An Exploratory Study on Information Manipulation by Doctors: Awareness, Actual State, and Ethical Tolerance
and Akira Akabayashi 3,7,*Round 1
Reviewer 1 Report
My conclusion will be to reconsider after major revisions, see the following comments Novelty of the study. What does the study add to knowledge in the area. How this study will contribute science on information manipulation is not elucidated in the introduction. There is no hypothesis presented in the introduction Presenting the Methods with subtitles as Study Design, Study population, Variables, Data collection, Data analysis, Statistical analysis would add clarity to the study. Also line 99 " at the time of informed consent" would benefit from some definition. Discussion The first paragraph should state the following: "This is what we have done, this is what we have found". Line 186 - 190 could be improved in this sense. However, stating that "analysis of their response suggested that ... line 190 - 192 is not obvious from the study itself . Line 196 Do nurses have the same viewpoint? Is not studied at all in this paper and cannot be mentioned as such. Line 213-214 in the discussion section: you are allowed to speculate. Sex differences: It would add to this trial if they compared with other similar studies. Above all they have not discussed what a small proportion of participants (< 30 %) actually means to the study. Dropout analysis should be performed? The authors draw conclusion both in the abstract and under Conclusions they have not studied. Line 33 - 38 information manipulation is not culturally dependent on cultural settings; trust between patients and medical personnel is crucial and is not needed when such a relation has been established.
Author Response
Reply to the Reviewer 1
My conclusion will be to reconsider after major revisions, see the following comments.
Novelty of the study. What does the study add to knowledge in the area. How this study will contribute science on information manipulation is not elucidated in the introduction.
Novelty of the study:
⇒The following are the strongest original elements of this physician’s data: 1) this was a large-scale study (N = 3,305) and 2) physicians from all specialties were examined (this study was not restricted to oncologists, such as in the 2008 Journal of Clinical Oncology study).
What does the study add to knowledge in the area:
⇒Our data has shown that, similar to the USA, Japan also has the problem of information manipulation in truth-telling and physicians have discretion over their patients’ information. We picked Japan as an example as this study implies that information manipulation by physicians may also be an essential issue in other countries.
How this study will contribute science on information manipulation is not elucidated in the introduction:
⇒As explained in the introduction, this is a preliminary fact-finding and awareness study. Although some studies show that patients are eager to know their prognosis, a study by Miyata et al. (2004) revealed that some patients want physicians and their families to decide. Why? Moreover, regarding the physicians’ perspective, we have shown that their real practice and awareness does not necessarily match. Furthermore, there are different types of practice and awareness among the related factors such as age or gender. Hence, these need to be examined further. Finally, our study has revealed many aspects requiring examination. Moreover, we have proposed a hypothesis that should be tested further. These are some of the significant contributions to this area.
There is no hypothesis presented in the introduction.
⇒Yes, this is correct. In fact, we asked physicians about their practice and awareness as this was a fact-finding and awareness survey. Thus, this study was exploratory, intended to demonstrate facts and awareness. Finally, we formulated our hypothesis. Our aim for the next study is to verify the hypothesis.
The first sentence of the Abstract: To what extent is information manipulation by doctors acceptable? To answer this question, we conducted an exploratory study aimed at obtaining basic data on descriptive ethics.
Presenting the Methods with subtitles as Study Design, Study population, Variables, Data collection, Data analysis, Statistical analysis would add clarity to the study.
⇒According to your suggestions, we have rewritten the Methods section as follows:
- Methods
Study Design：Anonymous self-administered questionnaire. In February 2014, we conducted an anonymous self-administered questionnaire survey by mail.
Study population: Members of the Japan Medical Association and doctors responsible for clinical training at basic clinical training hospitals and university hospitals throughout Japan. A total of 2,291 doctors representing all specialties were chosen by randomly selecting 2.5% and 10% of the members of the Japan Medical Association whose primary medical background was internal medicine and surgery, respectively. We also included doctors responsible for clinical training at all basic clinical training hospitals and university hospitals (total: N = 3.305).
