What a terribly uninformative paper. This paper says two things:
medical professionals have different opinions on different kinds of people.
treating people differently may affect healthcare outcomes.
There are no statistics, no controls, no correlations presented, but they heeeeeavily imply a causative connection despite never showing one.
Professionally, I definitely have opinions between one type of clients and another, because one type will almost always follow instructions and the other is a coin flip. They both get the same quality of work, but if you asked me honestly which I would rather have, it would be the former.
In a study such as this, I would show a strong “bias”, because several decades of experience have shown that one particular type of client doesn’t give a shit about safety rules and laws.
Ironically, that particular type of client would show up again in the second part, because they predictably also have much worse outcomes after I’ve done my work (because they didn’t actually do it).
And this, bots and droids, is why you put some controls in your paper.
I don’t know if you have actually read the paper, but this is a review on existing information gathered from other papers. Many sources have been mentioned, of which contain the things necessary to reach such conclusion. Only one such mentioned papers is https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/oby.20637 that talks about bias in medical field students against obese patients in statistics.
If we remain to the context of body size, judging a person’s willingness to follow healthcare only by their body type is indeed discriminatory. In a field such as public health you cannot cherry pick who you treat, and reaching the maximum amounts of positive outcomes is essential and what modern medicine aims to do on paper. You just cannot expect everyone to maintain the same level of professionalism and quality when dealing with something they disagree with.
You can absolutely do a literature review and include relevant statistics. That’s a very common thing to do in pretty much every field. That’s why this is so strikingly weird, but that’s probably because this paper is literature reviews crammed into a single article, hoping that proximity might create a casual connection.
one such mentioned papers is
Yes, that says “doctors have different feelings about obese people”. It says nothing else. And that’s my point.
judging a person’s willingness to follow healthcare only by their body type is indeed discriminatory
No. Giving that person a different and worse level of treatment for unfounded reasons is discriminatory.
That’s something this paper completely skips over. It reports the doctor’s feelings, it reports the patient’s outcomes, waves it’s arms and goes “seeee, this might be casual! It really probably might be. It might cause it!”
This is “icecream sales cause shark attacks” levels of science.
If no new data is available to show during a review, there is no reason to force it in. It would be an exercise in futility to gather new data if data already available reaches a certain conclusion, unless that original data is either found insufficient or have been collected or filtered with a conflict of interests.
“doctors have different feelings about obese people”.
…and a good number don’t feel prepared to treat them and might feel frustrated doing so. This is not the fault of obese individuals.
You cannot expect everyone to act with the same level of professionalism when finding themself in front of someone they are biased on. That’s just not how humans work.
Giving that person a different and worse level of treatment for unfounded reasons is discriminatory.
Warning: Long rant incoming about actually backing your sources with data instead of narrative.
A review doesn’t create new data, but it should actually use the existing data. This paper basically goes “People feel more negative to BMW drivers. BMW drivers suffer worse outcomes from car crashes. This might be the fault of bias against them. Here are some ways you can stop bias”.
There are 4 statements there, and plenty of evidence for number 1, 2 and 4, and absolutely none of the actual “conclusion” in statement number 3. And I use this example because there is a very simple fact that causes both 1 and 2, namely that BMW drivers do in fact create more and more dangerous accidents because they engage in more dangerous behaviour.
I don’t deny that doctors feel different about obese people, because the linked papers say so, pretty clearly.
What I deny is that the paper shows that those feelings are unwarranted, because the paper doesn’t show any of that. NONE of the cited papers mention it either, and yes, I read ALL the abstracts (not the entire paper, granted). There is a massive gap in data, that is filled in with “may”, “could” and “might”.
I also deny that the conclusion exists, because the paper doesn’t actually show that A causes B, just that both exist. Hence: Icecream sales increase shark attacks.
From the article:
There are several mechanisms by which provider attitudes may affect the quality of, or potential for, patient-centred care.
Emphasis mine. Yes, it definitely may, but the paper doesn’t show that it DOES. You’ll find that word quite a lot in the paper.
A common explicitly endorsed provider stereotype about patients with obesity is that they are less likely to be adherent to treatment or self-care recommendations (23,24,55,56).
Well, are they? I don’t know, the paper doesn’t actually say. The cited papers also don’t actually say. A common stereotype about Dutch people is that they’re tall, the stereotype exists because, in fact, Dutch people are really tall. If you were to say “I think Dutch people are tall”, that’s not a bias, that’s a reality.
