Patients’ experiences with fluctuations in persistent physical symptoms: a qualitative study

Objectives To explore patients’ experiences with fluctuations in persistent physical symptoms (PPS) and to understand which factors—from their viewpoint—play a role in these fluctuations. Design Qualitative study using semistructured interviews and thematic content analysis. Setting This qualitative study is part of a multicentre prospective cohort study on the course of PPS. Patients were recruited in general practices and specialised treatment facilities for PPS throughout the Netherlands. Participants Interviews were conducted with a sample of fifteen patients with PPS to explore their experiences with fluctuations in symptom severity. Results We identified three themes in the analysis: (1) patterns in symptom fluctuations (2) perceived causes of symptom exacerbations and (3) Patients’ strategies in gaining control over symptom exacerbations. Daily and weekly fluctuations in symptoms were an important element in patients’ experiences. In particular anticipating on the worsening of symptoms impacted their daily routines and posed various challenges. Symptom exacerbations were attributed to overstepping physical limits and/or the impact of negative emotions. Resigning to physical limits, adjusting ones daily planning, weighing personal needs and learning to say ‘no’ were described as different strategies in gaining control over symptom exacerbations. Conclusions Fluctuations in the severity of symptoms—and in particular daily and weekly symptom exacerbations—are an important element of the symptom experience in patients with PPS and poses various challenges. Patients attributed symptom exacerbation to overstepping physical limits and/or negative emotions. Patients described different strategies in gaining control over symptom exacerbations.

P4 line10: Please elaborate a little how symptoms may vary in severity and how severity may impact impairment. P4 line13: Definitions and terminology are much debated in this area. Please define more clearlyare all FSS also PPSand vice versa? P4 line 19 -21: The studies you are referring to are on MUPS, somatization and hypochondriasis. Are all of these covered under your definition of PPS? The definition of PPS from Box 1 might be helpful in the introduction, with an elaboration on how it relates to other terms used in the paper.

Method:
Overall the context, sampling strategy and data collection are well described. Box1: very helpful elaboration of the overall study. I have only a minor comment for this section: P5 line 17: Please elaborate the definition of symptom clusters P 7: The sections on data-analysis, the description of reflexivity, analysis and trustworthiness need more work. First off, the research question in itself suggests that researchers entered this project with a hypothesis, or preconception if you want, that fluctuations in symptom severity are of importance for patients with PPS. This however is also one off the main conclusions. And while this can be the case, some consideration of what preconceptions were and how these may have influenced the analysis is relevant. In the methods section, consensus decision and reliability are mentioned. However, reliability as defined in quantitative research cannot be applied directly in a qualitative study. Please elaborate how reliability is defined and how the validity of findings is supported. You might find inspiration in the following papers: Results: The results section needs substantial work. Again, you might want to draw on the two papers by Malterud for inspiration.
Throughout the results section there is emphasis on commonalities, consensus and on "what most patients say". However, in a qualitative inquiry, focus should also be on different perspectives on a theme, on understanding the variations in how fluctuations are experienced and influenced rather than the mere statement that they are. This could be kept in mind throughout the revision process. Further the results section lacks qualitative rigor. The authors present their own thoughts and interpretations, sometimes conclusions, between quotations, but there is not always sufficient support for the statements made in the quotations used. Quotations should support the statements made and the reader should be able to follow the researcher"s line of thought. I will provide a specific example: P 9 line 47, crossing emotional boundaries: A number of patients mentioned not just the importance of respecting their physical limitations, but also the importance of respecting their emotional boundaries.
"And sometimes I think: I am not going to do that. If I am not well and it"s not something I really enjoy. No … I evaluate: is it worthwhile, does it do me any good? Is it something I enjoy? If not, I say no. You also need to learn to say "No". I didn"t do that when this started." (P12, female, 70 yrs) A couple of patients indicated that their symptoms also worsened following acute emotionally stressful events. In these cases, they felt not capable to control symptoms.
