Diagnostic Test Accuracy Questions and Answers

Diagnostic Test Accuracy Questions and Answers evidence based medicine why are critical appraisal skills important? what is meant by the evaluation bypass? this is where due to enthusiasms, convictions, commercial pressure cause unevaluated procedures to be used in health services without being underpinned with good quality evidence what evidence is there for commonly used therapies? not a lot used examples of diagnostic tests demographics: age/gender/ethnicity/social class symptoms questionnaires (for example the mini mental state examination) physical signs blood tests imaging: x-ray, ultrasound, CT, MRI biopsy post-mortems when is a diagnostic test useful? a test is useful if it changes our ability to predict whether a person has a condition or not - this then guides treatment choices essentially: the probability of the conditions before the test should be altered by the test result reasons for carrying out a diagnostic test detection/exclusion; rule-in/out a disease to make sure we are not missing something important (reassurance) for medico-legal or financial reasons bc we are following a protocol bc it is easier to order a test than think about why we are doing it ultimately we should be doing a diagnostic test bc the pt is better off for having it how a test is useful: abdo pain example - shows how asking gender is a diagnostic test Gender of the pt has altered the probability of ectopic pregnancy and endometriosis H+E should be considered as tests in their own right why is improved patient outcome not guaranteed by improved test accuracy? no - need to consider harms and costs approaches to making a diagnosis patient appearance H+E investigations response to tx in GP what % of diagnoses are established by H+E? 88% in hospital what % of diagnoses are established by H+E? 50% by H alone and 75% by H+E mammography and BC- tests being used at different points in the clinical pathway Screening programme- well population of women invited for screening test Positive- further testing- make a diagnosis of Bc not Diagnosis- have another mammography + biopsy- FNAC Women who receive a diagnosis- surveillance for ongoing disease or disease recurrence 3 different applications of the same test: mammography Screening- most women will be well Repeat mammography- filtered out those without so Population will have higher prevalence. Screening will have picked up the more severe cases. Changed during diagnostics compared to screening When used for surveillance- women have already had tx. May have had mastectomy/lumpectomy. Population has changed in terms of women who are likely to have disease, disease recurrence, type of disease, also what mammography is being used on. Women may have had surgical procedures. Scar tissue might affect the accuracy of mammography. Accuracy of diagnostic test cannot always be carried onto different contexts. Evaluation in screening context Transferability of accuracy likely to be low when considered as surveillance testing what questions are involved in evaluation of tests? avoid the evaluation bypass Assess whether intevrvention is likely to benefit pts. Cost considerations- two options with same efficacy., go with least costly. Without evaluation- varies from intervention to intervention More likely that unevaluated test will make its way into health services compared to unevaluated tx. Happens medical tests are not regulated in the same way as tx ensures that testing kits are safe and measure what they say they do (for e.g. a cholesterol test must measure cholesterol levels) but whether that measurement helps diagnose or predict a disease is not regulated. tests will still give false positive and false negative results and no-one has to produce evidence about the overall benefit of the test (for patients) or how it is best used. Screening/diagnostic context. Should the test be used alongside others? Before or after? what is a big contributing factor to inappropriate testing? commercial pressure Overdiagnosis Public and HCPs considered testing as beneficial to detect disease early People are starting to question the harms of testingcon consequences of inappropriate testing waste of money that could be better used elsewhere anxiety, discomfort, pain, unecessary further testing and tx with associated adverse effects tests make errors, even if they are correctly performed effects of false positives The effect of FP and FN will depend