Variables: The questionnaire used the item, “Regarding explanation by manipulating information with good intentions” (hereinafter, “explanation with manipulation”). This was done with reference to the explanation provided while obtaining informed consent at the time of medical treatment (hereinafter, “explaining informed consent”). Concerning “explanation with manipulation,” it was first defined as giving a false explanation (e.g., describing a malignant disease as a benign disease to a patient to eliminate their anxiety); an understated explanation (e.g., understating the degree of side effects to a patient to encourage continued medication); or an exaggerated explanation (e.g., overstating the degree of exacerbation of a disease to a patient to encourage medical attention). Next, the respondents were asked whether “explanation with manipulation is necessary in some cases” with the two possible answers of “Agree” or “Disagree.” Furthermore, the participants were asked whether they “have provided an explanation with manipulation,” with two possible answers of “Yes” or “No”; and whether “explanation with manipulation is ethically acceptable,” with two possible answers of “Agree” or “Disagree.” Additionally, they were asked whether they “request a nurse to be present at the time of explaining informed consent” (hereinafter, “the presence of a nurse”), with two possible answers of “Yes” or “No.” Finally, the doctors were asked to provide information on their attributes, including sex, age, primary specialty, and their respective medical institutions.
Data collection: Online via e-mail .
Data analysis and Statistical: For the analysis, the three items related to “explanation with manipulation” and “the presence of a nurse” were grouped by attributes of the doctors, their place of medical employment, their primary specialty, and the number of hospital beds. We examined whether there were differences between the groups concerning awareness, action, and ethical views. We then examined whether there were differences in the proportion of those who request “the presence of a nurse,” depending on the awareness and action regarding “explanation with manipulation.” Lastly, the association between the age of the doctors and the number of hospital beds in their respective medical institutions was examined. The χ2 test was used for all statistical tests.
　Statistical analysis was performed using SAS version 9.4 software (SAS Institute Inc., Cary, NC, USA), and p < 0.05 was considered statistically significant.
Also line 99 “ at the time of informed consent” would benefit from some definition.
⇒Thank you very much for this suggestion. In this regard, we have added the following: “at the time of informed consent, when the attending physician explains the diagnosis and treatment options with the physician recommendation, inviting patient’s (and family’s) questions, and after enough conversation, when the patient has decided to choose one treatment option, then the patient signs the document of agreement with the treatment option. When the patient cannot decide at that point, the physician stops the process and gives more time to the patient unless the case in question is an emergency.”
Discussion The first paragraph should state the following:
"This is what we have done, this is what we have found".
⇒Thank you very much for the recommendation. This has been added.
Line 186 - 190 could be improved in this sense. However, stating that "analysis of their response suggested that ...
⇒Thank you very much for the recommendation. We have added following statements.
Analysis of their responses such as:
Although 98% said their usual practice is to tell terminally ill patients that they will die, 48% specifically described communicating terminal prognoses to patients only when specific preferences for prognosis information were expressed. Forty-three percent said they always or usually communicate a medical estimate of time as to when death is likely to occur, and 57% reported sometimes, rarely, or never giving a time frame [1 from abstract].
Amongst respondents who wanted full disclosure about the diagnosis without delay, 117 (69.6 %) respondents agreed to follow the doctor's discretion, whilst 111 (66.1 %) respondents agreed to follow the family member's decision. For respondents who preferred to have the diagnosis and prognosis withheld, 59 (26.5 %) agreed to follow the doctor's decision, and 79 (35.3 %) of respondents agreed with following family member's wishes [16 from abstract].
These findings suggested that physicians hesitate to tell the diagnosis to patients in the US [1]. Moreover, patients do not want to decide their treatment options by themselves, rather entrusting physicians or family—at least in Japan [16].
line 190 - 192 is not obvious from the study itself .