For me as the reader, the chosen quotation neither supports the above nor the below statement. If it does so, the authors don"t succeed in explaining how, sufficiently, for me to understand. The patient describes learning to say "no". Saying no however, can be a way to respect personal physical boundaries rather than emotional boundaries. Further the quote also suggests that saying no depends on other factors than possible symptom exacerbation. The patient describes how she evaluates potential activities based on possible personal gain and enjoyability. So it is not so much about avoiding "overstepping of boundaries", rather it is about gaining something, setting priorities. Further, it is very interesting to learn how P12 has acquired the skills to say "no", -is she more aware of what is good for her, is she less afraid of possible consequences…? However, the quote doesn"t say anything about thisbut maybe that could be an interesting theme on its own? I don"t have sufficient context to provide a meaningful analysis of a single quotation off course, so this should just be seen as inspiration.
The above general comments are relevant throughout the results section. Below, I will present some more specific comments to individual subthemes.
P8: Theme 1the first section lists a number of quotations which support what you already knew from your sampling -that you have included patients who do in fact experience fluctuation. So this is hardly an interesting or surprising finding. The quotations, however, also pertain to "patterns of fluctuation"this might be a slightly different focus to explore in this theme? P8 line 12: in your presentation of participant selection you emphasize the importance of selecting participants with fluctuating and stable symptoms respectively. Please elaborate how that relates to the fact that all participants in fact do experience fluctuation and how this may affect your findings. P8 line 21: paid or unpaid jobs, what do you mean? Does it matter? P 8 line 45: In these final quotes, patients talk about how the anticipation of symptom exacerbation, rather than experienced fluctuations, affect their daily life and activities. This may be a very relevant theme to explore. However, it does not fit in the title "experiences with fluctuation".
P9: Theme 2. This theme addresses the patient experience of why symptoms may fluctuate, which is very relevant. Maybe the title could reflect this. The introduction provides a very interesting and relevant conclusion and may be placed at the end of the subtheme description. The results have to support the claims though. The section on "balancing physical limits" contains relevant quotes which support the statements. The section on "crossing emotional boundaries" needs more work. As highlighted above, I cannot follow how quote 1 and 2 pertain to emotional boundaries. P10 line 12: Quote 3 in this section is concerned with the association of negative emotions and symptom experience. I don"t know if it is about emotional boundaries? However, the influence of negative emotions on symptom experience is extremely interesting an could be elaborated further. P9: Theme 3: Dealing with fluctuations: experienced difficulties in balancing limits and boundaries The theme is relevant and interesting; however, the introducing statement or conclusion again does not have sufficient support in the quotations and interpretations. Overall, I think that the quotes rather are concerned with the patients" ways of dealing with symptoms, priorities and life as such, rather than with dealing with fluctuations.  However, the claim made that lack of diagnosis/validation leads to overstepping of boundaries which again leads to symptom fluctuation is simply not one that can be supported by any qualitative data. Qualitative inquiry does not provide information about causation. Please revisit this section.
In the final section on page 11: Importance of resigning to limits and boundaries, the authors make statements about how patients have gained control over symptoms. Again, the quotes do not support the claims made. The final quote rather suggests that "not resigning into symptoms" is experienced as gaining control. The text does suggest so, however the overall conclusion you present later is that: overallresigning is a good thing. An interesting analysis might be the different perspectives on "gaining control" that you can find in your data. 1) "resigning", 2) "consciously overstepping", 3) " evaluating an activity and making choices based on personal preferences" (as was suggested in a previous quote). 4) …. I am sure you have a lot of data to provide perspectives on this matter. For the clinician, this could be very important informationgaining control is achieved differently by various patients, possible methods of gaining control were….

Discussion
I will not comment on the discussion in too much detail, as it probably will change when the results section has been revised. P11 line 16: The concluding remarks of the results are very interesting, however not sufficiently supported as described above. P11 line 47: I absolutely agree with the strength of and relevance of this study P12 line 13: This section starts with the claim that letting go of the search for a diagnosis results in less fluctuation which again results in improved wellbeing. Again, carefull with claiming causation. Maybe you can find inspiration for the discussion of this point in the literature suggestions above.
