Trials on therapist-delivered treatments (for illnesses or symptoms that don't have biomarkers yet) that used objective outcomes as primary endpoints

As the title indicates.

Looking for examples of such trials.

Systematic reviews (*) of trials that excluded (or downgraded to very low quality) non-blinded trials that used only subjective outcomes as primary endpoints also welcome.

Suggestions on amendments to the threads' title or topic much appreciated, too.

(*) Other than the NICE guidance committee's assessment on the evidence of interventions for ME.

 

A PubMed search expression that can be helpful: (accelerometer) AND (fatigue[Title] OR pain[Title])

The same one with filters for clinical trials (controlled or not), systematic reviews and meta-analyses: https://pubmed.ncbi.nlm.nih.gov/?te...dcontrolledtrial&filter=pubt.systematicreview

 

Example of a systematic review of trials using health care utilization as an objective outcome

That's not a good example because it didn't investigate effects on a symptom or on disability, but on an emotional state -- reassurance.

I am not sure if health care utilization usually counts as an objective outcome for effevtiveness of interventions. Also did not checke whether the authors actually excluded studies that only used 'measured reasssurance' but not health care utilization, and maybe some people count 'measured reassuranace' as an objective outcome, too?

But perhaps it illustrates some difficulties with reasonable definitions of objective outcomes, so just leave that here:

1) Traeger AC, Hübscher M, Henschke N, Moseley GL, Lee H, McAuley JH. Effect of Primary Care–Based Education on Reassurance in Patients With Acute Low Back Pain: Systematic Review and Meta-analysis. JAMA Intern Med. 2015;175(5):733–743. doi:10.1001/jamainternmed.2015.0217

https://jamanetwork.com/journals/jamainternalmedicine/fullarticle/2204032

Design Systematic review and meta-analysis of randomized and nonrandomized clinical trials.

Study Selection To be eligible, studies needed to be controlled trials of patient education for LBP that were delivered in primary care and measured reassurance after the intervention. Eligibility criteria were applied, and studies were selected by 2 independent authors.

Main Outcomes and Measures The primary outcomes were reassurance in the short and long term and health care utilization at 12 months.

 

An example of a RCT showing no effect on the main objective outcome but improvement on secondary subjective outcomes:

The effect of education and supervised exercise on physical activity, pain, quality of life and self-efficacy - an intervention study with a reference group

Background: Individuals with knee and hip osteoarthritis (OA) are less physically active than people in general, and many of these individuals have adopted a sedentary lifestyle. In this study we evaluate the outcome of education and supervised exercise on the level of physical activity in individuals with knee or hip OA. We also evaluate the effect on pain, quality of life and self-efficacy.

Methods: Of the 264 included individuals with knee or hip OA, 195 were allocated to the intervention group. The intervention group received education and supervised exercise that comprised information delivered by a physiotherapist and individually adapted exercises. The reference group consisted of 69 individuals with knee or hip OA awaiting joint replacement and receiving standard care. The primary outcome was physical activity (as measured with an accelerometer). The secondary outcomes were pain (Visual Analog Scale), quality of life (EQ-5D), and self-efficacy (Arthritis Self-Efficacy Scale, pain and other symptoms subscales). Participants in both groups were evaluated at baseline and after 3 months. The intervention group was also evaluated after 12 months.

Results: No differences were found in the number of minutes spent in sedentary or in physical activity between the intervention and reference groups when comparing the baseline and 3 month follow-up. However, there was a significant difference in mean change (mean diff; 95% CI; significance) between the intervention group and reference group favoring the intervention group with regard to pain (13; 7 to 19; p < 0.001), quality of life (- 0.17; - 0.24 to - 0.10; p < 0.001), self-efficacy/other symptoms (- 5; - 10 to - 0.3; p < 0.04), and self-efficacy/pain (- 7; - 13 to - 2; p < 0.01). Improvements in pain and quality of life in the intervention group persisted at the 12-month follow-up.

Conclusions: Participation in an education and exercise program following the Swedish BOA program neither decreased the average amount of sedentary time nor increased the level of physical activity. However, participation in such a program resulted in decreased pain, increased quality of life, and increased self-efficacy.

https://pubmed.ncbi.nlm.nih.gov/30037339/

 

Good idea!

Unfortunately, I won't be able to contribute much to this thread myself, so any input and discussion much appreciated.

If anyone feels up to have a look at the Cochrane reviews, that would be great, too.

I did a search a while ago on reviews on exercise for fatigue in MS. That was disappointing in terms of asessment of evidence. Hope will find the time to search my old posts.

Also a very quick search on the Cochrane Work Group's reviews was disappointing at a first glance -- but would need to be looked at more thoroughly, too.

https://www.s4me.info/threads/indep...led-by-hilda-bastian.13645/page-6#post-242495

 

I think that illustrates one problem very well:

It seems to me possible that pain and quality of life actually improved although the underlying illness didn't.

Since pain relief and better quality of life matter very much in a chronic illness, any benefit would be great.

Due to not having an adequate control group, though, it seems to be impossible to have even a bit of evidence on whether the secondary (subjective) outcomes actually improved or whether the results show a (short-term) placebo effect or whether they are due to different kinds of bias?

Edited to add the point about the adequate control group.

Would my questions also apply if the trial were adequately controlled?

 

Adding to the post above:

Perhaps also relevant: It could be that the pain improved but it just was not relevant for the tial participants to increase their activity and be less sedentary?

