Reviewer Number: 1

Reviewer number: 1

1. Major Compulsory Revisions

None

1. Minor Essential Revisions
2. There are a few minor grammar errors that I invite the authors to correct. I marked such errors on my file, which I cannot upload with my review report but would be available to send to the authors upon their request.

We have thoroughly revised the text with help of persons well familiar in English language.

2.2. Data collection: please give more details on participants’ recruitment. For instance, how were suicide attempt cases “recognized by medical staff” and what strategy was used to ask participation in the reviews? Where were they interviewed, and how? Authors should also clarify if the WHO scale used has been validated in Vietnam. If not, this should be stated as a study limitation.

Under “Data collection”, we have clarified the methods for data collection, in particular regarding individuals who were not hospitalized. We note as a limitation in the “Discussion” that the instrument has previously been used in Vietnam but not rigorously validated.

2.3. Incidence and methods used of suicide attempt: how many of 104 individuals interviewed were from the hospital records and how many were recognized by medical staff (and not in the records)? This information could give an indication of the extent of the under-reporting of suicide attempts.

90 persons were identified though hospital records, and 14 identified by medical staff after treatment at home. We have added this information in the beginning of “Results”

2.4. Results: in this section, research findings are sometimes compared to the available literature. This is what the discussion is for so I suggest moving these parts in the discussion.

While lacking an appropriate control group re socioeconomic conditions, we have compared with data from a similar rural area. We added an explanation of this under “Statistical analyses” and therefore consider it as part of “Results”

2.5. Use of health care services: First of all, I feel that the researchers have missed a very good opportunity to explore why 44.4% of the cases refused professional care. The researchers have chosen to use a WHO scale that is generally use in emergency settings or in psychiatric wards and has merely epidemiological purposes. They interviewed participants time after the attempt, thus not in a crisis situation and had, therefore, the chance to explore the important issue of access to mental health care (which was one of the aims of the study) more in depth. I wish they had asked participants why they refused professional help. We already know that many people who attempt suicide do not want professional help: now we need to understand the reasons behind this and what should be improved to give them the help they need. I feel the researchers missed the opportunity to explore this issue in Vietnam. The only data given are percentages and this needs a clarification: 44.4% of refusal to referral and 35.4% of acceptance do not add up to 100%: write that 22.2% were not sure if they accepted or had no referral.

We agree it would have been valuable to explore reasons for not attending professional care, and hope to be able to address this in future studies. We have added the missing information on the 22.2% under “Use of health services”.

2.6. Conclusions: This part should give some suggestions for future studies (e.g. see comment 3.2) and implications for suicide prevention. For instance, the authors stated that the suicide attempts might be due to psychological difficulties. They could suggest some prevention strategies that might address such difficulties. As a final comment, the conclusion emphasizes means control as a prevention strategies. Please do not forget to mention that, although it is a good prevention strategy, we have evidence that it is only partially effective and people turn towards other methods. This “means swapping” is currently being shown in Sri Lanka. Thus, means control might be effective in preventive a portion of impulsive suicides but can only be a piece of any suicide prevention program.

We have added suggestions for additional strengthening of mental health services in the community, better knowledge about mental health problems among community health workers, policy to reduce stigma associated with mental illness at the end of the “Discussion”. We have added the point about possibility of “swapping” between methods, and have taken the liberty to use the wordings of the reviewers, that we found excellent.

2.7. References: please make sure that the references match BMC referencing style (e.g. the use of et al, Nguyen’s article contains a mistake)

We are sorry the references had not been edited properly, this has now been done.

2.8. Table 2: please indicate what’s the number and what’s the percentage N (%).

We have modified the table as suggested

1. Discretionary Revisions

3.1. I suggest adding “self-harm” among the keywords and refrain from using the

expression “commit” (a suicide attempt/suicide), which has legal and moralistic

connotations.

We have modified as suggested

3.2. Discussion: Perhaps it would have been more interesting comparing the data

from Vietnam with data from other Asian countries rather than Canada, Norway

and Nigeria (e.g. Thailand, that is also a Buddhist country?). An interesting data

from the study is that, while international (i.e. Westerners/high income countries)

literature shows that living alone is a risk factor for suicide attempt (see

Introduction), only 3% of the sample lived alone. This is an important data, which

should receive some attention from the authors. This kind of knowledge help us

to question the cross-cultural applicability of risk factors models, which have

been in great part based on white middle-high class populations (mainly from US

and UK studies). I totally agree with what the authors wrote about suicide attempt

be due not only to mental health issues and I am glad the authors stated this.

The authors go even further than this and develop the hypothesis that this view

might be one of the reasons why Vietnamese people who attempt suicide so

often reject professional care. As I already stated, the researchers missed the

opportunity to ask this question themselves. However, they should write this as a

suggestion for future studies in the conclusions.

We have added the last point as suggested, and will carry other suggestions forward for future studies

Reviewer number: 2

1. The language at times is a little convoluted and would benefit from English editing. For example, it is not usual to refer to “committed suicide attempts”, and the simple use of the words “attempted suicide” is preferable.

See above, we have revised the text re grammar and style.

1. The data collection is a little unclear to me. If it is a retrospective case note survey, I am not sure who all those who had attempted suicide were interviewed face-to-face by trained medical staff. That is particularly the case for those who were reported to have “stayed at home”. This needs clarifications.

See above, we have clarified the methods for data collection.

1. The brief sentence about approval on medical ethical grounds is also unclear and needs rewording.

We have reworded this sentence using the correct wording by Hanoi Medical University.

1. Other examples where the language is a little unclear and needs revision include the use of the term “majority” in regard to those who have had somatic care, when in fact only about a third of persons had had such care; the “high case fatality associated with pesticides” would usually be referred to as “high lethality associated with pesticides”; and the words “easier available” should be “more easily available”. There are other examples, particularly late in the discussion where at times it is a little difficult to understand what is meant when reference to “manual farming and temporary works during leisure period” were referred to, but these are all relatively minor points which could be clarified by the careful attention of English speaking assistant.

Thank you for pointing out these inadequacies. We have worked on the language and in particular the examples given by the reviewer.

1. The authors are aware of the limitations of the study and the paper is well referenced. However, in Table 1 it is not clear what “rigid subjects” refers to, and it could be argued both in the Tables and also in the text that the term of “moral” support probably means general emotional support, as the word “moral” has probably unwarranted overtones to its generally accepted meaning in the current context.

Again, thank you for alerting us. We have changed the wordings: The category in the table should be “Using blunt object” and “Moral support” has been changed to “emotional support”. Actually, while “Moral support” is the term used in the English version of the instrument, referring to Bille-Brahe et al, in the paper by Bille-Brahe et al (Arch SuicideRes 1999;5:215-231) based on this instrument, they consistently use the term “Emotional support”. We chose to follow the terminology used by Bille-Brahe et al.