Checklists comments: 2. The abstract will need changes after the revision 8. The references used are relevant and up to date. However, the list reflects focus on quantitative literature. More relevant qualitative literature might prove helpful. . The impact of conveying the diagnosis when using a biopsychosocial approach: A qualitative study among adolescents and young adults with NES (non-epileptic seizures). Seizure, 24, 107-113. doi:10.1016/j.seizure.2014.09.006 9.-11. Please look to the comments on the results section above I hope that you find the comments helpful and that they will allow you to revisit your very interesting study and data with new inspiration. I want to thank you for allowing me to comment on your work.

GENERAL COMMENTS
The authors have performed an original and important study on a very common medical problem. Before their study it was indeed unknown how patients perceive the fluctuations in their symptoms and I find that my patients often try to find (sometimes in vain) any correlation between the fluctuations and any behavior or other attribute. Although the authors acknowledge that their data stems from a selected group of patients with PPS and might therefore not apply to all patients with PPS, I have learned a few points and so should other healthcare providers, to improve the care for these often "difficult to treat" patients. That said, I have a few comments. First, in the methods selection both patients with fluctuations and those with (seemingly) stable course were included, however this distinction does not trickle down in the rest of the article. Can the authors explain why, was there no two different group of patients (if so, explain table 1 and this seems to contradict the first paragraph of the results, line 14-15 page 8) or did they not differ in their discussed themes? This could help personalizing the care for PPS patients. More-over I would like to see more details about the patients. What kind of PPS do they have? The authors state it is over more domains, but to give a clear picture to readers, it would help to see that xx% have pain, xx% have fatigue, etc etc. Furthermore, can the authors give more quantification on the level of fluctuation over both long and short term. Are they really fluctuations or only perceived as such, if real, how much and is that amplitude of fluctuation related to outcome e.g. are high fluctuators more impacted on daily life that low fluctuators. Another point might be about the external validity. Can the authors elaborate more how applicable these results might be in a culturally different population than Dutch people, since I think culture and perspection of PPS are two very important issues. Can we use these data on patients from other cultures living in the Netherlands / Western Europe? Can the authors also give 1-2 extra examples on the topic of "importance of resigning to limits and boundaries"? Finally, I am not sure whether all readers have the same understanding of the term "resignation" which is used frequently. I suppose the authors use a different wording or explain the term. In all I think the authors will contribute substantially to the medical field of PPS / MUPS if these data are published.

VERSION 1 -AUTHOR RESPONSE
Reviewer #1: Reviewer Name Caroline Mitchell Institution and Country University of Sheffield, UK Please state any competing interests or state "None declared": none declared Please leave your comments for the authors below: Overall this is a good qualitative study with attention paid to fulfilling rigor in standards of conduct of research. Participants were positively selected and interviews continued to data saturation . Data analysis : term 'reliability' is not usually used in reference to independent verification of themes in qualitative research -cf Lincoln and Guba criteria -please rephrase (line 34 page 7) Response: Thank you for your helpful suggestion, we are aware that there is considerable debate about the concept of reliability in qualitative research, we therefore replaced it by more suitable terminology. We therefore draw upon Malterud"s (2001)  -"In all phases at least two authors were involved (HB, EW, FB) to increase reliability." was changed into: "In all phases at least two researchers were involved (HB, EW, FB) to enrich the analysis." -"…improve reliability" was changed into: "…improve internal validity" Reviewer #2 Reviewer Name Ditte Roth Hulgaard Institution and Country University of Southern Denmark Denmark Please state any competing interests or state "None declared": None declared Dear authors, First off, I want to congratulate you on your wonderful PROSPECTS study, which is relevant and needed in the field of PPS. Further I applaud the choice to use both quantitative and qualitative methods in order to broaden perspectives and knowledge.
The current study has a relevant focus and appropriate methodology. However, the paper needs substantial work with regard to qualitative rigor, analysis and interpretation. Thus, I have minor comments for the introduction and methods section, while results and discussion need a substantial revision.
