Everything We Ought to Know About the USPSTF

Clinical Preventive Service Recommendations

Clinical preventive services are hugely important in primary care. With dozens of available preventive services designed to keep patients healthy, and limitations of cost and time, how do clinicians decide which services to offer patients? How do patients ascertain that a preventive service has value for them?

Such decisions are greatly aided by recommendations from experts and groups who examine the evidence and provide recommendations to clinicians. One such group is the US Preventive Services Task Force (USPSTF), an independent, volunteer panel of national experts in prevention and evidence-based medicine. The Task Force makes recommendations to help primary care clinicians and patients decide together whether a preventive service is right for a patient's needs.

Just because a service is labeled "preventive" doesn't mean it's beneficial, or that it poses no risks or harms to the patient. Consideration of both the benefits and harms of a preventive service is therefore necessary before informed decisions can be made about the value of that service.

One of the challenges in the clinical preventive services landscape is that recommendations issued by key expert panels can differ, resulting in confusion for clinicians as well as patients. Another challenge is that the available evidence may have gaps that impede the development of evidence-based recommendations.

To explore these and other issues that arise in the course of developing clinical preventive service recommendations, and to help clinicians understand and best apply the recommendation statements issued by the USPSTF, Medscape spoke with the chair of the USPSTF, David C. Grossman, MD, MPH

The Who, What, and How of the USPSTF

Medscape: Most people in the healthcare field have heard of the USPSTF—usually in connection with a new guideline—but may not know much about the Task Force itself. Can you tell us about the mission and scope of the Task Force's work? What are your chief aims?

Dr Grossman: The Task Force was created in 1984, and its purpose has changed very little since then. Our mission is to improve the health of the US public by providing recommendations on clinical preventive services for well people. The targets of these recommendations are primary care physicians, nurse practitioners, and other types of providers in primary care.

Our focus is on well people—people who do not have signs or symptoms of a disease. We do not get into "tertiary prevention," meaning chronic disease management or prevention of complications from chronic disease.

Our Task Force is composed of 16 primary care leaders, who are experts in evidence-based medicine. They come from a variety of specialties, including pediatrics, family medicine, internal medicine, women's health, and nursing. We have always had a nurse on the task force, typically a nurse practitioner.

I should point out that we don't refer to our work product as "guidelines." We call them "recommendation statements."

Medscape: Who supports the work of the Task Force? Where does your funding come from?

Dr Grossman: The Task Force is supported by the Agency for Healthcare Research and Quality (AHRQ), which is part of the Department of Health and Human Services. The Task Force members are all volunteers. However, AHRQ plays a pivotal role in our ability to do our work. First, the agency funds the evidence-based practice centers that conduct the systematic evidence reviews that are used to inform our recommendations. AHRQ also has a scientific team of superb medical officers who are experts in preventive medicine, and they assist the Task Force work groups and ensure that the processes that we use are consistent.

Medscape: Does AHRQ or any other agency have input into your recommendations? Must they "sign off" on your guidelines?

Dr Grossman: No. The Task Force is independent—we are neither permanent nor temporary government employees. The USPSTF is not a government task force, nor is it a government advisory committee. It is an independent task force that speaks directly to the US primary care community as well as increasingly to the public, as a result of an increased interest in the Task Force's activities.

Our recommendations are not reviewed or approved by any agency of the US government, including AHRQ. They support us in terms of our ability to draft the recommendations, but they don't have influence over what we recommend.

The Task Force has a very stringent financial conflict of interest policy. We look for new Task Force members who have very few financial conflicts of interest. When we begin work on a recommendation statement, we ask each member whether they have any specific conflicts related to that topic, and we make a judgment about whether they can be included in the work group, participate in discussions, or even be allowed to vote. That is a carefully monitored process to ensure that conflicts of interest do not influence our work.

The Recommendations Development Process

Medscape: How does the Task Force decide when a new recommendation statement is needed?

Dr Grossman: The Task Force has three standing work groups—a topic prioritization committee, a methods work group, and a dissemination and implementation work group. The topic prioritization work group constantly reviews our library of topics and helps the Task Force decide which are in greatest need of updating.

As a general rule, we strive to update our topics every 5 years. Some topics may be given lower priority because the evidence in the literature has not changed much, and an update in 5 years might not be appropriate. Conversely, we might conduct an early update on a topic if important new evidence emerges in that area.

