Probability of test being positive given that disease is present
Ability of a test to correctly identify people without a disease
If a test result is positive, how often does the patient have the disease?
Positive Predictive Value
True Negative/(true negatives + false negs)
Negative predictive Value
In a clinical scenario, how can a test result from a test with high specificity can be most useful? A negative result can be used to rule out disease A positive result can be used to rule in disease
A positive results can be used to rule in disease. Tests with high specificity do a good job identifying people without disease (lots of people without disease have a negative test; there are many true negatives). The flip side is that someone who does have the disease might also be incorrectly classified as having the disease by having a positive test result (false negatives). However, most positive results produced from a test with high specificity will likely be true positives. Thus, a positive result from a test with high specificity can be used to rule in disease (SPIN). Tests with high sensitivity do a good job identifying people with disease (true positives), but someone who doesn’t have the disease might also incorrectly have a positive test result (false positives). Most negative results produced from a test with high sensitivity will likely be true negative results. Thus, a negative result from a highly sensitive test can be used to rule out disease.
When a woman is screened for breast cancer and has a positive test result but ends up not having breast cancer, what is this called?
False positive result A false positive occurs when a person without disease has a positive test results. The frequency of false positive and false negative test results is an important consideration in the evaluation of screening tests.
What is “overdiagnosis” in the context of breast cancer screening?
When a woman has cancer diagnosed but that cancer would never cause any health problems in the future - Overdiagnosis pertains to slow-growing cancers. Cancers that are overdiagnosed can be cured but do not need to be cured. The proportion of cancers that are overdiagnosed is estimated at a population level. As screening increases and disease diagnosis increases, mortality due to the disease of interest is considered. If mortality rates remain relatively stable despite these increases, this is suggestive of overdiagnosis (screening is detecting more cases, but not preventing death).
Consider the test performance of mammography for the detection of breast cancer. According to the Kerlikowske article, how many women aged 40-49 years need to be screened in order to prevent one death, based on data from summary mortality reductions?
How does using risk-based screening criteria impact the predictive value of a test?
Increase s PPV - Using risk-based criteria to define a group of people eligible for screening creates a population at a higher risk for disease. Thus, the disease will be more common in this population (the prevalence will be higher). When the disease is more common, a person who has a positive test result is more likely to actually have the disease. The probability of disease given a positive test result (PPV) is higher.
According to the Kerlikowske article, what is the 10-year risk of a women having a false positive mammography result for breast cancer is she starts annual screening at age 40?
What is one of the primary harms associated with false positive mammography results?
Overdiagnosis leading to unnecessary treatment
Which of the following might be considered secondary prevention of cancer?
Getting a screening mammography to promote early detection of breast cancer - The purpose of screening mammograms is to detect disease early and initiate treatment before disease progression. Secondary prevention involves detecting existing disease early in the disease process to prevent the development of adverse outcomes. The other options represent primary prevention activities aimed at preventing cancer before it occurs.
A 54-year-old man presents to your clinic for a new patient evaluation. He states that he is concerned about prostate cancer. He reports that his friend was recently screened, diagnosed, and treated for early-stage prostate cancer. He has also read that screening is controversial. He asks if he should be screened. What do you tell him?
Prostate cancer screening using the PSA test is not recommended due to potential downstream harms of screening. The United States Preventive Services Task Force (USPSTF) does not recommend prostate cancer screening using the PSA test. The potential harms of screening weight the possible slight benefits of screening.
A healthy 50-year-old man presents to your clinic for a routine well-person exam. When determining what types of preventive screening the man should consider, what is the primary agency looked to by many U.S. physicians devoted to developing evidence-based recommendations for cancer screening that you should consult?
The USPSTF is made up of a panel of health care experts that evaluate the latest scientific evidence on clinical preventive services. Evidence summaries are peer-reviewed and published in scientific journals. Subsequent recommendations are drafted and presented for public comment. USPSTF produces clinical guidelines for cancer screening and is considered the gold-standard guideline organization.
A 28-year-old woman presents to your office for a routine Pap smear. A friend of hers just found out she has breast cancer, and she is now concerned about her own risk. She doesn't smoke. Her BMI is 23. She exercises for 30 minutes per day 5 days/week. What is the most appropriate action she could take now to reduce her future risk of getting breast cancer?
- Obtain annual mammograms
- Eat 5 or more servings of fruits and vegetables daily
- Take vitamin E supplements
- Take beta carotene supplements
- Wear protective clothing and sunscreen when outside
Of the 4 recommended healthy lifestyle habits, this 28-year-old woman may only be lacking in recommended fruit and vegetable intake (she already exercises, maintains a healthy weight, and does not smoke). Breast cancer screening by mammogram is not recommended for an average risk 28-year-old woman.
An advertisement printed in a women's magazine urges women in their 40s to get screened for breast cancer and touts that the 5-year survival rate for breast cancer when caught early is 98%. This likely is an example of?
