Review discussion of Population Survey of the reported ownership of Asthma Action Plans

Index of discussion

1. Review by Elizabeth Comino 30.6.99
2. Review by Nicholas Saunders 1.7.99
3. Review by Christine Jenkins 13.7.99
4. Review by Sally Reagan 21.7.99
5. Editorial comments (requesting revision) by Ruth Armstrong 28.7.99
6. Author's response by R Ruffin 5.8.99
7. Author's response by R Ruffin accompanying revised version 25.8.99
8. Editorial response and decision by Ruth Armstrong 26.8.99

DateComment
Wed Jun 30 15:39:54 1999 1.

I think that this is a very good paper, which presents some important population-based information on asthma in the community.

This paper reports data collected during a multistage area sample of adults aged 15 years and over. The study provided population-based evidence of the prevalence of asthma and on the use of asthma management plans. Management issues were compared for subjects with asthma who used or did not use an asthma management plan.

The study used area based methods based on the use of the Australian Bureau of Statistics Collector Districts and is recognised as a reliable means of identifying a population-based sample. Weighting for age gender and geographic area standardised the results to the South Australian Population. These methods are well recognised and are widely used in studies of this type. The diagnostic criteria were adequate for a field study of this type. The questions used have been used in previous studies and have good face validity.

The results are presented in a concise fashion for a study of this type. The two tables that summarise these results are generally clearly presented and provide a useful overview of the results.

The discussion provides a useful synthesis of the results and does not attempt to over-interpret the findings of the study. The concluding correctly identifies the need to a new look at innovative ways of improving the quality of education about asthma in the general practice setting. Further research in this area is badly needed.

I would recommend this paper for publication



Elizabeth Comino

Thu Jul 1 14:11:35 1999This paper reports the results of a carefully conducted survey of the South Australian population about the possession of a written action plan among current asthmatics. The picture painted is bleak with regard to the adoption of best practice guidelines promulgated by the National Asthma Campaign: ownership of a written plan is reported by only one in three asthmatics, down from 42% reported just a few years earlier. Asthmatics with poorer control (as judged by nocturnal symptoms) reported more frequent ownership of an action plan than those who did not report frequent nocturnal awakening; however, such patients were significantly less likely to report ease of access to medical care or comfort with such care!

The survey method employed by the investigators is sound and has been previously validated. The number of asthmatics sampled is relatively small (approx. 350) but sufficient to explore the main hypotheses. It would be helpful if the authors provided more details about the asthmatics in the sample so that the reader can get a more complete picture of the profile of the asthmatics assessed. For example, frequent nighttime awakening was reported by only 15.2%. Does this reflect "mild asthma" in the remainder or good control by medication use despite the low prevalence of asthma plans?

The title of the manuscript refers to the "use" of action plans whereas the text more correctly refers to "ownership". This survey did not address the issue of actions taken by those in possession of an action plan (ie. use or effectiveness of the plan) nor whether a plan had been prescribed but ignored (ie. not reported) by the asthmatic. Perhaps the title of the manuscript should be amended to make this point clear.

Similarly, the title includes the phrase "Where to from here?" and the Discussion spends some time on "The way forward". Yet the article does not provide clear advice about what needs to be done now. Most of the discussion focuses on the effectiveness of education to change patient behaviour, but what about the two thirds of patients who did not report possession of a plan? Had they ignored their plan, allowing it to lapse, or had the medical practitioners caring for the majority of asthmatics not provided an action plan in the first place? I think the discussion should be broadened to include mention of adoption of best practise guidelines by the profession and promulgation of action plans that are known to be effective when implemented appropriately.

I found the tables difficult to read at first glance but this may simply be a function of pulling the manuscript off the web site. I also had some difficulty in reconciling the text with Table 1. For example, last sentence, para 3, p.6 implies that 13.4% of asthmatics without nocturnal asthma possessed an action plan. But is this the case? I think the table shows that 13.4% of asthmatics without an action plan awoke at night compared with 19.8% of those with an action plan; a different conclusion.