Introduction: -P4 line10: Please elaborate a little how symptoms may vary in severity and how severity may impact impairment.
Response: We believe that elaborating on how symptoms may vary in severity and impact functioning goes beyond the scope of this part of the introduction -where we aim at introducing the societal relevance of PPS research in general. We do, however, pay attention to this topic in the paper.
-P4 line13: Definitions and terminology are much debated in this area. Please define more clearlyare all FSS also PPSand vice versa?
Response: In our study we focus on overlapping experiences among patients with PPS. In order to be clear on our definition, we introduce it here -as you also suggested in your next comment-and provided some more background about the debate.
Changes (Introduction p. 4): The following was added: "There has been extensive debate about definitions and terminology in this field of research. Whereas some emphasize commonalities and overlap in symptoms and characteristics (Nimnuan 2001, Fink 2007, Fink 2010, Budtz-Lilly 2015, others differentiate between particular functional somatic syndromes (FSS), such as fibromyalgia, chronic fatigue syndrome and irritable bowel syndrome (Wessely & White 2004, Jones 2019, Abbi 2013). The importance of studying both similarities as well as differences has also been highlighted (White 2010). In this study we focus on similarities and overlap in patients" symptom experiences. We defined PPS as symptoms, which last at least several weeks and for which no sufficient somatic explanation is found after proper medical examination by a physician. This is in line with the current Dutch multidisciplinary and general practice guidelines for MUS We added the words in italics: According to a number of studies conducted in primary and secondary health care settings, 50-75% of patients with PPS showed symptom improvement over time, whereas 10-30% worsened (olde Hartman 2009). In a cohort study that we conducted on the course of PPS we found improvement (63%) and deterioration (27%) rates were in line with prior literature, when using total changes scores based on two measurements.

Method:
Overall the context, sampling strategy and data collection are well described. Changes (Methods, page 5), we adjusted/ added the following: "Almost all of them (N=14) had symptoms in at least two of the following symptom clusters: 1) gastrointestinal; 2) cardiopulmonary; 3) musculoskeletal/pain and 4) general symptoms (headache, dizziness, memory impairment, concentration difficulties, fatigue). These symptom clusters were identified in a prior study by Fink et al (2007)  For additional changes concerning the symptoms see table 1 on page 6.
-P 7: The sections on data-analysis, the description of reflexivity, analysis and trustworthiness need more work.
First off, the research question in itself suggests that researchers entered this project with a hypothesis, or preconception if you want, that fluctuations in symptom severity are of importance for patients with PPS. This however is also one off the main conclusions. And while this can be the case, some consideration of what preconceptions were and how these may have influenced the analysis is relevant.
Response: This is indeed a relevant point. We therefore added a few sentences on our preconceptions to the methods section.
Changes (data collection, p. 7): The following was added: "Based on our prior quantitative study (Claassen-van Dessel 2018), our preconception was that patients might experience fluctuations in symptoms and that these might be relevant to them. Based on theoretical sampling, we selected "fluctuating" as well as "seemingly stable" patients. We expected more prominent accounts on fluctuations in the "fluctuating" patients. Whilst we had this preconception, we asked open questions in both "fluctuating" as well as "seemingly stable" patients about the experienced symptoms over time (a day, a week, a month etc.) when interviewing the patients." -In the methods section, consensus decision and reliability are mentioned. However, reliability as defined in quantitative research cannot be applied directly in a qualitative study. Please elaborate how reliability is defined and how the validity of findings is supported. You might find inspiration in the following papers: Response: Thank you for your helpful suggestions. Reviewer #1 also commented on our use of the term "reliability". As we addressed in our response to reviewer#1, we are aware that there is considerable debate about the concept of reliability in qualitative research and we therefore replaced this term by more suitable terminology in qualitative research. For changes see our response to the comment of reviewer#1.

Results:
The results section needs substantial work. Again, you might want to draw on the two papers by Malterud for inspiration.
Throughout the results section there is emphasis on commonalities, consensus and on "what most patients say". However, in a qualitative inquiry, focus should also be on different perspectives on a theme, on understanding the variations in how fluctuations are experienced and influenced rather than the mere statement that they are. This could be kept in mind throughout the revision process.
