A new mathematical model could help clinicians prioritize which US Preventive Services Task Force recommendations to use with their patients and may improve life expectancy, researchers write in an article published in the August 6 issue of the Annals of Internal Medicine.
"The [USPSTF] makes recommendations for 60 distinct clinical services," write Glen B. Taksler, PhD, from New York University School of Medicine, New York City, and colleagues. "Yet, although receipt of preventive health care services has improved in the past decade, only approximately one half of recommended services are provided."
Use remains "alarmingly low" for some services, such as screening for colorectal cancer, which is not done in 48% of individuals. In addition, use of some services is uneven, which may add to disparities in health outcomes, the authors write. For example, 37% of blacks older than 65 years receive pneumococcal vaccinations compared with 62% of whites of similar age.
" An important reason why the U.S. health system does not implement prevention guidelines consistently may be insufficient personalization and prioritization at the point of care," the authors write. "The time available to fully evaluate and implement all relevant recommendations is widely considered inadequate and would require more than 7 hours each day for a typical practice panel of 2500 patients."
This makes prioritization essential, they add.
To facilitate personalized decision making, the authors developed a mathematical model that estimates the health benefit for each of the 25 USPSTF grade A and B guidelines applicable to nonpregnant adults.
In their model, which is a proof-of-concept effort, they examined each USPSTF recommendation in the context of a person's risk–benefit profile and how life expectancy could be influenced by the intervention.
The model showed that the rank order of benefits varied considerably on the basis of demographic characteristics, medical conditions, and lifestyle choices. For example, a 62-year-old obese man who smoked and had high cholesterol, hypertension, and a family history of colorectal cancer would benefit most from USPSTF recommendations on tobacco cessation, weight loss, and blood pressure control. In contrast, the man would not derive as great a benefit from screening for colorectal cancer, HIV, and abdominal aortic aneurysms.
The authors made certain assumptions to keep the model generalizable and feasible in a patient setting. They assumed that age, race, and sex were sufficient to represent average baseline life expectancy but did not include other factors such as ethnicity, geography, and socioeconomic status.
They also assumed that personalized risks affected mortality at constant multiplicative rates or relative risks. For example, a patient with hypercholesterolemia had 1.44 times the relative risk for abdominal aortic aneurysm, regardless of other risk factors. However, relative risks were allowed to vary across age, race, and sex groups when estimates were available from existing literature.
The third assumption was that baseline benefits of preventive care would depend on the relative risk reduction from a preventive care service, the proportion of the population adhering to that service, and the delay for benefits to occur.
Adam and Bill: 2 Hypothetical Patients
The authors depicted scenarios featuring 2 hypothetical patients with chronic conditions and health risks.
"Adam" was a 62-year-old white man with a body mass index (BMI) of 30.4 kg/m2 who smoked, had hypercholesterolemia, hypertension (150/90 mm Hg), and a family history of colorectal cancer.
"Bill" had the same conditions plus uncontrolled type 2 diabetes, with a hemoglobin A1C level of 9%.
The authors calculated that Adam's life expectancy was 13.1 years, and that Bill's was 9.6 years.
Using the model, the prioritized recommendations for Adam were as follows:
quit smoking, which would add 2.8 years to his life;
lose 7.3 kg or more, to add 1.6 years;
lower blood pressure to 120/80 mm Hg to add 0.8 year;
eat more fruits and vegetables (0.3 year);
use daily aspirin (0.3 year);
reduce cholesterol levels (0.3 year);
have decennial screening colonoscopy (0.2 year); and
have ultrasonography to screen for abdominal aortic aneurysm (0.1 year).
For Bill, the prioritized recommendations were as follows:
control diabetes to lower his hemoglobin A1C level less than 7% to add 1.8 years;
quit smoking (1.5 years); and
lower blood pressure (1.4 years).
"In general, both our model and a recent Lancet Global Burden of Disease series consistently rank tobacco cessation, diabetes control, weight loss, and hypertension control as high priorities," the authors write. "However, we shifted the focus from population burden of preventable morbidity and mortality to individual priorities."
They are now using their model to support decision making in a primary care clinic as part of a pilot study.
Future versions of this decision aid should include patient-level adherence rates, they write. "For example, if a person has a 50% probability of adhering to hypertension medication but only a 25% probability of successfully completing a tobacco cessation program, then the relative importance of tobacco cessation should decline by one half," they write.
They also say that including quality of life and patient preferences in addition to life expectancy in the model would improve "patient-centeredness" or individualized care.
Adding variables such as income and education should also be incorporated, the authors note.
"I think this is very interesting work and promising," Douglas K. Owens, MD, from Veterans Affairs Palo Alto Health Care System and the Center for Primary Care and Outcomes Research and Center for Health Policy, Stanford University, California, told Medscape Medical News.
Although he is a member of the USPSTF, he emphasized that this is his personal opinion. "I'm not speaking on behalf of the USPSTF."
Dr. Owens and colleague Jeremy D. Goldhaber-Fiebert, PhD, also from Stanford University, wrote an accompanying editorial in which they note that the model helps physicians understand the relative importance of different interventions for a specific patient.
"How to prioritize what to do in the limited time of a clinical encounter is a pervasive problem in primary care," they note. "While broader strategies are developed to provide all recommended care, the tool developed by Taksler and colleagues has promise to help make sure the interventions with the greatest effect are delivered. Whether its use will improve patient outcomes is a question worthy of further study," the editorialists conclude.
"The results are provocative and important because of the enormous differences in magnitude between various preventive services," said David F. Ransohoff, Professor of Medicine at the University of Carolina at Chapel Hill, North Carolina.
"The average life gained is years by control of diabetes, cessation of smoking, and weight loss. That is a huge amount for a preventive service," Dr. Ransohoff told Medscape Medical News. "In contrast, gains from other preventive services, like breast cancer screening, may be much smaller in magnitude but receive disproportionately intense public attention," he said.
"This analysis highlights for clinicians, patients and policy makers, the challenge of how to prioritize and implement preventive strategies when there are real-world limitations to time, resources, or patient interest."
The study was funded by New York University School of Medicine. Dr. Taksler has disclosed no relevant financial relationships. Dr. Owens reports that he is a member of the USPSTF, that he works on guidelines for the American College of Physicians, and that he was paid as a consultant by the college for work on their initiative on high-value healthcare. Dr. Goldhaber-Fiebert and Dr. Ransohoff have disclosed no relevant financial relationships.
Ann Intern Med. 2013;159:161-168. Abstract
Medscape Medical News © 2013 WebMD, LLC
Send comments and news tips to email@example.com.
Cite this: Model Prioritizes USPSTF Recommendations for Individuals - Medscape - Aug 06, 2013.