There are numerous points in the manuscript where grammar and expression could be improved, including sentences that appear incomplete (eg. third sentence para 3, p.5 commencing line 29; and seventh sentence, same paragraph commencing line 38).

This article is an important contribution to improving the care of asthmatics in Australia. I recommend its publication after revision.


N.A. Saunders

Tue Jul 13 02:01:42 1999This paper is a valuable addition to the ongoing evaluation of the implementation of the Australian 6 step Asthma Management Plan. In assessing the relationship between possession of a written action plan and morbidity, perception/comfort associated with managing asthma and adequacy of information provided about asthma, the study advances our current knowledge.

The methodology of the study is reliable and has been used when asthma data from previous SAHOS's have been reported by the authors. The questions on morbidity have been used before and their validity has been assessed. I am uncertain about the reproducibility and validity of the questions on perception. It would be helpful to have all the questions provided by the authors in a table. I have the impression that the questions as listed in par 21 have been paraphrased /shortened for the sake of the flow in the text. Is this the case? It would help readers to assess the validity of the authors' conclusions if they could read the questions in full. This is fairly standard practice in questionnaire reporting, especially when there are new questions.

I find it difficult to understand Table 1. The legend needs to spell out more clearly the relationship between the OR's and the headings in the 2 columns. Perhaps an example beneath the table would help, or in the text. For example, in par 25 it is stated that ownership of an action plan was more likely to be associated with taking inhaled steroids - but interpreting this from Table 1, it could be that it was 2.56 times more likely that people with an action plan did not take inhaled steroids. Somehow interpretation of this table needs to be made easier.

I am unsure of the difference in the observations in par 25 and par 27.

In the Discussion, the authors comment that there has been a drop in the ownership of written action plans, compared to the SAHOS in 1995, from 42% to 33%. They refer to the paper in ANZJM,1997. In another paper (MJA 1997), reporting the same study, 9-12% had a written asthma management plan (cf 46% total, the remainder being verbal), and 39.5% had an action plan (not specified in the text if written). Perhaps the authors could make some comments on these differences, and in particular whether the differences in possession of a WRITTEN action plan are likely to be significant.

Also in the Discussion, there is scope to discuss the underlying problem further - if action plan possession is down, is this a failure of implementation by health professionals, especially doctors? Par 32 implies this, but mention could be made of the recent review of step 6 (educate and review regularly) published in the 1998 asthma Management Handbook and on the Cochrane Collaboration Website. If so, the issues of effective uptake of guidelines by doctors deserves some discussion. In par 33, some of the problems of adoption of good management concepts by the patient are discussed. It would be appropriate to refer to the NAC's recently published Adherence Guide for Health Professionals - an acknowledgement that this end of te therapeutic relationship can be improved by the adoption of practical skills which enhance communication and effectiveness of advice.

Could action plans be a marker for something else - eg recognition by the doctor of an "at risk" person? This is suggested by the description of this "person" in par 25 - ie some features of suboptimally controlled asthma. Perhaps one problem is failure to achieve good asthma control.

There are some grammatical errors, especially in par 32.

This is a valuable paper which should be published after the authors have addressed the issues raised.