There’s also a fair bit of selective quoting going on, for example:
For example, in one study of primary care providers randomly assigned to evaluate the records of patients who were either obese or normal weight, providers who evaluated patients who were obese were more likely to rate the encounter as a waste of time and indicated that they would spend 28% less time with the patient compared with those who evaluated normal-weight patients (59)
When you read the actual paper, it says that the providers would LIKE TO spend less time on them. Not that they don’t, just that they would prefer. There is no actual data presented on time spent with patients.
Finally, physicians **may **over-attribute symptoms and problems to obesity, and fail to refer the patient for diagnostic testing or to consider treatment options beyond advising the patient to lose weight. In one study involving medical students, virtual patients with shortness of breath were more likely to receive lifestyle change recommendations if they were obese (54% vs. 13%), and more likely to receive medication to manage symptoms if they were normal weight (23% vs. 5%) (23).
Wow, that really sounds like doctors are treating obese patients much more than normal-weight patient. Good thing the actual study say this:
Analyses revealed more negative stereotyping, less anticipated patient adherence, worse perceived health, more responsibility attributed for potentially weight-related presenting complaints and less visual contact directed toward the obese version of a virtual patient than the non-obese version of the patient.
And I’ll note that neither the review nor the cited paper mentions ANYTHING about comparing anticipated patient adherence with actual adherence. The cited paper specifically mentions that doctor’s had a worse percieved health of obese patients, because they used identical patients, except one was obese. So, one patient with 5 medical conditions, and the other has 6. If you then say “Huh, maybe this patient with a bunch of conditions AND obesity is in worse health”, they rate that as negative stereotyping. I think that we can all agree that a patient with two broken bones SHOULD BE percieved as having worse health than one with a single broken brone.
In contrast, there was no clear evidence of bias in clinical recommendations made for the patient’s care.
Emphasis mine. This cited paper literally says the opposite of the review. I can only conclude reviewers didn’t actually read their own citations.
Impact on patients
This chapter is actually amazing. NONE of the cited papers actually discuss impact on patients as a result of obesity-related discrimination. They all fall into one of four types: “Obese patients do things that are bad for them”, “stress is bad for you”, “Racist doctors are bad for patients”, “Obese people dislike what they recieved”.
Note that the third group is actually good science. Those three papers (67, 68, 69) are exactly the sort of paper that should be used here. If only they were actually about the topic being discussed here, and not something tangentially related and worked into a story without actual support.
To summarize: This paper does NOT support that any “Impact of weight bias and stigma on quality of care and outcomes for patients with obesity” exists, at all. It merely says that it MAY exist. It also doesn’t show that bias is unfounded. It does, however, cite at least one paper that shows the exact opposite.
Thus, pulmonary inflammation could play a role in cancer initiation or promotion.
Definitive statements are almost never done in public health papers because not everything will end in the same result. There is no use to put so much weight on those terms because that’s how open research works.
–
A common explicitly endorsed provider stereotype about patients with obesity is that they are less likely to be adherent to treatment or self-care recommendations (23,24,55,56)
Well, are they? I don’t know, the paper doesn’t actually say.
23,24,55, and 56 all say this.
“I think Dutch people are tall”, that’s not a bias, that’s a reality.
For example, in one study of primary care providers randomly assigned to evaluate the records of patients who were either obese or normal weight, providers who evaluated patients who were obese were more likely to rate the encounter as a waste of time and indicated that they would spend 28% less time with the patient compared with those who evaluated normal-weight patients (59)
When you read the actual paper, it says that the providers would RATHER spend less time on them. Not that they don’t, just that they would prefer.
Directly from the paper’s extract:
…though physicians prescribed more tests for heavier patients, F(2, 107) ¼ 3.65, P < 0.03, they simultaneously indicated that they would spend less time with them, F(2, 107) ¼ 8.38, P < 0.001, and viewed them significantly more negatively on 12 of the 13 indices
I don’t see where you found the preference of spending less time, when the same words that the 2001 paper used were reported on the 2015 one too. This is the full paper of 59: https://www.mikkihebl.com/uploads/9/0/2/3/90238177/8.pdf
From the full paper:
The results of the Patient follow-up questionnaire provided support for the notion that physicians viewed and responded to patients differently depending on their weight. The heavier the patients were, the more negative the attitudes and the distancing behaviors were. Such patterns can
be observed from examining the means, the Fs from ANOVAS, and the ts from the linear trend analyses, all of which are presented in Table 2. In particular, physicians reported that they would spend significantly less time with patients the heavier they were (M ¼ 31.13 min with average-weight patients, M ¼ 25.00 min with moderately overweight patients, and M ¼ 22.14 min with severely overweight patients).
Wow, that really sounds like doctors are treating obese patients much more than normal-weight patient.
