Inaccuracy in my blog on diabetes screening??

I just received a comment from the USPSTF (see below) suggesting that there was an inaccuracy in how their diabetes screening recommendation was presented in my recent blog http://gmodestmedblogs.blogspot.com/2016/07/uspstf-diabetes-screening-misses-most.html  . This blog reviewed a PLoS article which found that a majority of patients with dysglycemia (glucose intolerance or diabetes) would be missed if we followed the USPSTF recommendations. The gist of the USPSTF concern was that the original PLoS article and I only included their formal  recommendation summary, which indeed was included accurately (ie, "The USPSTF recommends screening for abnormal blood glucose as part of cardiovascular risk assessment in adults aged 40 to 70 years who are overweight or obese", and was given a grade "B" recommendation), but not the full content of their recommendation as outlined in their text.

In general, when I get significant feedback on my blogs,  I send that around but do not comment further: I want to encourage people to make their comments, so I have printed their comments in full and do not want the responder to feel undercut by my further comments. But in this case, I will respond as a means to articulate my intentions overall in writing these blogs:

it is true that buried in the USPSTF text is their additional comment "Persons who have a family history of diabetes, have a history of gestational diabetes or polycystic ovarian syndrome, or are members of certain racial/ethnic groups (that is, African Americans, American Indians or Alaskan Natives, Asian Americans, Hispanics or Latinos, or Native Hawaiians or Pacific Islanders) may be at increased risk for diabetes at a younger age or at a lower body mass index. Clinicians should consider screening earlier in persons with 1 or more of these characteristics." (my emphasis). This comment did not have a letter rating indicating their assessment of its importance/generalizability, nor further clarification of what it means to consider screening. I would like to make the following observations:

    --the formal USPSTF formal recommendations (in this case: to screen people 40-70 who are overweight/obese) are the ones essentially universally cited in applying USPSTF recommendations to clinical practice, and they are the ones used by insurance companies when they determine which preventive services to cover. It is also the only part of their recommendation that receives their letter grade indicating the relative importance of the recommendation.

    ​--the point of the PLoS article and my critique was that applying this USPSTF formal recommendation would miss a very large number of people with dysglycemia, I think a rather important shortcoming of their formal recommendation (and the reason i did the blog).

    --I did, in fact (and as per usual) read the full USPSTF report and felt that their additional comment that we should consider screening some younger than age 40 to be neither helpful nor practicable (what does consider mean? seems to be a rather indefinite term left to the quite variable discretion of the clinician). and what does “earlier” means. At birth? Age 5? Age 25?.... If they wanted to highlight this option, perhaps in their formal recommendation they could have stated:  "though there is not enough evidence at this point, we urge clinicians to consider screening patients earlier if there are 1 or more risk factors since diabetes is so prevalent, with so much associated morbidity/mortality, and the risks of screening are so small....".

    --and, I think that an organization such as the USPSTF really should be welcoming of the PLoS study, as it provides some evidence that their recommendations might be broadened.​ Their response seemed to me to be rather strongly put, a tad defensive, and did not reflect how we (and insurance companies) generally use their recommendations by relying on their formal recommendations with their graded rating systems, and not by scouring the detailed text for other ungraded comments.

I should also add that I really do try to look at the cited articles in authors’ references when they make assertions that I do not know to be reasonable/true. and, honestly, it is a bit shocking the number of times that a reference does not confirm the authors' assertions. For example, I recently sent out recommendations on HIV treatment (see http://gmodestmedblogs.blogspot.com/2016/07/2016-hiv-treatment-guidelines.html ​ ), which included the comment that MAC prophylaxis was not necessary in this era of better HIV drugs. My concern is that some patients with very low CD4 counts may not increase above 50 with meds to make MAC prophylaxis unnecessary. So, I looked at the cited article, and in fact there were no data to suggest that such people do not need prophylaxis. My blog migrated to Paul Sax (whose opinions I really do value); he makes it clear he does not think MAC prophylaxis is necessary. However, on reading his blog, in fact he does state that it might be reasonable to consider MAC prophylaxis in those who did not get a significant bump in their CD4 counts (ie, he acknowledges this is still a reasonable concern).

so, I bring this up mostly to assure you that I do take these blogs seriously, I do read the full articles and even check occasional references cited, and I try to put the articles (hence my critique/commentary) into what seems to me to be a primary care perspective for practicing clinicians ("should this change our practices?", "do these guidelines from the experts make sense to apply to the patients we see in our primary care practices?", “what are the biases of the authors or methodology which limit their applicability in primary care?”, etc)

thanks.  geoff

---

​My name is Lauren and I am a media contact for the U.S. Preventive Services Task Force. I read your blog post titled "Primary Care Corner with Geoffrey Modest MD: USPSTF Diabetes Screening Misses Most People" and was hoping we could work together to clarify the Task Force's recommendation statement on Screening for Abnormal Blood Glucose and Type 2 Diabetes Mellitus.

Unfortunately, the study that was referenced in this article only looked at the topline of the recommendation statement, missing a key clinical consideration. In the recommendation, the Task Force did recognize that “persons who have a family history of diabetes, have a history of gestational diabetes or polycystic ovarian syndrome, or are members of certain racial/ethnic groups (that is, African Americans, American Indians or Alaskan Natives, Asian Americans, Hispanics or Latinos, or Native Hawaiians or Pacific Islanders) may be at increased risk for diabetes at a younger age or at a lower body mass index.” In the recommendation statement, the Task Force encourages clinicians to consider screening earlier in persons with 1 or more of these characteristics.

So, it is misleading when the study findings suggest that the Task Force did not consider racial ethnicity in their criteria for screening.

All of this being said, I think that we could make a quick and easy addition to this article to more accurately reflect the Task Force’s position. I think the most opportune place to do this is up front, by changing the first line "A study looked at the sensitivity/specificity of the current USPSTF guidelines for diabetes screening” to “In a narrow view of the clinical considerations in the Task Force’s recommendation, this study looked