Christine Jenkins

Wed Jul 21 15:42:00 1999This is a descriptive study of prevalence of asthma, asthma action plans and severity of asthma in the South Australian population. The study also looks at the relationship between the ownership of asthma plans and asthma morbidity to patient perceptions of their asthma and patient self management.
Methods
The method of data collection appears to be appropriate and reliable. The survey size is large and response rate good. It should be made clear that in this study a person was classified as having asthma if they satisfied all three criteria a), b) and c). This is not absolutely clear and if the subjects did not meet all criteria but were included in the study it would introduce selection bias (par 20)
It is interesting that the researchers had several measures of disease morbidity but only chose one for the outcome measure. Was there any reason for this?
I agree with Dr Jenkins. I am unsure about the reproducibility and validity of the questions on perception. It is difficult to determine these without the questions printed in full.
The analysis of the data appears to be appropriate.
Results
The number of cases needs to be included in the text not just the percent. (par 24) e.g 346 people had asthma.
The lay out of table 1 makes it difficult to follow. The results need to line up with the variables.
It needs to be clear that the information in par 26 relates to the univariate analysis.
Discussion
The opening sentence in par 30 shows over interpretation of the results. No significant differences in demographic characteristics were found. This does not mean that testing asthma management on a population basis will be difficult. Further over interpretation of the results is shown at the end of the paragraph. The results show that there was no significant association between having an action plan and feeling comfortable about managing asthma. It should not be interpreted as deficiencies in comfort of management of asthma. The same applies to the other variables mentioned in that sentence.
In par 31, line 3; it is non-smokers who waken frequently not no-smokers. Non smokers and smokers have been analysed separately. You can't pool the results and discuss asthmatics with nocturnal symptoms without referring to the smoking status(par 31).

Overall, this is a good study, but I think the main point has not been emphasised enough. The results show that the subjects with an asthma plan are more likely to use corticosteroids regularly, understand asthma, have enough information on asthma and have some form of self management(peak flow meter). If those on corticosteroids are cases with more severe asthma, the results suggest that only the more severe cases have asthma action plans and this may relate to a better understanding of the disease and self management. If this is so,the results of this study suggest that more work is required in the education and development of action plans in less severe cases of asthma.

Sally Reagan

Wed Jul 28 14:02:15 1999Editorial advice re "A population survey of the use of asthma action plans. Where to from here?"

Thanks to the authors for the submission of this manuscript and to the reviewers for their participation. There seems to be unanimous agreement that it is important to publish this study pending a satisfactory revision and the reviewers have provided useful comments to make this possible. The manuscript has also been discussed at the editorial committee meeting.

From an editorial point of view, we would offer the following comments

TITLE: You may wish to reconsider this, as per NS "The title of the manuscript refers to the "use" of action plans whereas the text more correctly refers to "ownership". This survey did not address the issue of actions taken by those in possession of an action plan (ie. use or effectiveness of the plan) nor whether a plan had been prescribed but ignored (ie. not reported) by the asthmatic. Perhaps the title of the manuscript should be amended to make this point clear".
ABSTRACT: We are not sure what you mean by "written relationships" (par 8);
The conclusion (par 14) should be more specific (see several of the referees' comments on the need to emphasise the important findings of the study and make specific recommendations).

INTRODUCTION: This sets the scene well but is a bit too long. You probably don't need to include the details of what you looked for (par 18) at this point.

METHODS: Is there a reference for the methods of the SA health omnibus survey(par 19)?
Did the classification as asthmatic require all three criteria (par 21)?
Please provide your working definition of an asthma action plan (par 21), particularly if you distinguished between written and conceptual.
Please reference the statistical package (Par 22).
Please address SR's question re why you chose only nocturnal awakening as a morbidity measure.
Did you ask the subjects how often they saw a doctor with their asthma? This is a variable which is often analysed, and which you have examined before.
RESULTS: Please include numbers as well as %s (par 24)
Two referees have asked for the exact wording of the questions to be reported in a table (or could they be included in table one?) We agree that this is a good idea.
Overall, the presentation of the results is a bit confusing. Please clarify table one as per CJ's suggestion. We also share her confusion about the differences between variables associated... (par 25) and variables which described... (par 27). See also NS's query. This may be solved by recording the number (%) of people with an AA plan and the number (%) who woke at night somewhere on the table. See also SR's comments on this section.
DISCUSSION: CJ raises a good point about written/verbal AAPs. You should clarify this and discuss, particularly as many of our readers will have read the MJA article. CJ and SR also provide some good ideas for focussing the discussion on the main messages. Word limits will not allow you to increase the discussion length too much, so you should consider pruning some of the information in paragraphs. 30 and 33. You should also mention some of the limitations of your data.