Further the results section lacks qualitative rigor. The authors present their own thoughts and interpretations, sometimes conclusions, between quotations, but there is not always sufficient support for the statements made in the quotations used. Quotations should support the statements made and the reader should be able to follow the researcher"s line of thought. I will provide a specific example: P 9 line 47, crossing emotional boundaries: A number of patients mentioned not just the importance of respecting their physical limitations, but also the importance of respecting their emotional boundaries. "And sometimes I think: I am not going to do that. If I am not well and it"s not something I really enjoy. No … I evaluate: is it worthwhile, does it do me any good? Is it something I enjoy? If not, I say no. You also need to learn to say "No". I didn"t do that when this started." (P12, female, 70 yrs) A couple of patients indicated that their symptoms also worsened following acute emotionally stressful events. In these cases, they felt not capable to control symptoms.
For me as the reader, the chosen quotation neither supports the above nor the below statement. If it does so, the authors don"t succeed in explaining how, sufficiently, for me to understand. The patient describes learning to say "no". Saying no however, can be a way to respect personal physical boundaries rather than emotional boundaries. Further the quote also suggests that saying no depends on other factors than possible symptom exacerbation. The patient describes how she evaluates potential activities based on possible personal gain and enjoyability. So it is not so much about avoiding "overstepping of boundaries", rather it is about gaining something, setting priorities.
Further, it is very interesting to learn how P12 has acquired the skills to say "no", -is she more aware of what is good for her, is she less afraid of possible consequences…? However, the quote doesn"t say anything about thisbut maybe that could be an interesting theme on its own?
I don"t have sufficient context to provide a meaningful analysis of a single quotation off course, so this should just be seen as inspiration.
The above general comments are relevant throughout the results section. Below, I will present some more specific comments to individual subthemes.
Response: Thank you for your extensive commentary; this helped to strengthen our results section considerably. We revised this section thoroughly and checked all quotations. With regard to "learning to say no": in an earlier version of the manuscript we included a theme focused on gaining control, in which we explored several strategies. This got lost over the various versions (and shortening) of the manuscript (there"s a 4000 word limit). We agree it is however an interesting theme that was important in relation to symptom exacerbations, so we decided to discuss it in our revised version. Because we quite rigorously revised the results section, please see the manuscript for changes made, these are highlighted (pages 8-13).
P8: Theme 1the first section lists a number of quotations which support what you already knew from your sampling -that you have included patients who do in fact experience fluctuation. So this is hardly an interesting or surprising finding.
Response: We don"t agree with this, what was surprising was that in the interviews there was no distinction in experienced fluctuations between the "fluctuating" and "seemingly stable" sample of patients. So, in that sense, it is quite surprising that both "groups" of interviewed patients experienced fluctuations (over the day, week, longer periods). To emphasize the relevance of this finding, we added a sentence to underline this in the results section. Besides, we gave more emphasis to the "two groups" ("fluctuators" and "seemingly stable") in our sample in the methods section and in the discussion.
Changes (results, p. 8): The following was added: This meant that both the selected patients with fluctuations, as well as the seemingly stable patients in our sample experienced these fluctuations.
The quotations, however, also pertain to "patterns of fluctuation"this might be a slightly different focus to explore in this theme?
Response: Good point. In a prior version of the manuscript, we actually spoke more about patterns (short-term fluctuations; long-term fluctuations) and we now decided to bring this back in the paper.
See amendments made on page 8 of the manuscript (results section).
P8 line 12: in your presentation of participant selection you emphasize the importance of selecting participants with fluctuating and stable symptoms respectively. Please elaborate how that relates to the fact that all participants in fact do experience fluctuation and how this may affect your findings.
Response: We indeed purposively sampled both patients who fluctuated as well as patients who seemed stable based on our quantitative findings. As mentioned above, our interviews actually showed no particular difference between the experiences of these two groups; fluctuations in symptoms were a central theme in the accounts of all interviewed patients. Therefore, although we anticipated differences, this was in fact not the case. In particular short-term fluctuations led patients to a continuous search to what caused exacerbations of symptoms to anticipate on exacerbations and prevent symptoms from worsening.