This is the Impact On Providers sections. Yes, they are and this is not contested. This paper is not about refusal to give prescriptions.
There is evidence that providers’ communication is less patient-centred with members of stigmatized racial groups (37–43), and other stigmatized groups including patients with obesity (44), and that provider attitudes contribute to this disparity (45–47). Implicit attitudes have also been found to be associated with lower patient ratings of care (46).
This is not about prescriptions, but communication. Exactly what the paper is talking about. 53 supports the importance of communication in outcomes and quality, too. The fact that it is not centered about obesity specifically is irrelevant, especially because it doesn’t specify any difference between different issues.
This is also sustained by the extract:
Stigma can reduce the quality of care for patients with obesity despite the best intentions of healthcare providers to provide high-quality care.
Impact on patients
To be honest I have no idea what you are saying here. If a person felt embarrassed, out of place, or misunderstood or fears any of those to be present that is lower quality of care. And a person postponing or refusing checkup/treatment due to what they feel about following trough is a negative outcome.
I don’t think they read it or else they’d see it’s a literature review. Those don’t tend to have “controls”, they instead summarize multiple studies on the sane topic to see if anything interesting come up.
Literature reviews don’t generally work by grabbing two seperate topics, and placing them in close proximity to eachother so you can them claim “This MIGHT influence that”. This paper has a qualifier in every single sentence of the abstract that talks about conclusions.
Show that it DOES, leave the speculation for the discussion portion.
Fat shaming destroys lives
https://pmc.ncbi.nlm.nih.gov/articles/PMC4381543/
What a terribly uninformative paper. This paper says two things:
medical professionals have different opinions on different kinds of people.
treating people differently may affect healthcare outcomes.
There are no statistics, no controls, no correlations presented, but they heeeeeavily imply a causative connection despite never showing one.
Professionally, I definitely have opinions between one type of clients and another, because one type will almost always follow instructions and the other is a coin flip. They both get the same quality of work, but if you asked me honestly which I would rather have, it would be the former.
In a study such as this, I would show a strong “bias”, because several decades of experience have shown that one particular type of client doesn’t give a shit about safety rules and laws.
Ironically, that particular type of client would show up again in the second part, because they predictably also have much worse outcomes after I’ve done my work (because they didn’t actually do it).
And this, bots and droids, is why you put some controls in your paper.
What?
I don’t know if you have actually read the paper, but this is a review on existing information gathered from other papers. Many sources have been mentioned, of which contain the things necessary to reach such conclusion. Only one such mentioned papers is https://onlinelibrary.wiley.com/doi/pdfdirect/10.1002/oby.20637 that talks about bias in medical field students against obese patients in statistics.
If we remain to the context of body size, judging a person’s willingness to follow healthcare only by their body type is indeed discriminatory. In a field such as public health you cannot cherry pick who you treat, and reaching the maximum amounts of positive outcomes is essential and what modern medicine aims to do on paper. You just cannot expect everyone to maintain the same level of professionalism and quality when dealing with something they disagree with.
The patients values and wishes have to always be respected. https://pmc.ncbi.nlm.nih.gov/articles/PMC9677014/
You can absolutely do a literature review and include relevant statistics. That’s a very common thing to do in pretty much every field. That’s why this is so strikingly weird, but that’s probably because this paper is literature reviews crammed into a single article, hoping that proximity might create a casual connection.
Yes, that says “doctors have different feelings about obese people”. It says nothing else. And that’s my point.
No. Giving that person a different and worse level of treatment for unfounded reasons is discriminatory.
That’s something this paper completely skips over. It reports the doctor’s feelings, it reports the patient’s outcomes, waves it’s arms and goes “seeee, this might be casual! It really probably might be. It might cause it!”
This is “icecream sales cause shark attacks” levels of science.
If no new data is available to show during a review, there is no reason to force it in. It would be an exercise in futility to gather new data if data already available reaches a certain conclusion, unless that original data is either found insufficient or have been collected or filtered with a conflict of interests.
…and a good number don’t feel prepared to treat them and might feel frustrated doing so. This is not the fault of obese individuals. You cannot expect everyone to act with the same level of professionalism when finding themself in front of someone they are biased on. That’s just not how humans work.
And those unfounded reasons are created from bias. https://dictionary.cambridge.org/dictionary/english/bias
Warning: Long rant incoming about actually backing your sources with data instead of narrative.
A review doesn’t create new data, but it should actually use the existing data. This paper basically goes “People feel more negative to BMW drivers. BMW drivers suffer worse outcomes from car crashes. This might be the fault of bias against them. Here are some ways you can stop bias”.