As mentioned, we consider all the referees' comments to be worth your consideration. Please address them specifically in a covering letter and highlight changes you make to the manuscript (using bold font or underlining). Please also adhere to our limit of 2500 words.

We are interested in publishing the manuscript (in print) for asthma week at the beginning of October. If possible we would therefore like you to make your revisions by Friday August 27 1999. This is shorter than the usual time period given so please let us know as soon as possible if this is untenable.

We look forward to hearing from the authors and thank everyone once again for their contributions.
Ruth Armstrong

Thu Aug 5 14:22:33 1999Comments from the authors with regard to the Referees Reviews

The first point that we accept is that Table I was not set out in an ideal fashion and this is commented on by all Reviewers. We therefore changed the format of Table I (new table now on web). It now includes numbers of the population within each variable that has been talked about. The percentages are those within that group who either own an asthma plan or the waken weekly or more frequently. Changing the lines on the table hopefully makes this table clearer.

With respect to individual comments about the paper our responses in order are:-

Prof Saunders
*We agree that the title is a little misleading and his suggestion of using ownership is a good one. We would therefore suggest that the title reads "A Population Survey of the Reported Ownership of Asthma Action Plan; Where to from here?"
*It is difficult to provide a more complete picture of the profile of asthmatics assessed. This was a face-to-face survey but without objective measurements. The amount of data that we can collect in this survey has been limited by the practicalities and although we have knowledge as to whether people have used inhaled corticosteroids regularly we don't have a dose. Therefore we think the modification of the Table now presents all of the information that we have to give the profile of asthmatics. We agree that we are left with questions which can only be answered in a prospective study which includes objective measurements of lung function, drug doses, symptoms etc e.g. distinguishing between mild asthma only, and good control with a lot of treatment can not be decided on current information. The discussion will be been broadened as suggested to include adoption of best practice guidelines etc. The ownership of action plans not being a distinguishing feature of wakening with asthma will be altered in the last sentence para 26.

Dr Jenkins
Dr Jenkins makes a valid point about the questions. The reproducibility of the perception questions have not been assessed. These questions (or all questions) can be presented in an appendix or obtained on request. There is a potential problem with the length of the article with some of the additions in the discussion. Perhaps an editorial decision is required here.

The assumption that the ANZJM 1997 and MJA 1997 articles are including the same study is incorrect. The MJA 97 study was a one off - done out of synchrony with the spring SAHOS and using different questions. Within that study action plans were reported at 39.5% but it was the asthma management plans "a co-ordinated method of management etc" that Dr Jenking quotes figures for. The spring SAHOS from 1992 has used the same question to self reported doctor diagnosed asthmatics each year "Do you have an asthma action plan i.e. written instructions of what to do if your asthma is out of control?" Therefore the only valid comparison is with the reported ANZJM 1997 data.

The issue of guidelines has been addressed earlier and will be included. The use of the NAC documents in the reference list to support this discussion seems appropriate however this report is not a survey of uptake of plans by doctors.
Action plans could be a marker for something else, but this is a topic for a prospective study including qualitative research

Dr Reagan
In the methods section para 21 it is stated "a person was classified as suffering from asthma if (a) they had ever had asthma (b) their asthma had been confirmed by a doctor, and (c) they still had asthma" We are uncertain as to how this should be altered to make it absolutely clear.

The outcome measure of nocturnal wakening was chosen because of its clinical applicability - currently and in the future. It is one of the 3 to 5 questions for assessment of control used by clinicians generally. Hospitalisation is a relatively uncommon event (~5%) in a community survey of asthma.

The issue about perception questions has been discussed previously. Changes will be made to the discussion to cover the over-interpretation that has been commented on.
The point Dr Reagan makes about the main point of the study is relevant. The conclusion section of the abstract and of the body of the paper will be altered to take this issue into account.