We typically develop one or two new recommendation statements annually. Screening for latent tuberculosis, screening for sleep apnea, and autism screening are some new topics we have tackled recently.

We also have an open nominations process for new recommendations, located on our website. Anyone in the United States can nominate a topic, and all of those nominations are reviewed by the topic prioritization work group to decide whether they are in scope and appropriate for new recommendation statements. Our methods manual outlines explicit criteria for what constitutes an ideal new topic. (Note: you can nominate a new topic here.)

Medscape: What goes into the development of a new recommendation statement? In particular, can you tell us about the stages during which clinicians can have input into the recommendations?

Dr Grossman: The first step in the life cycle for either a new topic or an update is the development of a research plan, which provides both the analytic framework and the key questions that we're trying to address. Nothing is more important than making sure you have the questions framed correctly and that you're asking the right questions.

That framework and research plan are posted for public review and comment, so that we can get feedback. We also reach out to stakeholder organizations and professional specialty societies that might have an interest in this area, to actively solicit their feedback.

After we gather that public feedback on our research plan, the evidence-based practice center conducts the systematic evidence review. That takes 12-14 months, and when it is completed, the topic work group convenes to review the evidence and look at each key question to find out whether the evidence is sufficient to make a recommendation. If so, then they begin to look into what the potential net benefit and harms might be of providing that preventive service.

The topic work group is a subset of the Task Force—typically it's three or four primary care members who have the most familiarity with the topic. For example, if it's a pediatric topic, that group might be composed of pediatricians and family physicians. A geriatric topic might include a geriatrician as well as other adult providers. We also actively engage subspecialists and expert consultants in these domains.

Once that topic work group has decided on a draft recommendation, they bring it forward to the full Task Force for a vote. We then post the draft recommendation statement on our website for public comment for 1 month, a process we started in 2010. It's been very valuable, because it's an opportunity to see whether we got the evidence right. Did we miss any evidence? And are we communicating clearly? We get lots of feedback on how we are communicating, and as a result almost always end up revising the draft statement. We rarely change the grading of the recommendation, but we commonly reword parts of the recommendation statements on the basis of feedback.

Figure 1 illustrates the main steps in the recommendations development process. Other USPSTF resources are available here.

Figure 1. US Preventive Services Task Force recommendations development process. Republished with permission of the US Preventive Services Task Force; Rockville, Maryland

Medscape: Does posting the draft recommendations ever cause confusion among clinicians or the public, who might mistake them for final recommendations?

Dr Grossman: Our process of transparency has greatly benefited the Task Force, and has been helpful in managing the concerns of stakeholders who might otherwise feel that they didn't have any input. But members of the media, who take great interest in and often report on our recommendation statements, sometimes make it sound as though the recommendations are final, when in fact they are still at the draft stage and likely to undergo further revision. Clinicians and others who read those news reports might not be aware that they aren't final. The draft recommendations are taken down once we have issued the final recommendation statement.

Balancing the Benefits and Harms

Medscape: How does the grading system work?

Dr Grossman: Figure 2 illustrates our grading system. The grades are A through D, plus the I statement, which means that the current evidence is insufficient to assess the balance of benefits and harms of the service.

Figure 2. The US Preventive Services Task Force grading system: definitions. Republished with permission of the US Preventive Services Task Force; Rockville, Maryland.

Medscape: You mentioned that you assess the potential net benefit of a preventive service recommendation. What is "net benefit"? Is cost a consideration in terms of benefits or harms?

Dr Grossman: The net benefit is benefits minus harms, and represents the overall potential benefits of the preventive service to the average asymptomatic individual. We look at what the evidence says about potential harms of providing that service. Then, we look at the magnitude of those benefits and harms.

It's a qualitative judgment. We assign a grade of A through D, on the basis of the certainty of a net benefit. For example, if a preventive screening has only small benefits but poses major harm, the net benefit of that screening will be negative. If the screening has large benefits and small potential harms, then the net benefit will be positive.

We don't factor cost at all into our decisions—cost is not a consideration in our grading or construction of recommendations. Benefits and harms are the only variables that go into a grade. We have a grid on our website that shows the three levels of certainty (low, moderate, high) about the evidence used to determine net benefit.

