Avaliação da reprodutibilidade e validação do instrumento pronto para checagem de prescrições de antineoplásicos
Arquivos
Data
2023
Autores
Macedo, Rodrigo Spineli 
Orientadores
Santoro, Ilka Lopes
Tipo
Tese de doutorado
Título da Revista
ISSN da Revista
Título de Volume
Resumo
Objetivos: Avaliar a reprodutibilidade e realizar a validação do instrumento PRONTO,
checklist criado para auxiliar farmacêuticos especialistas com pouca experiência na revisão de
prescrições de quimioterapia. Métodos: Para a reprodutibilidade, foi realizado um estudo
transversal retrospectivo com 100 prescrições de quimioterapia de pacientes adultos em
tratamento de câncer. Estas prescrições foram checadas por dois farmacêuticos que utilizaram
um checklist baseado em 6 domínios (Patient, Regimen, Organ Function, Numbers, Toxicity,
Order Entry) e 30 questões. Na análise da reprodutibilidade intraobservador, o farmacêutico 1
realizou checagem com o instrumento e repetiu o processo depois de quinze dias. Para a análise
da reprodutibilidade interobservador, o farmacêutico 2 realizou checagem única das prescrições
com o instrumento. Após estes dois processos, foram realizadas a análises da reprodutibilidade
intra e interobservador a partir dos coeficientes de Correlação Intraclasse e Kappa. Para a
validação, foi realizado estudo randomizado incluindo 30 farmacêuticos com vivência mínima
de 1 ano em Oncologia. Na 1ª fase do estudo, os sujeitos foram alocados em dois grupos,
ANÁLISE LIVRE e CHECKLIST e realizaram análises de 10 prescrições – teste elaboradas
pelo autor e contendo 22 erros de medicação. Na 2ª fase, 22 dos 30 farmacêuticos participaram
e foram submetidos a um treinamento online sobre verificação de prescrições de quimioterapia
para analisarem 5 prescrições – teste elaboradas pelo autor e que continham 5 erros de
medicação. Foi realizada análise estatística descritiva e comparativa com testes de qui quadrado e t de Student. Resultados: A reprodutibilidade intraobservador foi excelente (CCI
>0,75; p<0,001) em 8 questões e regular a boa (0,40-0,75; p<0,001) em 14 questões. A
reprodutibilidade interobservador foi moderada a muito boa (Kappa >0,40-0,92; p<0,001) em
6 questões. Na FASE 1 da validação, os farmacêuticos do grupo CHECKLIST alcançou uma
taxa média de acertos de 35,60% (± 10,94) enquanto os farmacêuticos do grupo ANÁLISE
LIVRE alcançou uma taxa média de acertos de 37,27% (± 9,59). Na FASE 2, o grupo
CHECKLIST alcançou uma taxa média de acertos de 42,85% (±25,84) e o grupo ANÁLISE
LIVRE alcançou uma taxa média de acertos de 55% (±25,63). Conclusão: Na análise de
reprodutibilidade, o checklist PRONTO apresentou reprodutibilidade intraobservador
regular/boa a excelente na maioria dos domínios e baixa reprodutibilidade interobservador,
demonstrando diferenças na detecção de erros de medicação entre diferentes farmacêuticos. O
instrumento PRONTO se mostrou tão eficaz quanto a análise live na análise de prescrições.
