occurred intra-operatively. My findings also reveal that unsafe care in dentistry is not limited to human error, but can also be ascribed to the presence of other administrative or organisational flaws that contribute to the reported incidents. Future initiatives to improve and research clinical practice should focus on improving administrative processes to reduce delays in treatment. Also, the reduction of procedural errors through the standardisation of x-rays, medication prescription and other clinical procedures is needed. Lastly, I have constructed the first comprehensive international list of NEs for primary care dentistry. I believe my findings, including the list of NEs, can provide an evidence-base which will encourage researchers to further expand the patient safety research and development agenda in dentistry, as well as encouraging decision-makers and professional bodies to translate my findings into quality improvement strategies.