P8 line 21: paid or unpaid jobs, what do you mean? Does it matter?
Response: With unpaid jobs we meant volunteer work. What we meant is that, it didn"t matter whether they had paid work or volunteer work. We decided to slightly change this formulation, because indeed it does not really matter.
Changes (results, p. 8): ""Worsening over the work week was described by all patients with (un)paid jobs."" Was changed into: ""Worsening over the work week was described by all patients who worked."" P 8 line 45: In these final quotes, patients talk about how the anticipation of symptom exacerbation, rather than experienced fluctuations, affect their daily life and activities. This may be a very relevant theme to explore. However, it does not fit in the title "experiences with fluctuation".
Response: We agree with your remark. When revising the results section, we replaced this section under theme 3 ("patients" strategies in gaining control over symptom exacerbations"), where we made a subheading named "Adjusting daily planning".
Changes: Because we quite rigorously revised the results section, please see the manuscript for changes made, these are highlighted.
P9: Theme 2. This theme addresses the patient experience of why symptoms may fluctuate, which is very relevant. Maybe the title could reflect this. The introduction provides a very interesting and relevant conclusion and may be placed at the end of the subtheme description.
The results have to support the claims though. The section on "balancing physical limits" contains relevant quotes which support the statements. The section on "crossing emotional boundaries" needs more work. As highlighted above, I cannot follow how quote 1 and 2 pertain to emotional boundaries.
Response: Thank you for your helpful remarks. We agree on adapting the title of theme 2 to better reflect its focus and changed it into "perceived causes of symptom exacerbations". We have critically revised the claims we made and removed concluding remarks and/ or claims suggesting causality from the results section. The section on "crossing emotional boundaries" has been altered and is now more focused on emotions and symptoms ("negative emotions"), which was an important (sub)theme in relation to the experience of why symptom may worsen.
Changes: Because we quite rigorously revised the results section, please see the manuscript for changes made, these are highlighted in the results section.
P10 line 12: Quote 3 in this section is concerned with the association of negative emotions and symptom experience. I don"t know if it is about emotional boundaries? However, the influence of negative emotions on symptom experience is extremely interesting an could be elaborated further.
Response: We agree. As mentioned in the prior response, we revised this section as mentioned above and discussed this in a subtheme about how patients linked symptom exacerbations to "negative emotions".
Changes: Because we quite rigorously revised the results section, please see the manuscript for changes made, these are highlighted in the results section (p. 10+11).

P9: Theme 3: Dealing with fluctuations: experienced difficulties in balancing limits and boundaries
The theme is relevant and interesting; however, the introducing statement or conclusion again does not have sufficient support in the quotations and interpretations. Overall, I think that the quotes rather are concerned with the patients" ways of dealing with symptoms, priorities and life as such, rather than with dealing with fluctuations.
Response: We looked at previous versions of the manuscript and came to the conclusion that we should reintroduce strategies in gaining control over symptoms as a third main theme. So, we revised this entire theme and renamed it "patients" strategies in gaining control over symptom exacerbations". Because we quite rigorously revised the results section, please see the manuscript for changes made, these are highlighted in the results section.

P10: Lack of recognition and validation of symptoms
The experiences with lack of diagnosis are well described in the qualitative literature and an important issue to address. However, the claim made that lack of diagnosis/validation leads to overstepping of boundaries which again leads to symptom fluctuation is simply not one that can be supported by any qualitative data. Qualitative inquiry does not provide information about causation. Please revisit this section.
Response: Because we needed to shorten our manuscript due to the major revisions of our results section and because this is a well-known -though important-issue, we decided to highlight this only shortly in theme 3 "patients" strategies in gaining control over symptom exacerbations" under the subheading "resigning to physical limits" (without making any claims about causation).
Changes: Please see the manuscript for changes made on this part, these are highlighted in the results section (p. 12).
