SciELO - Scientific Electronic Library Online

 
vol.26 issue1The evidence of Bayesian A/B testing in the contrast of clinical events by COVID-19 author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Services on Demand

Journal

Article

Indicators

Related links

  • On index processCited by Google
  • Have no similar articlesSimilars in SciELO
  • On index processSimilars in Google

Share


Infectio

Print version ISSN 0123-9392

Infect. vol.26 no.1 Bogotá Jan./Mar. 2022  Epub Nov 13, 2021

https://doi.org/10.22354/in.v26i1.1004 

RESPUESTA A CARTA AL EDITOR

Reply to the letter: The evidence of Bayesian A/B testing in the contrast of clinical events by COVID-

Jeel Moya-Salazar1  2 

Hans Contreras Pulache2  3  * 

1 Escuela de Medicina, Facultad de Medicina. Universidad Privada Norbert Wiener. Lima, Perú. https://orcid.org/0000-0002-7357-4940

2 Hospital Nacional Docente Madre Niño San Bartolomé, Lima, Perú

3 https://orcid.org/0000-0003-2450-9349


Dear Editor:

We reviewed the comments made by Cristian Ramos-Vera with great interest, and we thank him for his comments. His comments will allow us to lead the reading from the “stan dard framework of frequentist statistics based on significan ce assumptions” towards Bayesian models and thus avoid dichotomous biases for future interests in the manuscript.

Our team has applied Bayes’ theorem in cervical cancer diag nostic studies1, however, as Bayesian methods are diverse, in this study we did not apply this analysis, which according to the analysis of the letter agrees with our findings.

We know that case-control studies are subject to the action of different biases, so in this study, we take the necessary precau tions to reduce and avoid them in the context of the Peruvian lockdown due to COVID-192. Our main objective was not to generalize their findings but to know the risk factors of the rural Andean population for COVID-19, however, we agree that the Bayesian analysis deepens in refining the statistical conclusions and provides better quality to the clinical response3.

Given that the p-value is not sufficient for clinical interpreta tion4, certain factors (such as effect size and sample size) can undermine the conclusions, resulting in large amounts of bias. One of our limitations was the sample size since in the context of confinement and social distancing it has been difficult for us to voluntarily enroll patients in the study. The applied Ba yesian analysis has highlighted this small sample size as the cause for the wide ranges for arterial hypertension and Diabe tes mellitus type 2 as risk factors. Further studies are required to evaluate on a large scale the risk factors for COVID-19 in the Andean and Amazonian populations, improving the Peru’s investment priorities in science and technology5.

As noted in the letter, the Bayesian approach can be very useful in SARS-CoV-2 research, improving the traditional sta tistical approach, thus, scientific paradigms are changing as is being seen in levels of evidence, levels of significance, and scientific writing 6-7. We believe that we have contributed to the discussion about the risk factors of the Andean Peruvian po pulation against COVID-19 in a perfect way.

References

1. Moya-Salazar J, Huarcaya J, Rojas-Zumaran V, Vazquéz D, Contreras- Pulache H. Quality and performance of Papanicolaou test using the Clinical and Laboratory Standards Institute (CLSI) EP12-A2 guidelines: a single-center study in Peru. J Cytol. 2021; in press.Links ]

2. Gobierno del Perú. Decreto Supremo N° 008-2020-SA, Decreto Supremo que declara en Emergencia Sanitaria a nivel nacional por el plazo de noventa (90) días calendario y dicta medidas de prevención y control del COVID-19. Diario El Peruano, 11 marzo 2020. Lima, 2020. Available on: https://cdn.www.gob.pe/uploads/document/file/1206594/DS_N__008- 2020-SA.pdf Access: 07/01/21 [ Links ]

3. Jeffreys H. Theory of probability. Oxford: Oxford University Press; 1961. [ Links ]

4. Solla F, Tran A, Bertoncelli D, Musoff C, Bertoncelli CM. Why a P-Value is Not Enough. Clin Spine Surg. 2018; 31(9):385-838. doi: 10.1097/BSD.0000000000000695. [ Links ]

5. Moya-Salazar J, Gomez-Saenz L, Cañari B, Contreras-Pulache H. Scientific research and innovation response to COVID-19 in Peru. F1000research 2021; 10: 399. https://doi.org/10.12688/f1000research.51400.1Links ]

6. Haynes RB. Of studies, syntheses, synopses, summaries, and systems: the “5S” evolution of information services for evidence-based healthcare decisions. Evid Based Med. 2006; 11(6):162-4. doi: 10.1136/ebm.11.6.162-a. [ Links ]

7. Marcus G, Davis E. GPT-3, Bloviator: OpenAI’s language generator has no idea what it’s talking about. MIT Tech Rev. 2020. Available on: Available on: https://www.technologyreview.com/2020/08/22/1007539/gpt3-openai-language-generator-artificial-intelligence-ai- opinion Access: 07/01/21 [ Links ]

Cómo citar este artículo: J. Moya-Salazar, et al. Reply to the letter: The evidence of Bayesian A/B testing in the contrast of clinical events by COVID-19. Infectio 2022; 26(1): 101

There was no financing. No conflict of interest

Received: July 26, 2021; Accepted: July 26, 2021

* Autor para correspondencia: hans.contreras@uwiener.edu.pe Av. Arequipa 440, Oficina 801, Escuela de Medicina Humana, Facultad de Ciencias de la Salud, Universidad Privada Norbert Wiener, Lima 51001, Perú. Teléfono: +511 959736855

Creative Commons License This is an open-access article distributed under the terms of the Creative Commons Attribution License