The lead time is the length of time between the detection of a disease (usually based on new, experimental criteria) and its usual clinical presentation and diagnosis (based on traditional criteria).
A 35-year-old woman comes to your office for a contraception refill and asks whether or not she should be screened for breast cancer. She has no known family history of ovarian or breast cancer. When considering routine breast cancer screening through mammography for her, what is the most important risk factor to consider?
- Breast density
- Smoking status
- Alcohol consumption
Age - The risk of breast cancer increases as age increases. This is the most important factor to initially consider when deciding if an average-risk patient should be screened.
The mayor of a small town in Kansas is interested in promoting a healthy lifestyle among her constituents. Because she only has a small staff, she wants to focus on one area that will apply to the general population. In order to reduce the cancer risk of her constituents, what one activity should she focus on providing?
Development of a clean air act prohibiting smoking in public places -Tobacco exposure is the number one cause of preventable cancer deaths. By restricting smoking in public places, the mayor can impact tobacco exposure in two ways: reducing exposure to secondhand smoke and promoting smoking cessation by limiting the number of places that smokers can smoke outside of their home.
A 64-year-old man presents to your clinic for a new patient evaluation. He has a BMI of 32 and occasionally smokes a cigarette (<1 pack/month). His blood pressure is normal. He has not been to the doctor in a long time. What preventive cancer screening do you recommend for him?
Colorectal cancer screening using a colonoscopy - The most appropriate screening for this man is colorectal cancer screening using a colonoscopy, a sigmoidoscopy, or a fecal occult blood test. Colorectal cancer screening is recommended for adults aged 50-75 years. While this man does smoke, he is not a heavy smoker; thus, lung cancer screening is not recommended.
You are a member of an objective panel reviewing evidence around a new cancer screening test. What test performance characteristic is the most important to consider prior to implementing a screening program using this test?
Sensitivity - While it is desirable for a screening test to have both high specificity and high sensitivity, sensitivity is the most important test characteristic to consider. Highly sensitive tests produce few false negatives; since these tests do a good job correctly identifying people with disease, most people with disease will have a positive test. Because there are few false negatives, most negative results are true negatives. Thus, a negative result from a highly sensitive test can be used to rule out disease. When screening someone for cancer, you can be pretty sure that a negative test means that cancer is not present.
An evaluation was conducted on the utility of annual screening for lung cancer with low-dose computed tomography (CT) among heavy smokers. The subjects underwent baseline evaluations in 1996, with repeat annual examinations the next two years. Confirmation of a lung cancer diagnosis was based on cellular pathology. A summary of the results of all three screenings combined is provided in the table below. What is the specificity of this test?
13,198 / 13,730 - Specificity is the ability of a test to identify correctly those who do not have disease. Specificity is calculated as the probability of testing negative given the absence of the disease: true negatives/(true negatives+false positives). In this question, there were 13,198 true negative results and 532 false positive results. Thus, specificity is 13,198/(13,198+532), or 13,198/13,730. The specificity of the test is 96.1%.
As part of a well-woman exam, you order a screening mammography for your average-risk 62-year-old patient. Her mammography results come back as positive for breast cancer. Based on the test performance characteristics of screening mammography, what can you conclude about this finding?
Mammography is a highly specific test, so you conclude that she likely has breast cancer. As a screening test, mammography is about 92% specific among 50-54 year-old women. This means that the test correctly identifies about 92% of all women who do not have breast cancer. Because the test does a relatively good job at identifying women who don’t have disease, we can be pretty confident that a positive test result is a true positive.
The benefits of cancer screening must be balanced by a variety of risks or harms associated with screening. Breast cancer screening by mammography is recommended for women of average risk aged 50-74 years. What is the primary reason that breast cancer screening is not widely recommended for younger women?
False positive screening results can lead to further diagnostic work-up - The biggest risk of screening for cancer in average-risk populations is the risk of false positive screening results. A false positive result often leads to extensive and costly downstream effects including diagnostic work-up, treatment, and follow-up, in addition to psychological stress.
A 12-year-old adolescent female is visiting your office with her mother for a sports physical before school starts. She is in good overall health, maintaining a healthy body weight and exercising regularly. What is the most important step for you to recommend to the 12-year-old for the prevention of cancer?
Get the HPV vaccination before she leaves. - HPV has been linked to numerous genital and oropharyngeal cancers, and the HPV vaccine is recommended for all adolescent females and males. This is the most important prevention recommendation that you can make at this visit, especially given her overall health.
In 2015, a study was published that evaluated the 10-year relationship between breast cancer screening, incidence, and mortality. The authors reported that breast cancer screening and incidence were associated, but that increased screening was not correlated with decreased mortality. What type of bias are the findings of this study addressing?
Overdiagnosis bias - Overdiagnosis occurs when a cancer was diagnosed (and treated), but it didn’t need to be diagnosed and treated because it wasn’t going to lead to clinical signs or symptoms. Cancers that are overdiagnosed can be cured but do not need to be cured. By showing that screening increased the number of incident cancers detected, but didn’t decrease mortality, the authors’ findings suggest substantial overdiagnosis of cancers detected by screening.