In the final section on page 11: Importance of resigning to limits and boundaries, the authors make statements about how patients have gained control over symptoms. Again, the quotes do not support the claims made. The final quote rather suggests that "not resigning into symptoms" is experienced as gaining control. The text does suggest so, however the overall conclusion you present later is that: overallresigning is a good thing. An interesting analysis might be the different perspectives on "gaining control" that you can find in your data. 1) "resigning", 2) "consciously overstepping", 3) " evaluating an activity and making choices based on personal preferences" (as was suggested in a previous quote). 4) …. I am sure you have a lot of data to provide perspectives on this matter. For the clinician, this could be very important informationgaining control is achieved differently by various patients, possible methods of gaining control were….
Response: Thank you for your helpful remarks. We looked at previous versions of the paper and went back to our data and came to the conclusion that we should indeed address strategies in gaining control over symptoms as a third main theme (as we also mentioned in prior responses) and extensively revised this part of the result section.
Changes: Because we quite rigorously revised the results section, please see the manuscript for changes made, these are highlighted in the results section (p. 11-13).

Discussion
I will not comment on the discussion in too much detail, as it probably will change when the results section has been revised.
P11 line 16: The concluding remarks of the results are very interesting, however not sufficiently supported as described above.
Response: We revised our results, and part of these remarks became part of theme 3 (subtheme: resigning to physical limits), to support the statement we added a clarifying sentence and a quote.
Changes (results, p. 13), we added the following: Response: After our revisions of the results, we also revised the discussion section of our manuscript and made sure that any claims on a causal nature were taken out and the focus on patient experiences was highlighted.
Changes (discussion, p. 14+15): Please see for changes made, the highlighted parts.
Checklists comments: 2. The abstract will need changes after the revision Response: We revised the abstract and article summary after revising the manuscript.
Changes (abstract p. 2): Because we quite rigorously revised the results section and discussion, the abstract has been altered quite extensive as well. Please see the manuscript for changes made, these are highlighted. Response: Thank you for these suggestions. In this new version we included different qualitative studies in our discussion section to broaden the perspective on our findings (discussion, p. 14+15 highlighted version).
9.-11. Please look to the comments on the results section above I hope that you find the comments helpful and that they will allow you to revisit your very interesting study and data with new inspiration. I want to thank you for allowing me to comment on your work. Best whishes Response: It has been really helpful to receive your extensive and very helpful commentary, which definitely helped to strengthen our paper.

Reviewer #3
Reviewer Name MWF van den Hoogen Institution and Country Erasmus MC Rotterdam Please state any competing interests or state "None declared": None declared Please leave your comments for the authors below: The authors have performed an original and important study on a very common medical problem. Before their study it was indeed unknown how patients perceive the fluctuations in their symptoms and I find that my patients often try to find (sometimes in vain) any correlation between the fluctuations and any behavior or other attribute. Although the authors acknowledge that their data stems from a selected group of patients with PPS and might therefore not apply to all patients with PPS, I have learned a few points and so should other healthcare providers, to improve the care for these often "difficult to treat" patients. That said, I have a few comments. First, in the methods selection both patients with fluctuations and those with (seemingly) stable course were included, however this distinction does not trickle down in the rest of the article. Can the authors explain why, was there no two different group of patients (if so, explain table 1 and this seems to contradict the first paragraph of the results, line 14-15 page 8) or did they not differ in their discussed themes? This could help personalizing the care for PPS patients.
Response: Thank you for your positive remarks. Regarding your question, we can understand that our results evoke some questions about the distinction between the patients we selected "with fluctuations" and "with a seemingly stable course". The themes we identified were mentioned in both groups of patients-so indeed, there were no differences in the discussed themes and they both indicated fluctuations. The other themes were also identified in both groups. Because we probably did not clarify this well enough (Reviewer#2 also commented on this), we added some additional sentences on this in our results and discussion section.
Changes: The following was added (see highlighted version): This meant that both the selected patients with fluctuations, as well as the seemingly stable patients in our sample experienced these fluctuations (results, p. 8).