⇒You are absolutely right. Hence, we have changed the sentence as follows:
We speculate that the analysis of their responses may indicate …
Line 196 Do nurses have the same viewpoint? Is not studied at all in this paper and cannot be mentioned as such.
⇒Thank you very much, and you are absolutely correct with regard to this matter. To assert the nurses’ positions, we need other studies—both quantitative and qualitative—which target nurses. Accordingly, we have changed the expression as follows:
We speculate that this result may be interpreted as follows.
Line 213-214 in the discussion section: you are allowed to speculate. Sex differences: It would add to this trial if they compared with other similar studies.
⇒To our knowledge, there are no—or few that we cannot find—papers that study sex differences. Even in the classic US oncologist study [1], which features female participants being placed into the multivariate analysis, sex was not significant.
As shown in Table 3, results from multivariate analysis reveal that older age (OR _ 0.76; 95% CI, 0.64 to 0.92; P _ .004) and being Jewish (OR _ 0.59; 95% CI, 0.38 to 0.91; P _ .018) were associated with being less likely to report communicating to patients a medical estimate, or time frame, as to when death would occur. (JCO, 2008, p. 5890)
⇒We want to take this matter seriously. Given this, we need more studies, such as a large-scale quantitative study for female doctors, in-depth quantitative studies, etc. We hope you understand our position.
Above all they have not discussed what a small proportion of participants (< 30 %) actually means to the study.
⇒This point is discussed in the Limitations section (number 1).
- Low response rate (however, compared to previous studies, the present study is a large-scale sampling and is superior as it targets not only oncologists but also doctors from all medical departments).
Regarding this study, which could be considered large-scale, a response rate of 30% is acceptable in a sociological survey, as long as the low response rates are described in the Limitations section. Please refer to several other large-scale surveys in the sociology field.
Dropout analysis should be performed?
⇒As this was an exclusively online study, we could not analyze those who did not answer the anonymous survey. Therefore, in this study design, a dropout analysis was impossible.
The authors draw conclusion both in the abstract and under Conclusions they have not studied. Line 33 - 38 information manipulation is not culturally dependent on cultural settings; trust between patients and medical personnel is crucial and is not needed when such a relation has been established.
⇒Yes, you are absolutely right. We have deleted the sentence about cultural settings.
Deleted: … indicating that information manipulation by doctors is not culturally dependent.
However, just stating our hypothesis may be allowed, as this is the hypothesis we discovered while analyzing this study’s data. Of course, we have not confirmed the hypothesis; it needs further review to examine whether it is true or not. I hope you will be satisfied with our explanation. If you wish to not accept it, we will delete the hypothesis.
We hope these changes are acceptable for you.
Once again, thank you very much for your valuable comments on our article.
Reviewer 2 Report
At first, the authors should present their definition of information manipulation. Did they assume it as lying or not to disclose the truth? These two are far different issues related to information manipulation.
More information about the questionnaire is needed. Did the researcher develop it? How? It's content? How they evaluated its relevancy?
Why is there an 8-year gap between the study conduction and paper submission?
]It is a little bit confusing, because the informed consent is different from truth-telling. Why the physician should ask a nurse at the time of explaining the information for informed consent?
Is there a role for number of beds in truth-telling?
Author Response
Replay to the Reviewer 2
At first, the authors should present their definition of information manipulation. Did they assume it as lying or not to disclose the truth? These two are far different issues related to information manipulation.
⇒We define information manipulation as “not disclosing the truth at the discretion of the physician for the benefit of the patient.” Thus, ethically speaking, it is an action based on the principle of beneficence and non-maleficence.
More information about the questionnaire is needed. Did the researcher develop it? How? It's content? How they evaluated its relevancy?