We often hear someone say, "You didn't use this study," or "What about that study?" We rate studies for quality, and poor-quality studies aren't used in making our recommendations. We look at the strength of the evidence and the quality of the research that has been done. When we look at why we didn't use a particular study, we often find that we did consider it, but it was a deeply flawed study, so it was excluded.

Medscape: How are the final recommendation statements communicated to the primary care community?

Dr Grossman: All of our recommendations, as well as the evidence reviews on which they are based, are published on the USPSTF website, and 99% of our final recommendations are also published in a peer-reviewed or partner journal. Since 2016, that journal partner has been JAMA, which uses not only its main journal, but also its network of journals, to disseminate our recommendation statements.

They also typically invite one or more experts to write an editorial, which is published alongside the recommendation statement and evidence review. These editorials are independent commentaries on the implications of the recommendation statement.

Uptake and Implementation

Medscape: So, now you have a final recommendation statement posted on your website and published in JAMA. How should clinicians interpret the new recommendations?

Dr Grossman: The fact that we have issued a recommendation statement at all means that there was sufficient evidence to recommend for or against that preventive service. Within the statement, the grading system is designed to help clinicians judge whether to offer that preventive service to their patients.

In general, clinicians can view an A or a B grade similarly. An A grade means that we have more substantial evidence to support it, but both an A and a B grade mean "do it."

A grade of C means that those services should be offered selectively, on an individual basis. The magnitude of benefit of a C recommendation is smaller compared with an A or B recommendation. Some clinicians may choose to prioritize A and B recommendations, but we think that C recommendations should also be brought to the attention of patients.

A recommendation of D means that the clinician should not offer that service. Finally, an I statement means that the evidence was insufficient, and therefore the Task Force cannot recommend for or against the service.

Medscape: Does the USPSTF make suggestions about how to implement the recommendations?

Dr Grossman: Our recommendations should be used in well care for adults, children, or pregnant women as part of routine practice. We have many different tools to help with implementation.

Each recommendation statement has a section called "clinical considerations." It's important for clinicians to read beyond just the very top line of the recommendation statement, because a lot of the nuance and how it should be implemented is found in the clinical considerations section. For those who are really interested in the topic and want to dive deep, there is a discussion section that goes through the evidence supporting each recommendation.

We also have a mobile app called ePSS (Electronic Preventive Services Selector). It's a really cool little app, available in the iTunes library. The app asks for a few details about the individual patient—age; sex; and whether the patient is pregnant, has ever used tobacco, and is sexually active. On the basis of those details, you get a list of recommendations that apply to that patient. The app also contains useful tools and helpful information about screening, such as screening tools for depression or alcohol use. It provides body mass index calculators and other tools to help clinicians implement the recommendations in practice.

Some electronic health records have similar tools embedded in the patient records. Our mobile app is also useful for patients who want to know what preventive services they should be receiving.

Medscape: How do you allow for the role of patient choice and clinician judgment within your recommendations?

Dr Grossman: Although our recommendations are population-based, every patient represents an individual circumstance, and there may be compelling cases in which our recommendation statement doesn't apply to a given individual. Ultimately, the clinician has to be the quarterback to decide the applicability of our recommendations to his or her population of patients as well as to each individual patient. For some recommendations (those with an A grade) the evidence is very compelling that there is net benefit, so we would strongly encourage clinicians to offer that service to every single patient.

For other recommendations, it's a much closer call. Those are typically our C recommendations. For example, before conducting mammography screening in women between age 40 and 50 years, clinicians should have a discussion with those patients and determine their values and risk factors that might alter their decision of whether to be screened. What would ease the patient's concerns and fears about being screened, and what is the patient's willingness and tolerance to deal with a false-positive test? That kind of shared decision-making is most common with C-grade recommendations.

Medscape: Do you have any way of knowing how widely your recommendations are applied in practice, or whether clinicians are running into barriers in implementing them?

Dr Grossman: For the most part, we don't have any monitoring mechanism to understand how our recommendation statements are implemented in practice. Although we don't have a specific process for that, sometimes we get that information through the comment and feedback periods when clinicians write, "Here's where I have problems or issues." Ideally, the comment and feedback process is an opportunity for us to simplify the recommendations.