Would need to check if they compared the trial participants' activity levels to that of the patient population in general.

 

I was thinking of randomly selecting a set of say 100 cochrane reviews of therapist delivered interventions, maybe limiting to social and psychological interventions to leave out physiotherapy, to see how the land lies in terms of what outcomes the reviewers choose and if they choose to rely on subjective outcomes in the same way as triallists tend to do compounding the bias in the way the Exercise for CFS review did.

 

Maybe I should add that I hope people also find some trials or reviews we could refer to as good examples.

(cross-posted with Caroline)

 

Yes, indeed. I agree! When I worked at Cochrane I worked with a reviewer who I thought was really good. He is now the coordinating editor of the group I worked for (The pain group - PaPaS). He did a review in around 2010, and updated it a couple of times (after I had left) concluding in the end that despite many more studies being done with each subsequent review update, they were not good quality and so they still couldn't say whether the interventions were effective or not. His review is not a social or psychological intervention, but probably useless anyway! https://www.cochranelibrary.com/cdsr/doi/10.1002/14651858.CD008208.pub5/full#CD008208-abs-0002. (I made a comment on the review, about how I think Cochrane should go further in advocating for better trials). I really rate him as a review author. He said there was a big backlash from all the powerful proponents of this technique, and they ignored the review. Sounds familiar.

 

The reason being?

 

I'm interested in how many reviews have a negative conclusion (the treatment does not work, probably does not work, cannot be assumed to work). I have the suspicion that the conclusion is nearly always positive.

 

The editor Caroline mentioned is an exeption in this regard. Don't have the capacity to post the details now, but have a look at this and other reviews he was involved:

Unfortunately, the most recent proocol he co-authored seems not to bother about the merely subjective outcomes thing and I haven't found out if that actually is what he criticized and led to his conclusions in the previous reviews. It sems to be important to read the protocols.

If anyone could have a closer look at this, that would be great.

Another issue is that you can't do higher quality trials if there are no adequate objective outcomes. No idea if he acknowledged this anywhere.

To objectively measure a more direct impact of arthtritis pain on movement, one perhaps had to measure how / whether patients' movement patterns of the affected joints/ limbs change.

If it comes to outcomes in ME trials it also seems important to me to acknowledge that it could be even easier to develop more objective measurements than for pain.

There are discussions on potential objective measures in ME and also those in MS related fatigue elsewhere on the forum.

Just to leave here some preliminary thoughts:

I think there are several lines of arguments to respond if some people argue that there are no robust measures for ME.

One line of arguing could be that less robust objective measures are better than merely subjective measures and there may exist analytical/ statistical means to make them more robust.

Another one could be to conclude that is not possible to do high quality non-blinded trials on ME until there are better objective measures, and it's possible and needed to develop these.

 

No reason. I would be equally happy to keep them in. Would give better generalisability and not look like we're targeting psychiatrists.

 

Not sure if I entirely agree.

Maybe we need to differentiate between the trial investigators' flawed actions to hide and explain away objective outcomes they didn't like because they didn't support their claims on the one hand, and questions on how robust some of the objective measures actually are, on the other hand?

The first has to be criticized and makes the investigators as well as their results even more untrustworthy than their insufficient trial design per se.

Taking the objective outcomes into account in a sound manner (= not fiddling the figures and not misrepresenting them) also seems to provide evidence that their was no treatment benefit or even a deterioration of symptoms instead.

The second (how robust are the existing objective measures) I think needs to be discussed.

We repeatedly discussed that in mild to moderate ME people can do certain activities at cost of other activities. To a certain extrent, people who are not on the severe spectrum can chose between rest and activity and also different kinds of activity, and influence their symptoms that way. That decision can influence objective measures like activity trackers. And if people are only mildly affected or if they have sufficient support, perhaps some people chose not to do more physically but cognitively?

The objecive measures that were used in some trials surely are better than subjecive measures with regard to reliable evidence.

But do they provide sufficient/ reliable evidence? Or in GRADE terms: evidence of very low, low, moderate or high certainty?

Sorry to not be able to give some examples at the moment. Would need to include references to other threads.

 

Just parking that here -- not checked if this is on the thread about Placebos and/or in the Tack/Struthers/Tuller paper about blinding, just stumbled across that reference in Faulkner (2016) :

Kamper SJ, Apeldoorn AT, Chiarotto A, Smeets RJ, Ostelo RWJG, Guzman J & van Tulder MW (2014) Multidisciplinary biopsychosocial rehabilitation for chronic low back pain. Cochrane Database of Systematic Reviews CD000963. http://www.ncbi.nlm.nih.gov/pubmed/25180773

Faulkner (2016):

 

Again just parking a reference:

https://www.s4me.info/threads/indep...ed-by-hilda-bastian.13645/page-71#post-349789

 

This Cochrane database on physical therapies could be helpful? (*)

https://www.cochrane.de/sites/cochr.../gfb/final_cochrane_fur_die_pt_datenbank.xlsx


And there's that poster by another group:

Physical activity/exercise for health outcomes: an overview of Cochrane systematic reviews | The 26th Cochrane Colloquium

https://colloquium2019.cochrane.org...outcomes-overview-cochrane-systematic-reviews

No idea how to find the corresponding paper though.


(*) From this group (google translate link) :
https://translate.google.com/transl...siotherapie-den-zugang-zu-evidenz-erleichtern

original website: https://wissenwaswirkt.org/cochrane-fuer-die-physiotherapie-den-zugang-zu-evidenz-erleichtern