Assignment 6 week 6
Sneha Dhareshwar
DPT
A.T. Still University, AZ
October 2018
1. Sensitivity is the ability of the test to obtain a positive test when the target condition is really present, or the true positive rate. The sensitivity is calculated as a/ (a + c) on the 2×2 table. (Portney & Watkins, 2015).
2. Specificity is the test's ability to obtain a negative test when the condition is really absent, or the true negative rate. It is calculated by d/ (b + d). (Portney & Watkins, 2015).
3. A positive predictive value (PV +) estimates the likelihood that a person who tests positive actually has the disease. The positive predictive value gives us a number that gives information on the true positives and implies that a test with high positive predictive value will provide a strong estimate of the actual number of people who have the disease.
Positive predictive value (PV+) = a/(a + b). (Portney & Watkins, 2015).
4. Negative predictive value is the number that states that a test performed was negative in results and the person does not have the disease. Therefore, the Negative predictive value gives information on true negatives. It is calculated by d/(c+d). (Portney & Watkins, 201
5. The sensitivity of the mammogram which is the ability to accurately diagnose patients with diagnosis of breast cancer is calculated as a/a+ c. Therefore, the sensitivity is 0.64.
6. The sensitivity of the mammogram is the ability of the test to correctly identify the diagnosis of breast cancer. In this case, about 64% of women which is about 90 women were correctly identified with the diagnosis of breast cancer.
7. The specificity of the test is ability to correctly identify the patients who do not have the disease. Specificity is calculated as d/ (b + d) =0.936
8. Specificity is the property of the test that accurately identifies patients who do not have the disease. In this case, about 93.6% of the women were accurately identified as not having breast cancer. In a total of 940 patients, 93% of women which is 740 were correctly identified as not having the disease.
9. Likelihood ratio is the number that indicates how much more likely it is to have the disease when the test is conducted. The positive likelihood ration (LR+) was calculated at 9.14 and shows almost nine times it was likely to obtain a positive test in patients with the disorder than those without the disorder. A high LR+, to indicate that the disorder is likely to be present with a positive test. (Portney & Watkins, 2015).
10. A positive likelihood ratio indicates that the disease is present. A high positive likelihood ratio indicates that there is a greater likelihood of the disease to be present. Positive values that are greater than 5 indicate that there is high likelihood of the disease being present. (Portney & Watkins,2015). This indicates, that mammogram showed high likelihood of obtaining positive results of breast cancer.
11. A negative likelihood ratio suggests that the confirmation of the disease being absent and improves the diagnostic conclusion. In this case, the LR- was calculated at 0.387.
12. The LR– indicates how many times more likely a negative test will be seen in those with the disorder than in those without the disorder. Portney & Watkins, (2015) state that a LR- less than 0.2 shows that the test is significant, suggest that the mammogram is relatively important in the diagnosis of breast cancer.
13. Using the given data, the Positive predictive value is calculated as a/(a + b) which 833/983 and totals as 0.84 or 84%
14. The Negative predictive value is the d/(c + d) and equals 65/63650 which 0.001
15. With the given data and calculation, the positive predictive value (PPV) is 84% which suggests that 84% of the people were acutely diagnosed with the disease when they had the disease with positive result.
16. The negative predictive value (NPV) is calculated as 0.0001 which suggests the proportion of subjects without the disease with a negative test result in total. This indicates that 0.1% of the patients may show negative results is actually disease free.
Part II:
Article:
The study by Van der Post, A., Noorduyn, J. A., Scholtes, V. B., & Mutsaerts, E. R. (2017) was conducted to evaluate the sensitivity and the specificity of the duck walk test in diagnosing medial meniscal tears and to know if other factors like cause and location of meniscal tears, associated ACL insufficiency led to differences in sensitivity and specificity of the test.
With the inclusion and exclusion criteria clearly defined, a convenience sample of 136 patients was chosen to participate in the study and the duck walk test was implemented by an expert orthopedic surgeon. The results demonstrated that overall 51% of the patients were diagnosed with meniscal tears. The sensitivity of the test was calculated at 71% (95% CI, 59%- 81%). Specificity was totaled at 39% (CI 95% 27-52). Similarly, the Positive Predictive Value (PPV) was calculated at 54% and the Negative Predictive Value at 57%.
The results imply that out of 136 patients, 69 patients were confirmed with diagnosis of one or more meniscal tears. Therefore, the prevalence of meniscal tears in this population was 51%. Porteny & Watkins, (2015) define sensitivity as the ability of the test to accurately identify patients who have the disease. The duck walk test identified 71% of the patients having meniscal tears accurately. Therefore, the sensitivity of the test was 71%. In addition, the sensitivity was much lower than 71% due to verification bias was also reported as one of the limitations of the study. Specificity is the proportion of patients that test negative and are disease-free. The results in the study by Van der Post et al. (2017) totaled specificity at 39% which suggests that the duck walk test could only identify 39% of the subjects accurately who did not have meniscal tears and showed negative results in investigations. The specificity shown was lower than that seen in other provocative tests by Thelasy 200 test and McMurrays test. Likelihood ratios are alternative statistics for summarizing diagnostic accuracy. The positive likelihood ratio(+LR) was 1.16 which suggests that there was greater probability of a positive diagnosis of meniscal tear since likelihood ratio greater than 1 indicates that the test result is associated with the presence of the disease. LR+ > 10 and their positive result has a significant contribution to the diagnosis. (Deeks & Altman, 2004). Positive predictive value (PPV) is the probability that an individual with a positive test truly has the disease. (Gogtay & Thatte (2017). The PPV calculated at 54% suggests that only 54% of the 136 patients who were tested showed that could possibly test positive for have the meniscal tears. Although a high PPV is always desirable, 54% represents a proportion of patients with positive test result in total of subjects with positive result showing only moderate clinical significance.
In conclusion, complete information about the diagnostic test is important in clinical decision-making process. The study by Van der Post et al. (2017) provided important data in the analysis of the duck walk test. Diagnostic tests should be judiciously evaluated to be implemented in patients for accurate results. Because sensitivity and specificity are only test performance features and do not address the problem of prevalence in diverse populations; positive and negative predictive values are important in the interpretation of test results. (Gogaty & Thatte, 2017). With the described sensitivity of 71% and specificity of 30%, the PPV at 54% the duck walk test in not the best clinical tool in the assessment of meniscal tears.
References:
Deeks, J. J., & Altman, D. G. (2004). Diagnostic tests 4: likelihood ratios. BMJ (Clinical Research Ed.), 329(7458), 168–169. Retrieved from http://p.atsu.edu/login?url=http://search.ebscohost.com.p.atsu.edu/login.aspx?direct=true&db=mdc&AN=15258077&site=eds-live.
Gogtay, N. J., & Thatte, U. M. (2017). Statistical Evaluation of Diagnostic Tests (Part 1): Sensitivity, Specificity, Positive and Negative Predictive Values. The Journal Of The Association Of Physicians Of India, 65(6), 80–84. Retrieved from http://p.atsu.edu/login?url=http://search.ebscohost.com.p.atsu.edu/login.aspx?direct=true&db=mdc&AN=28782317&site=eds-live.
Portney, L., & Watkins, M. (2015). Principles of measurement. In Foundations of clinical
research (3rd edition) (pp.63-75), Philadelphia, PA: F.A. Davis Company