The Centers for Disease Control and Prevention (CDC) has done some assessments looking at some preventive service recommendations and how often they are implemented in practice, and sometimes other researchers do that, but it's not part of our scope. Our scope is to make and disseminate the recommendations, not to assess their uptake in practice.

Dueling Recommendations

Medscape: Many specialty societies and professional organizations develop their own guidelines, often related to screening. How do you avoid the problems of duplication and overlap?

Dr Grossman: We have a referral process for when we know that another group follows a similarly rigorous, evidence-based process. It's uncommon, but one example is the Secretary's Advisory Committee on Heritable Disorders in Newborns and Children, who make newborn screening recommendations. We don't see a reason for us to duplicate that effort. Similarly, we don't make immunization recommendations, and refer our readers to the CDC and Advisory Committee on Immunization Practice recommendations.

We always look at what other specialty societies, as well as other national and even international evidence-based bodies, are saying on a specific topic, and we point that out in our recommendation statements. It's important for us to be mindful of what the external community is saying and doing and to understand their points of view.

One way in which we are different from recommendations put out by other groups is in the use of the I statement, which is a statement about the insufficiency of the evidence. An insufficient evidence statement means that the Task Force can't make a recommendation either for or against a service, owing to inadequate evidence of harms, benefits, or both. Most other guideline groups don't use the I statement, perhaps because they use other systems to grade the evidence. We recognize that I statements are not that helpful to clinicians, but it's important for clinicians understand where the boundary of evidence is—where they are practicing on sound evidence, and where they are practicing on the basis of clinical judgment.

It's part of our scope to be clear to the primary care community and others about significant gaps in the evidence, so that we can make recommendations for research. In our annual report to Congress—which is posted on our website—we highlight the burning, pressing issues that need to be addressed because they are interfering with our ability to make a clear recommendation on a topic. The National Institutes of Health disseminates that report among its institutes to encourage them to sponsor or promote the research necessary to close those gaps. And this works. Several I statements have changed over to an A or B statement after new trials were published.

An example is screening for childhood obesity. The American Academy of Pediatrics' Bright Futures program recommended obesity screening for children before we did. We said it was an I statement. Then several new trials were published demonstrating the effectiveness of interventions, and we changed that recommendation to a B.

Medscape: How often do your recommendations conflict with those of another group?

Dr Grossman: That issue has come up any number of times. One factor that comes into play in making different recommendations is the judgment involved in determining the net benefit and the balance of benefits and harms. We can't make a recommendation purely on the basis of benefit—we must also understand the potential harms of a screening test, for example, as well as the potential harms of any interventions that follow the screening test. But frequently, we are in agreement with specialty societies. It goes both ways.

Areas of Controversy

Medscape: What have been the USPSTF's most controversial recommendations?

Dr Grossman: Breast cancer screening is an acknowledged controversial area. We ran into this controversy both in 2009 and to a lesser degree during the most recent round in 2016.

One controversy is related to the appropriate time to start and the appropriate frequency for screening mammography. These are very difficult issues for which we end up having to rely not only on evidence, but also to some degree modeling work that we commission.

I'll give you an example. In 2009, overdiagnosis and overtreatment were relatively new concepts. We were one of the first groups to float this as an important potential harm. It wasn't widely acknowledged in the community that the potential for overdiagnosis of breast cancer was a real issue. We outlined that as a potential harm, and one of possibly greater importance to younger women in their 40s and 50s, because the benefits of mammography in that age group are smaller than in older women. Yet, the harms of overdiagnosis are about the same.

That's an example of when there was substantial controversy. The research and clinical communities have come around to recognize the possibility of overdiagnosis, and all groups, including the radiologists, now acknowledge it as a harm. Screening mammography is an area where we want to be absolutely certain about the net benefit—that what we are recommending is not only effective in preventing the condition we are trying to prevent, but also that we are not inducing harm in the process.

We have looked very carefully at the frequency of mammography screening and at what happens when you do annual vs biennial mammography. We used the evidence, along with modeling data, to determine the balance between the extra benefit provided by annual mammography and the harms caused by unnecessary biopsies, procedures, and treatments. The American Cancer Society changed their recommendations from annual to every 2 years starting at age 55 years, recognizing that there is potential harm and probably not much extra benefit to annual mammograms.

Medscape: What other types of criticism has the USPSTF received over the years?

