Dear Stuart Airey and Dr Tim Baird,
re: Irritable bowel syndrome (IBS), chronic fatigue syndrome (CFS) and fibromyalgia in those aged 45 and older – the impact of emotions and relationships on symptom severity.
I am writing to you with several concerns about this study which I participated in today.
My major concern is that I cannot see how the data collected from this study can fulfil the stated research aim indicated in the title. As you will be well aware, any correlations found between scores on questionnaires about symptoms, relationships and emotions cannot and should not be used to indicate a direction of causation. The title states that it you are seeking to find out 'the impact of emotions and relationships on symptom severity.' That implies a causal direction.
If a person has both difficulties with relationships/emotions and severe ME/CFS, FM or IBS, there are several possibilities:
1. Living with severe ME/CFS, FM or IBS and coping with severe symptoms and the life changes severe disabling illness causes may impact negatively the person's relationships, and may lead to natural feelings of sadness, anxiety and grieving for a lost active life, just as can happen with other chronic disabling diseases such as MS and Parkinson's Disease. The more severe the symptoms the more impact this may have on relationships and emotions.
2. The biological basis of ME/CFS is not yet known. It is possible that part of that biological dysfunction may affect brain chemistry, thereby affecting mood in ways not able to be influenced by thoughts or actions.
3. Lack of medical understanding, belief and medical and social care and support may lead to isolation, financial hardship, inappropriate or lack of care etc, which can lead to anxiety, despression, relationship breakdown etc. The more severe the symptoms the more impact this may have on relationships and emotions.
4. Individuals with ME/CFS, FM or IBS may, independently of their condition, also suffer from other comorbid conditions, including anxiety, depression, or may have problems with bad relationships generated by other individuals such as coercive control. There may be no causal connection in either direction between the symptom severity and these other circumstances.
5. Suffering from ME/CFS, IBS or FM may lead to negative changes in behaviour or thoughts which in turn lead to worsening symptom severity (I have seen no research evidence to support this last step).
6. People with particular personality types and/or behavioural patterns and/or beliefs about illness may make their symptoms more severe, or believe their disease is more severe than it really is with a combination of negative thinking and unhelpful behaviours such as avoidance of activity and social interaction. (repeated research has failed to support this proposition - I can give you references)
If you take the above items in turn and think about the causal direction in relation to your hypothesis, and your study, it should be crystal clear that there is no way your data from the questionnaires will be able to distinguish which if any of them may be correct. Your study title could equally have stated as its aim:
– the impact of symptom severity on emotions and relationships.
If you find no significant correlation between the answers on symptom severity and any of the measures you are attempting to find for relationship difficulties and emotional difficulties, I guess that would indicate that there is no causal explanation in either direction, or at least that your inadequate questions have not managed to elicit any.
If you do find some statistically significant correlations, after correcting, of course, for multiple comparisons, what then? You clearly will not be able to determine any causal direction, so will not have advanced understanding of chronic illness in any useful way. You will have no way of distinguising any of the 6 possiblities I listed above, or any other I haven't thought of.
_____________________
I have specific concerns about some of the questions too, which I summarise below:
Question 12:
For each symptom patients are asked to rate how much they are bothered by it. That can be interpreted as how bearable the symptom is, or how much it impacts your life and stops you doing things, or how much it makes you unhappy or angry or some other emotion, or how much it worries you, or how much you catastrophise about it. It's far too ambiguous to provide useful information. There is no question about the severity of the disease.
The recently published NICE ME/CFS guideline lists core symptoms, the most important of which, post exertional malaise, is not even included in your list. The guideline also lists severity levels with descriptors. If you want to rate participants disease severity, why not use these descriptors, or use one of the recognised severity level scales. (I can provide references).
So how are you measuring severity? Question 12 will not enable you to do so.
Question 13
Most of these are also ambiguous and therefore open to misinterpretation by both patient and researcher.
For several of them, I answered on the basis of current scientific research, so for questions 2, 3, 4 and 7 my answers were based on facts
1 - I have severe ME, so of course it affects my life a lot. There was no question about severity level, so this could be misinterpreted as me catastrophising, or giving in to symptoms, rather than being severely physically disabled.
2 - there is no cure in sight in my likely lifetime (I'm quite old), and people who have had ME as long as I have rarely if ever recover, so answering 'never' is a realistic answer, not an emotional one, but could be misconstrued as negativity, giving up, not wanting or trying to get better etc.