Objective: To evaluate the reproducibility and validate the PRONTO (Patient, Regimen, Organ Function, Numbers, Toxicity, Order Verification) instrument, a checklist designed to assist young oncology pharmacists in reviewing chemotherapy orders. Methods: For reproducibility analysis, a retrospective cross-sectional study of one hundred chemotherapy orders of adult patients undergoing cancer treatment was reviewed by two pharmacists using a checklist consisting of 30 questions grouped into six domains (Patient, Regimen, Organ Function, Numbers, Toxicity, Order Verification). For intraobserver analysis, pharmacist #1 reviewed all prescriptions twice with a 15-day interval between reviews. For interobserver analysis, pharmacist #2 checked all prescriptions only once. Descriptive statistics, validation analysis, and randomized trials were conducted by 30 pharmacists with at least one year of experience in oncology. In the 1st phase of the study, the subjects were divided into two groups, FREE ANALYSIS, and CHECKLIST, and performed analyses of 10 prescriptions written by the author containing 22 medication errors. In the second phase, 22 of the 30 pharmacists participated and underwent online training on chemotherapy prescription review to analyze five prescriptions written by the author containing five medication errors. Descriptive statistics, chi squared tests, and t-Student tests were performed. Results: Intra-observer reproducibility was excellent (ICC > 0.75; p<0.001) in 8 questions and moderate to good (0.40-0.75; p<0.001) in 14 questions. Inter-observer reproducibility was moderate to very good (Kappa > 0.40-0.92; p<0.001) in 6 questions. In Phase 1 of validation, pharmacists in the CHECKLIST group achieved an average accuracy rate of 35.60% (±10.94), while pharmacists in the FREE ANALYSIS group achieved an average rate of 37.27% (±9.59). In Phase 2, the CHECKLIST group achieved an average rate of 42.85% (±25.84) and the FREE ANALYSIS group achieved an average accuracy rate of 55% (±25.63). Conclusion: In the analysis of reproducibility, the PRONTO checklist showed moderate to excellent intra-observer reproducibility in most domains and low inter-observer reproducibility, demonstrating differences in the detection of medication errors among different pharmacists. In validation, PRONTO checklist showed as effective as to free analysis of chemotherapy prescriptions in detecting medication.
Objective: To evaluate the reproducibility and validate the PRONTO (Patient, Regimen, Organ Function, Numbers, Toxicity, Order Verification) instrument, a checklist designed to assist young oncology pharmacists in reviewing chemotherapy orders. Methods: For reproducibility analysis, a retrospective cross-sectional study of one hundred chemotherapy orders of adult patients undergoing cancer treatment was reviewed by two pharmacists using a checklist consisting of 30 questions grouped into six domains (Patient, Regimen, Organ Function, Numbers, Toxicity, Order Verification). For intraobserver analysis, pharmacist #1 reviewed all prescriptions twice with a 15-day interval between reviews. For interobserver analysis, pharmacist #2 checked all prescriptions only once. Descriptive statistics, validation analysis, and randomized trials were conducted by 30 pharmacists with at least one year of experience in oncology. In the 1st phase of the study, the subjects were divided into two groups, FREE ANALYSIS, and CHECKLIST, and performed analyses of 10 prescriptions written by the author containing 22 medication errors. In the second phase, 22 of the 30 pharmacists participated and underwent online training on chemotherapy prescription review to analyze five prescriptions written by the author containing five medication errors. Descriptive statistics, chi squared tests, and t-Student tests were performed. Results: Intra-observer reproducibility was excellent (ICC > 0.75; p<0.001) in 8 questions and moderate to good (0.40-0.75; p<0.001) in 14 questions. Inter-observer reproducibility was moderate to very good (Kappa > 0.40-0.92; p<0.001) in 6 questions. In Phase 1 of validation, pharmacists in the CHECKLIST group achieved an average accuracy rate of 35.60% (±10.94), while pharmacists in the FREE ANALYSIS group achieved an average rate of 37.27% (±9.59). In Phase 2, the CHECKLIST group achieved an average rate of 42.85% (±25.84) and the FREE ANALYSIS group achieved an average accuracy rate of 55% (±25.63). Conclusion: In the analysis of reproducibility, the PRONTO checklist showed moderate to excellent intra-observer reproducibility in most domains and low inter-observer reproducibility, demonstrating differences in the detection of medication errors among different pharmacists. In validation, PRONTO checklist showed as effective as to free analysis of chemotherapy prescriptions in detecting medication.