Researchers are designing a study to evaluate the effectiveness of a new cancer screening program. What study design is considered the optimal design to address this question?
Randomized controlled trial - A randomized controlled trial is considered the gold standard study design for evaluating screening programs. This study design allows for two exposure groups, one randomly assigned to no screening, and the other randomly assigned to screening. Both groups are the same on all factors except the exposure of interest (screening/no screening). Both groups are then followed forward in time to look at mortality over a set follow-up period.
A hypothetical new test designed to detect prostate cancer has been approved by the Food and Drug Administration (FDA). Experts have gathered to draft screening recommendation guidelines and have suggested that screening efforts target men aged 50 years and older due to the increased risk of prostate cancer as age increases. What test performance characteristic are experts aiming to improve by creating risk-based screening recommendations?
Positive predictive value - Increasing the prevalence of the disease that will be screened for results in an increase in the positive predictive value. Changes in prevalence can impact both the positive and negative predictive values, but do not impact sensitivity and specificity. The likelihood ratio is calculated using sensitivity and specificity.
A 55-year-old man presents to your clinic for an annual well-person exam. After reviewing his records, you determine that he is due for colorectal cancer screening. He is at average risk for colorectal cancer. He agrees to screening, and you run a test that is highly sensitive but has low specificity. A week later, his test comes back positive. What is the next step for you to take?
- Diagnose the man with colorectal cancer and refer him to an oncologist for treatment
- Rerun the same test again for result confirmation
- Suggest he return in 1 year for follow-up screening
- Run a different test for result confirmation
- Explain to him that the result is likely a false positive result
Run a different test for result confirmation - A test with high sensitivity is useful clinically if the result comes back negative. Negative results from a highly sensitive test can be used to rule out disease (since the test does a good job correctly identifying people with disease, most negatives will be true negatives). Because this test has low specificity, the positive result may not be conclusive. Tests with low specificity do not do a good job correctly identifying people without disease, meaning that some people without disease will have a positive test (false positives). The best course of action is to send a confirmatory test using a test with better specificity.
A 40-year-old woman is considering getting screened for breast cancer. She has no family history but is very concerned about her cancer risk and feels strongly that she should be screened. What is the most important possible harm of breast cancer that she should be aware of through her decision-making process?
A false positive screening test result - The biggest possible risk of breast cancer screening conducted among an average-risk 40-year-old woman is a false positive test result. About 40% of women who begin screening at age 40 and are screened every other year until age 49 will have a false positive test result.
The Centers for Disease Control and Prevention (CDC) is interested in reducing prostate cancer mortality. What primary prevention activity should they recommend for the most impact?
Implementation of work policies that allow two 10-minute exercise breaks daily - Primary prevention activities are aimed at preventing disease before it occurs. Increasing physical activity has been shown to reduce overall cancer risk. Prostate cancer tests, whether in an office or a self-administered prostate exam, are secondary prevention strategies aimed at detecting cancer early. Increasing prostatectomy training and facilitating treatment decisions would be directed toward tertiary prevention, diagnosing disease and treating to minimize adverse outcome.
In a study evaluating the effectiveness of systematic skin cancer screening for the prevention of skin cancer mortality, a total of 306,288 patients needed to be screened to detect 585 cases of melanoma. Assuming that 12.5% of these melanomas would have been fatal (n=73) and that screening is 100% effective, how many patients needed to be screened to prevent one melanoma-related death?
306,288/73 - To determine how many patients would need to be screened to prevent one death, divide the total number of people in the population (306,288) by the total number of deaths (73). This calculation results in 4196. In order to prevent one melanoma-related death, 4196 patients needed to be screened. This assumes that screening in 100% effective. In reality, screening programs are generally considered to be effective at 20% (for instance, a 20% reduction in mortality in the screened group compared to the unscreened group). If a 20% effectiveness is applied to the number needed to screen (73 deaths x 0.20, since only 20% would have been prevented), the number needed to screen would be 20,420 (306,288/15).
A population of 1200 men will be screened for colorectal cancer using a multitarget DNA test (sensitivity=92.3%, specificity=86.6%). The prevalence of colorectal cancer in this population is estimated at 1.0%. How many false positive test results could be expected to occur in this population?
159 - Of the 1200 men in the sample, 12 will have colorectal cancer (12/1200=1%), and 1188 will not. However, not all of the men without cancer will have a negative test result, since specificity is not 100% (thus, the ability of the test to correctly identify those without disease is not perfect). This means that some of those 1188 will be false positives. The formula for specificity is true negatives/(true negatives+false positives). Since we know that specificity is 86.6%, we know that only 86.6% of all of the men without disease (1188) will have a negative test (true negatives=1188/0.866=1029). The remainder of men without disease, 1188-1029=159, will be false positives.