Screen-And-Treat to Prevent Diabetes Doomed to Fail

Screening must be supplemented by broader public health approaches.

Screen and treat strategies to prevent type 2 diabetes are doomed to failure, according to a large new systematic review and meta-analysis published in The BMJ. 

Instead, the authors said and outside experts agreed, any effort to combat the already enormous and still growing problem of type 2 diabetes will need to target the broad population in addition to individual patients with “pre-diabetes.”

Although there is broad agreement that increasingly large numbers of people are likely to develop diabetes, there is neither a widely-accepted definition of prediabetes nor a consensus that the concept or label itself is helpful or necessary. Various health organizations have broadly divergent definitions of prediabetes. Critics of the concept point to the potential dangers of overtesting, medicalization, and overtreatment associated with the label.

The Findings

Trish Greenhalgh (University of Oxford) and colleagues analyzed data from 49 studies of screening tests and 50 intervention trials in people at elevated risk for developing type 2 diabetes. In the screening portion of the study, they found important deficiencies in both fasting glucose and HbA1c tests, the two main methods to detect diabetes risk.

There was little agreement between the tests in determining which people had pre-diabetes. The most widely used and convenient test, HbA1c, had very low specificity and sensitivity. The fasting glucose test had greater specificity but it did not have sufficient sensitivity to justify its expense and inconvenience.

The findings mean that “large numbers of people will be unnecessarily treated or falsely reassured depending on the test used,” according to the authors.

Therapeutic trials to prevent progression to diabetes showed that some — but by no means most — cases of diabetes could be prevented. Lifestyle intervention trials resulted in a 37% relative risk reduction in developing diabetes, cutting the risk from 239 out of 1,000 in the control group to 151 out of 1,000 in the treatment group during the trial period. But the reduction in risk dropped sharply during follow-up studies. Using the drug metformin resulted in a 26% reduction in relative risk, from 295 out of 1,000 in the control group to 218 out of 1,000 in the treatment group.

The authors warned, however, that treatment studies were mostly of low quality. They further noted that because of the large number of people who either might not meet the eligibility criteria of the trials or who wouldn’t fully participate in the intervention, “there is no scientific basis for extrapolating percentage risk reductions seen in trials to an equivalent reduction in incidence of diabetes across an entire community. Poor enrollment and completion of lifestyle interventions will limit the impact national prevention programs will have on the overall burden of disease.”

The authors recommend that “because of the low accuracy of screening tests and the limited reach of intervention programs, policymakers might want to consider supplementing screen and treat policies with population based approaches aimed at entire communities.”

In an accompanying editorial, Norman Waugh (University of Warwick) agreed that “the risk reductions reported by trials could be much greater than will be seen” in a real-life scenario. He offered support for “public health measures targeted at the whole population at risk,” including taxing sugary beverages and encouraging physical activity by supporting bicycle lanes separate from traffic.

Why Screen and Treat Is Cruel

One expert who has spent a good portion of his career thinking about the complex questions raised by these sort of problems is Victor Montori (Mayo Clinic). Asked to comment on the BMJ study, he agreed that screen and treat can be effective for a few patients but any effort to combat the rise in diabetes will require a “massive” response that goes well beyond screening individual people.

His provocative statement suggested that it will require a remarkable political transformation to achieve success in this area. Here is his full response:

“It is so hard to articulate the issues because there is obvious good in preventing bad things, but let’s give this another go:

• Type 2 diabetes is a bad thing when it reduces the quality of your life, because of its symptoms, complications or the burden of its treatment.
• So preventing diabetes is obviously a good thing.
• The scale of diabetes is huge and the proportion of people who live one step before that diagnosis is very large. (The review shows we will disagree in labeling who exactly is one step closer to the diagnosis depending on what definitions we choose and the ideology behind the definition selection.)
• Individuals who choose to live more actively and eat healthier meals do better and delay diabetes, but they do so by swimming against the current, which explains the very high rates of drop offs and ‘failures.’
• The response should be massive in scale and persistent in time directed at the determinants of the environments, at the environments themselves, and at the lifestyles that emerge as people adapt to those environments. These changes should make healthier lifestyles the easy default —= the direction of the current that drags those who are and are not interested in swimming.
• Screen and treat is a clinical response, individual, one-at-a-time. It seems ideally suited to people who already are chronic patients by virtue of their comorbidities and thus are already in the healthcare system as it requires the resources of the healthcare system for its success. However, any clinical success leaves the determinants of the environments and the environments unchanged, guaranteeing a steady stream of candidates for screen and treat forever. Furthermore, patients with prediabetes who ‘fail’ to improve with lifestyle interventions may be considered candidates for diabetes drugs like metformin – in essence they are preventing the diagnosis of diabetes by ensuring they get treatment for diabetes instead— a lousy proposition.
• Meanwhile people bemoan the low quality of treatment of type 2 diabetes, in part because of lack of time, training, and resources. These are lacks from the same system we are ready to load with people who screen positive for prediabetes. And since the epidemic hits the underserved hardest (suggesting again problems with the contexts in which people try to make a living rather than a massive epidemic of poor judgment among the poor and socioeconomically distressed) and these folks have trouble getting healthcare in the first place, a solution reliant on healthcare access, if effective, would make disparities in the incidence of diabetes worse.
• Thus, we need solutions that don’t leave the conditions that have created the epidemic intact, making the efforts of those set on improving their lifestyle often seem futile in the long run, producing more at-risk people, burdening the sick-care system with healthy people seeking wellness. In all these ways, policies of screen and treat are accidentally (I hope) cruel, particularly toward the sick and the needy, people living ‘in the shadows of life.’
• I wholeheartedly endorse the priority of preventing type 2 diabetes, but effective sustainable solutions are more likely to be found through evidence-informed deliberative democracy (the population version of shared decision making). The work there is to determine the kind of environments we want — for ourselves and our children — and the public health policies that must be implemented to realize them.
• Those who seek a more expedient solution to match the urgency of the problem would do best to start this long-term process as soon as possible rather than waste time, attention, and resources, in palliating the problem one screen-and-treated patient at a time.”