3 - it is impossible to 'control' the fact that we have this illness or its severity level, except by day to day managing as best we can to try to avoid worsening - so the question is ambiguous.
4. - there is no effective treatment for ME/CFS, so what is the expected answer that won't be judged as not trying?
5. No idea what is expected here. How much I experience symptoms can be taken to mean are they present all the time? or are they bad enough to dominate my life? or do I symptom focus too much? How will this be judged in correlating it with emotions?
6. How concerned am I? Well, it would be very odd having a life changing illness not to be concerned. But how long is a piece of string? Do you mean do I spend all day ruminating on my misfortune? Do you mean do I phone my doctor every week asking for help? Or do you mean do I spend a lot of my very limited energy following the research and helping to run a forum about my illness? I doubt you mean the latter. Again, will you judge participants negatively for saying they are 'concerned'?
7. A stupid question. Nobody understands ME, so how can any patient do so?
8. I have been ill for 32 years, how am I supposed to summarise the various emotions I have experienced about my illness over that period in a single number. Whatever random number I choose will be meaningless, and subject to misuse by a researcher with an agenda.
Question 14 on relationships
I was very concerned about this section. It seemed to me very intrusive while at the same time not providing any information of relevance to the person's physical health.
Collecting this amount of detail about what may be anything ranging from:
- a current happy relationship with a very supportive partner
- a far distant past failed relationship that happened before getting sick,
- the relationship of an adult child to their parent/carer,
- adults who have no relationships because of isolation in ill health
- a sick person who also has caring responsibilities
- and a host of other permutations
without asking anything about which of these applies is remarkably insensitive and by jumbling all of that together becomes completely meaningless.
And nowhere do you ask a direct question about whether the patient thinks their disease has changed their relationships, or about whether some aspect of their relationships has affected their disease severity.
Honest disclosure - in order to see the rest of the survey, that section had to be filled in, so I just went straight down the middle, selecting the same neutral answer for every statement, since none of them are relevant to me in the way implied by the research aim.
Questions on emotions
Again, an uncomfortably ambiguous and useless set of questions. Trying to lump together all emotions in this way seems silly to me. Do you mean just negative emotions? And whether one shares ones emotions with others can depend largely on one's home and life circumstances and personality, age, experiences, and the behaviour of others around us in ways completely unrelated to being ill. There's also a lot of judgement in these - is it a good or a bad thing to control emotions, or to look back on one's life and reflect on past experiences? Are you going to judge someone as emotionally inadequate if they keep their emotions to themselves, or is that a good thing? How can you possibly make a numerical score to use in statistical analysis from these questions without implying value judgements that may be completely inappropriate?
Questions relating to symptoms of anxiety and depression
Some of these may have more to do with the way others treat the individual and difficult home, financial and care circumstances. There is no way of telling whether these are related to the physical disease, social circumstances or comorbid unrelated anxiety or depression disorders.
___________________
I would like to ask you for a copy of the research proposal, hypothesis being tested, statistical analysis plan and ethics approval for this study, as well as answers to the questions my comments raise. I have posted my concerns and will share a copy of this email on the Science for ME forum. You would be welcome to join us there to further this discussion. I will share your reply on the forum too if you give me permission to do so.
In response to my comments about the study on the forum, Professor Jonathan Edwards, who was an expert witness to the NICE ME/CFS guideline committee, and whose statement can be read on the NICE website, said the following:
The main purpose of a PhD is to show that the candidate has learned what is bad methodology that just perpetuates your prejudices, and what is good methodology that shows you how to find out something new. With luck responses from patients will help to serve that purpose by making it clear that this project, in its current form, is firmly in the first category. You cannot infer impact from correlation.
I trust that as a PhD student and supervisor, you are both eager to improve your research methods and to learn from the experience of expert patients, carers and clinicians who have been following ME/CFS research for some time and unsurprisingly want to see high quality research that can find answers to real questions of value to patient care. This detalied response is offered in the spirit of helping you in learning to be better researchers. I hope it will be received in that light.
I look forward to your reply,
With best wishes,
Trish Davis
https://www.s4me.info/threads/uni-o...-45-and-older-–-the-impact-of-emotions.24087/