Dr Grossman: We know we are doing our work when we have supporters and detractors for any given recommendation. We typically get criticism from several areas. We are criticized by some specialists who don't necessarily have the framework or view of primary care clinicians. Our interest is in primary care, but the patients seen by many specialists are not the same as those seen in primary care. Specialists see people who are referred to them, and very rarely are these patients asymptomatic. That's not the population of patients for whom our recommendations are intended. Yet, specialists are a critical resource to our process, and we value their input very much.

There is often a misunderstanding about what screening is. Screening is not testing. Testing is based on a hypothesis, whereas screening is uniformly applying a tool to a population, on the basis of demographics or risk factors. This misunderstanding is the source of much of the criticism we receive.

We have been doing this work since 1984, and primary care groups, such as the American Academy of Family Physicians, the American College of Physicians, and other key stakeholders, have relied on our recommendations for decades. It wasn't until later, and more recently when the issues of linkage to insurance coverage arose, that we began to elicit attention from other groups, and evoke more criticism and concerns about that linkage. Although these concerns were raised, they have not changed the basics of our process or how we make our recommendations. Our methods, however, continuously evolve and improve.

We are proud of the fact that we are as transparent as we are, both in terms of putting our draft recommendations and evidence reviews out for public comment, and the transparency of our methods. We are always striving to improve, but we are proud of what we have accomplished.

The Far-Reaching Effects of Recommendations

Medscape: Even without a direct link, how have your recommendations influenced coverage decisions? Do insurers decide what to cover and not cover according to your recommendations?

Dr Grossman: I would refer you to talk to someone from an insurance industry trade group or the Department of Health and Human Services to answer those questions. We published an editorial that highlights our position on these topics in JAMA.^[1]

We recognize that the Affordable Care Act (ACA) mandates coverage for A and B recommendations. Our job is to figure out what the evidence supports, not whether it should be covered or excluded by insurance. Those decisions are made by legislators, Congress, and the insurance companies. We recognize that such coverage decisions are made externally, but in no way do considerations about coverage or cost enter into our decision-making or discussions about how to issue a grade. A grade is based only on the magnitude of net benefit to the patient and the certainty of that benefit.

Medscape: Will repeal or changes to the ACA affect what you do, particularly if payment for preventive services is largely eliminated from insurance coverage?

Dr Grossman: As long as we have a Task Force and we continue to receive support from AHRQ, we will continue doing what we have been doing for 40 years. We believe that access to preventive services is important, and we are nonpartisan and don't take political positions on any issues, including the ACA. Having said that, we fully recognize that access to preventive services is an important determinant of health.

Medscape: When a bill was introduced into the last Congress to "change the way in which decisions are made" by the USPSTF, did this result in any changes to your processes? For example, their requirement to add specialists to the reviewing/voting panel—is this something that the Task Force plans to do? And are you still being pressured by legislators or outside medical groups to change your processes? If so, are you planning any other changes?

Dr Grossman: The most important thing to note is that many of the issues raised in the proposed legislation are already part of the Task Force's existing policies and procedures. For example, the bill requires that the Task Force seek public comments on its materials, which we already do for every draft research plan, evidence review, and recommendation statement as part of our commitment to making our recommendations clear and our processes transparent.

As for the composition of the Task Force, our members are nationally recognized experts in prevention, evidence-based medicine, and primary care—the specialties to which the broad range of our recommendations are most directly applicable. Given the more than 100 different topics under our purview, we have found that the most efficient and effective approach is consultation with other specialists for the specific topic in which they have expertise, and we already have a robust approach for seeking the that feedback throughout the recommendation development process.

Medscape How much lobbying and pressure do Task Force members receive from industry representatives, and how do you handle this or prevent it seeping into your decisions?

Dr Grossman: The Task Force receives input and public comments from a diverse range of stakeholders, including industry representatives, throughout the recommendation development process. Although the Task Force considers all comments carefully as our recommendations are finalized, overall, we are guided by our rigorous evidence-based process for reviewing the available scientific research, which includes disclosure of all potential financial and nonfinancial conflicts by our members to ensure that our recommendations remain unbiased and focused solely on the evidence.

Medscape Family Medicine © 2017  WebMD, LLC

Any views expressed above are the author's own and do not necessarily reflect the views of WebMD or Medscape.

Cite this: Everything We Ought to Know About the USPSTF - Medscape - May 03, 2017.