⇒The seven researchers, a group which included physicians, nurses, legal scholars, and ethicists, discussed and created the questionnaire based on previous studies. In particular, as previous studies have shown that manipulation of information by healthcare professionals is a problem (e.g., 2008 Journal of Clinical Oncology [JCO]), the questions were designed to focus on information manipulation. First, to investigate the current situation in Japan, we conducted a highly representative sampling and asked a large number of physicians (N = 3,305) about their awareness and actual practice. Prior to implementation, 10 physicians were pretested and asked their opinions; subsequently, feedback was received, which stated that the questions were easy to answer and important. Specifically, the four short questions can be answered in less than five minutes.
Why is there an 8-year gap between the study conduction and paper submission?
⇒We published a letter in the Lancet in 1999 (one university hospital, physicians in all departments [N = 74]. Further, a 2008 JCO article (only oncologists in the US [N = 1,137] encouraged us to take a large grant in Japan and conduct a larger survey with more representative samples than the JCO paper in Japan (including physicians in all departments [N = 3,305]). As you can see, the data was quite interesting, showing that it is not only the US that is in trouble, rather it is an international phenomenon. At that time, we planned to conduct a follow-up survey in 2024, 10 years from the 2014 study. In the meantime, our research activities had to be temporarily suspended due to the COVID-19 pandemic, and when we reached the stage of applying for a grant for the 2024 project, we were asked to submit a paper in English for the 2014 survey. As a baseline survey, it is desperately needed for the next grant application. Notably, the topic of this paper is far superior to previous research because of the following: (1) it is a large-scale survey; and (2) the sampling is representative of physicians in all departments in Japan, although some small surveys have been conducted outside Japan. Seeing how it changes after 10 years will be academically useful. If the 2024 study is successful, we would like to submit it to your journal as soon as possible.
Akabayashi, A.; Kai, I.; Takemura, H.; Okazaki, H. Truth telling in the case of a pessimistic diagnosis in Japan. The Lancet. 1999; 354:1263.
Daugherty, C.K.; Hlubocky, F.J. What are terminally ill cancer patients told about their expected deaths? A study of cancer physicians’ self-report of prognosis disclosure. Journal of Clinical Oncology. 2008;26:5988-5993.
It is a little bit confusing, because the informed consent is different from truth-telling. Why the physician should ask a nurse at the time of explaining the information for informed consent?
⇒This is part of working as a team in the medical field. The nurse's presence relieves pressure on the parents and family, and they can fill in where there are difficulties in the physician's explanation. They can also communicate questions from the patient in a way that is easier to understand than the physician. In many hospitals in Japan, the head nurse is present at the explanation to the patient and family.
Is there a role for number of beds in truth-telling?
⇒Hospitals with a large number of beds are also teaching hospitals, because they have to teach residents and young doctors how to explain and obtain consent.
In conclusion, we hope these changes and replies are acceptable for you.
Once again, thank you very much for your valuable comments on our article.
Round 2
Reviewer 1 Report
my comments in the Discussion section typed in red should be deleted
Author Response
Reviewer 1 Comments and Suggestions for Authors
my comments in the Discussion section typed in red should be deleted
Yes, we did.
Once again, thank you for reading our paper in detail.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The main problem of the paper was the questionnaire. Questionnaire development and standardization is not an easy, simple method as the authors explained. It needs more clarifications about questionnaire development, how they evaluated its validity and reliability. How many items they generated at first and the number of items at the end.
Author Response
Reviewer 2 Comments and Suggestions for Authors
The main problem of the paper was the questionnaire. Questionnaire development and standardization is not an easy, simple method as the authors explained. It needs more clarifications about questionnaire development, how they evaluated its validity and reliability. How many items they generated at first and the number of items at the end.
We have asked a professor of Sociology, an empirical survey expert. His answer was:
This is an exploratory fact-finding and awareness survey. It is not a scale development such as psychological questionnaires. Standardization, evaluation its validity and reliability are not necessary, rather it is impossible. So, your description of the methods is enough.
We agree with what the professor said. Would you also ask your colleague who are experts of social survey? So, we would like to leave the description as it is. I hope you will be satisfied with this answer.
Once again, thank you for reading our paper in detail.
Author Response File: Author Response.docx