To the Editor
The earlier comments were directed towards the referees and I have not responded directly to your issue. The word written on para 8 should be deleted and we will change the conclusion as set out. I think I do need your advice about the inclusion of all the questions because it will make the table complex and whether an appendix is acceptable etc needs to be discussed. I think I have answered a range of other questions within my response to the individual reviewers questions.

I am assuming from your comment that you wish to have the manuscript in hard copy or do you wish to have it emailed? The other thing I need your advice on is that you ask about a previous reference for the health omnibus survey. There are references of the description and are you happy to have the methods section reduced by including that reference for the methodology of the survey?

Thank you for your help.
Professor R Ruffin

Wed Aug 25 18:40:20 1999
Authors' covering letter for the revised manuscript

I think that the changes made to the manuscript address the majority of the concerns and where we have not altered the manuscript we are providing some argument about why that is unnecessary.

I would like to go through and report the changes in a point by point format. You will note the changes have been underlined as requested in the accompanying document.

1. We agree with the suggestion that the title should be changed to continue to use reported ownership rather than use of asthma action plans.
2. The word "written" has been shifted in the objective as it does emphasise that we are examining written asthma action plans and it was in an incorrect position.
3. We have added the year of the setting of the study to the abstract. We have modified the results section to emphasis the positive results of the study. The conclusions have been modified to focus on where potential strategies should be examined.
4. We have taken on board the editorial comments about restricting the last paragraph of introduction.
5. A reference is now provided for the South Australian Health Omnibus Survey.
6. We have changed the wording to try and emphasise that each of the questions had to be positive for someone to be classed as having asthma.
7. There is a reference for the statistical package.
8. In the result section we have included the number of patients with asthma.
9. In the third paragraph of the results we have incorporated the specification that the association was at the univariate level.
10. In the same paragraph we have modified the section about the association of wakening with asthma and ownership of an asthma action plan to reflect the results.
11. In the last paragraph of the results section we have specified that this was a multivariate analysis.
12. We have included a small section on the limitations of the study. In this section we also partly address the issue of written action plans vs verbal action plans. The overinterpretation of results commented on by Dr Reagan has been taken on board and discussion modified around that. We have also incorporated Professor Saunders discussion point about whether people had good controls of severe asthma or poorly controlled mild asthma.
13. We have emphasised practice guidelines as being an area that needs to be examined and that this may be a possible explanation for a fall in asthma action plan ownership. The references that Dr Jenkins suggested have now been incorporated.
14. The conclusion again has been modified as in the abstract to emphasise areas which need to be examined before developing new strategies.
15. We have previously answered the question about why we only chose nocturnal wakening in our response to the review process and I will just repeat that argument.
We did not ask subjects how often they saw their doctor and we have now included the questions as an appendix.
16. There is a good deal of argument about the issue of written/verbal asthma action plans. I have previously commented on the potential misinterpretation by Dr Jenkins that the studies that she quoted were the same studies. There were indeed separate studies and had asked different questions. The question that was worded in this study is referred to the appropriate question in earlier data.
17. The abstract has been modified and includes the requested references. Other references have been pruned so the total remains at 25.
18. Table one has been modified to now include the number and the lines have been drawn differently so that I believe it becomes easier to follow.
19. The appendix now includes the survey questions.

I would be happy to discuss other issues.

Richard Ruffin

Thu Aug 26 16:44:32 1999

Editorial response to and decision on "A population survey on the reported ownership of Asthma Action Plans".

Thank-you to the reviewers for your helpful comments on this paper and to the authors for their timely and methodical response to the reviewers' comments. We are accepting this paper for immediate publication in our electronic journal and print publication in the October 7 issue.

(To Professor Ruffin) We are editing for the print journal over the next two weeks and hope that you will be available to read and amend/approve galleys. Please let us know as soon as possible if this is not feasible.

Thanks again for your submission to the MJA
Ruth Armstrong, Assistant editor