Sensitivity and specificity are two components that measure the inherent validity of a diagnostic test for dichotomous outcomes against a gold standard. Receiver operating characteristic (ROC) curve is the plot that depicts the trade-off between the sensitivity and (1-specificity) across a series of cut-off points when the diagnostic test is continuous or on ordinal scale (minimum 5 categories). This is an effective method for assessing the performance of a diagnostic test. The aim of this article is to provide basic conceptual framework and interpretation of ROC analysis to help medical researchers to use it effectively. ROC curve and its important components like area under the curve, sensitivity at specified specificity and vice versa, and partial area under the curve are discussed. Various other issues such as choice between parametric and non-parametric methods, biases that affect the performance of a diagnostic test, sample size for estimating the sensitivity, specificity, and area under ROC curve, and details of commonly used softwares in ROC analysis are also presented.
Sensitivity and specificity measure inherent validity of a diagnostic test against a gold standard. Researchers develop new diagnostic methods to reduce the cost, risk, invasiveness, and time. Adequate sample size is a must to precisely estimate the validity of a diagnostic test. In practice, researchers generally decide about the sample size arbitrarily either at their convenience, or from the previous literature. We have devised a simple nomogram that yields statistically valid sample size for anticipated sensitivity or anticipated specificity. MS Excel version 2007 was used to derive the values required to plot the nomogram using varying absolute precision, known prevalence of disease, and 95% confidence level using the formula already available in the literature. The nomogram plot was obtained by suitably arranging the lines and distances to conform to this formula. This nomogram could be easily used to determine the sample size for estimating the sensitivity or specificity of a diagnostic test with required precision and 95% confidence level. Sample size at 90% and 99% confidence level, respectively, can also be obtained by just multiplying 0.70 and 1.75 with the number obtained for the 95% confidence level. A nomogram instantly provides the required number of subjects by just moving the ruler and can be repeatedly used without redoing the calculations. This can also be applied for reverse calculations. This nomogram is not applicable for testing of the hypothesis set-up and is applicable only when both diagnostic test and gold standard results have a dichotomous category.
Lambda-Mu-Sigma and Box-Cox Power Exponential are popular methods for constructing centile curves but are difficult to understand for medical professionals. As a result, the methods are used by experts only. Non-experts use software as a blackbox that can lead to wrong curves. This article explains these methods in a simple non-mathematical language so that medical professionals can use them correctly and confidently.
ObjectiveTo describe the clinical profile and factors leading to increased mortality in coronavirus disease (COVID-19) patients admitted to a group of hospitals in India.DesignA records-based study of the first 1000 patients with a positive result on real-time reverse transcriptase-polymerase-chain-reaction assay for SARS-CoV-2 admitted to our facilities. Various factors such as demographics, presenting symptoms, co-morbidities, ICU admission, oxygen requirement and ventilator therapy were studied.ResultsOf the 1000 patients, 24 patients were excluded due to lack of sufficient data. Of the remaining 976 in the early phase of the epidemic, males were admitted twice as much as females (67.1% and 32.9%, respectively). Mortality in this initial phase was 10.6% and slightly higher for males and steeply higher for older patients. More than 8% reported no symptoms and the most common presenting symptoms were fever (78.3%), productive cough (37.2%), and dyspnea (30.64%). More than one-half (53.6%) had no co-morbidity. The major co-morbidities were hypertension (23.7%), diabetes without (15.4%), and with complications (9.6%). The co-morbidities were associated with higher ICU admissions, greater use of ventilators as well as higher mortality. A total of 29.9% were admitted to the ICU, with a mortality rate of 32.2%. Mortality was steeply higher in those requiring ventilator support (55.4%) versus those who never required ventilation (1.4%). The total duration of hospital stay was just a day longer in patients admitted to the ICU than those who remained in wards.ConclusionMale patients above the age of 60 and with co-morbidities faced the highest rates of mortality. They should be admitted to the hospital in early stage of the disease and given aggressive treatment to help reduce the morbidity and mortality associated with COVID-19.
Reporting of MLR in many Indian journals is incomplete. Only one article managed to score 8 out of 10 among 109 articles under review. All others scored less. Appropriate guidelines in instructions to authors, and pre-publication review of articles using MLR by a qualified statistician may improve quality of reporting.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.