There are 4 statements there, and plenty of evidence for number 1, 2 and 4, and absolutely none of the actual “conclusion” in statement number 3. And I use this example because there is a very simple fact that causes both 1 and 2, namely that BMW drivers do in fact create more and more dangerous accidents because they engage in more dangerous behaviour.
I don’t deny that doctors feel different about obese people, because the linked papers say so, pretty clearly.
What I deny is that the paper shows that those feelings are unwarranted, because the paper doesn’t show any of that. NONE of the cited papers mention it either, and yes, I read ALL the abstracts (not the entire paper, granted). There is a massive gap in data, that is filled in with “may”, “could” and “might”.
I also deny that the conclusion exists, because the paper doesn’t actually show that A causes B, just that both exist. Hence: Icecream sales increase shark attacks.
From the article:
Emphasis mine. Yes, it definitely may, but the paper doesn’t show that it DOES. You’ll find that word quite a lot in the paper.
Well, are they? I don’t know, the paper doesn’t actually say. The cited papers also don’t actually say. A common stereotype about Dutch people is that they’re tall, the stereotype exists because, in fact, Dutch people are really tall. If you were to say “I think Dutch people are tall”, that’s not a bias, that’s a reality.
There’s also a fair bit of selective quoting going on, for example:
When you read the actual paper, it says that the providers would LIKE TO spend less time on them. Not that they don’t, just that they would prefer. There is no actual data presented on time spent with patients.
Wow, that really sounds like doctors are treating obese patients much more than normal-weight patient. Good thing the actual study say this:
And I’ll note that neither the review nor the cited paper mentions ANYTHING about comparing anticipated patient adherence with actual adherence. The cited paper specifically mentions that doctor’s had a worse percieved health of obese patients, because they used identical patients, except one was obese. So, one patient with 5 medical conditions, and the other has 6. If you then say “Huh, maybe this patient with a bunch of conditions AND obesity is in worse health”, they rate that as negative stereotyping. I think that we can all agree that a patient with two broken bones SHOULD BE percieved as having worse health than one with a single broken brone.
Emphasis mine. This cited paper literally says the opposite of the review. I can only conclude reviewers didn’t actually read their own citations.
This chapter is actually amazing. NONE of the cited papers actually discuss impact on patients as a result of obesity-related discrimination. They all fall into one of four types: “Obese patients do things that are bad for them”, “stress is bad for you”, “Racist doctors are bad for patients”, “Obese people dislike what they recieved”.
Note that the third group is actually good science. Those three papers (67, 68, 69) are exactly the sort of paper that should be used here. If only they were actually about the topic being discussed here, and not something tangentially related and worked into a story without actual support.
To summarize: This paper does NOT support that any “Impact of weight bias and stigma on quality of care and outcomes for patients with obesity” exists, at all. It merely says that it MAY exist. It also doesn’t show that bias is unfounded. It does, however, cite at least one paper that shows the exact opposite.
From a paper about smoking and lung cancer:
Definitive statements are almost never done in public health papers because not everything will end in the same result. There is no use to put so much weight on those terms because that’s how open research works.
–
23,24,55, and 56 all say this.
No. That’s a stereotype, that is a perceived generalization. https://dictionary.cambridge.org/dictionary/english/stereotype . Stereotype is not bias, but bias can be born from a stereotype.
Directly from the paper’s extract:
I don’t see where you found the preference of spending less time, when the same words that the 2001 paper used were reported on the 2015 one too. This is the full paper of 59: https://www.mikkihebl.com/uploads/9/0/2/3/90238177/8.pdf
From the full paper:
This is the Impact On Providers sections. Yes, they are and this is not contested. This paper is not about refusal to give prescriptions.
This is not about prescriptions, but communication. Exactly what the paper is talking about. 53 supports the importance of communication in outcomes and quality, too. The fact that it is not centered about obesity specifically is irrelevant, especially because it doesn’t specify any difference between different issues.
This is also sustained by the extract:
To be honest I have no idea what you are saying here. If a person felt embarrassed, out of place, or misunderstood or fears any of those to be present that is lower quality of care. And a person postponing or refusing checkup/treatment due to what they feel about following trough is a negative outcome.
I don’t think they read it or else they’d see it’s a literature review. Those don’t tend to have “controls”, they instead summarize multiple studies on the sane topic to see if anything interesting come up.
Literature reviews don’t generally work by grabbing two seperate topics, and placing them in close proximity to eachother so you can them claim “This MIGHT influence that”. This paper has a qualifier in every single sentence of the abstract that talks about conclusions.
Show that it DOES, leave the speculation for the discussion portion.