Evidence Based Laboratory Medicine Notes
Evidence Based Laboratory Medicine
The Principles
- Definition: Evidence-Based (Laboratory) Medicine (EBLM) is a new term that embodies current best practice. It is defined as:
- "The conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individuals" (Sackett et al., 1996).
- Background:
- Developed in the 1990s.
- Widely accepted by health systems worldwide.
- Principles employed in laboratory medicine.
- Ensures a well-informed medical standard, resulting in optimal patient care (Guyatt, G et al., 1992, JAMA 268(17):2420-5).
- Integration necessities:
- Determining where laboratory medicine fits into the care pathway.
- Determining when testing is appropriate.
- Assessing the type and quality of evidence required to demonstrate the test's clinical utility.
- Understanding how the test impacts clinical actions.
- Identifying where changes in the care pathway will occur.
- Determining where benefit/value can be achieved.
- Goal: To provide the best laboratory procedures that benefit health care.
- Principles:
- Formulating the question.
- Searching for the evidence.
- Appraising the evidence.
- Applying the evidence.
- Assessing (auditing) the practice.
- Modifying the practice.
The Process
- Principles Diagram:
- Identify the question → Search for evidence → Critically appraise evidence → Apply to practice → Audit practice → Modify practice.
Why EBLM is Needed
- A request for a diagnostic test is part of clinical decision-making.
- Relevance of the test to the clinical situation is crucial.
- Critical appraisal of evidence is necessary in terms of quality.
- Evidence is continually evolving, especially for new tests.
- Quality in clinical practice is a political agenda item.
Diagnostic Test Considerations
- A diagnostic test is considered an intervention.
- It should be requested only when a question is being asked, and the result will provide an answer.
- The best evidence for the value of a test is improved outcome.
Contrasting Perspectives on Medical Testing
- Too much testing vs. Insufficient testing:
- Expensive testing vs. Laboratories expected to generate income.
- Tests have little impact on outcomes vs. Test results are essential for clinical decision making.
Emphasis on Outcomes
- The result of an intervention:
- Health outcome.
- Economic outcome.
- Differing expectations:
- Patient:
- Promptness.
- Effective treatment.
- Symptom relief.
- Service provider:
- Promptness.
- Delivery of effective care.
- Workload.
- Economic factors.
Why Focus on Outcomes?
- Increasing concerns about the quality of health care.
- Cost concerns.
- Increasing patient expectations.
- New delivery models (e.g., point of care testing).
- Continual availability of new evidence.
- Discovery of new technologies.
- Resource allocation is an increasing problem.
EBLM's Central Role
- EBLM is at the heart of what we do.
- Quality assurance.
- Audit.
- Research & Development.
- Education & training.
Factors to Achieve Quality in Health Care
- Safe – avoiding harm, minimization of risk.
- Patient-centered – respecting individual needs and values.
- Timely – minimizing delays; performance targets.
- Efficient – avoiding waste.
- Equitable – equal quality, guidelines.
- Effective – evidence-based, maximization of benefit.
- Reasonable cost (Institute of Medicine, 2001).
Findings from a Review of Health Care in the US
- Medical errors:
- Accounted for approximately 100,000 deaths per year.
- Costing approximately $$2 billion per year.
- Costs of poor quality: Approximately 30-40% of total health care spending.
- Conclusion: The health service was fragmented, dangerous, and wasteful (Institute of Medicine, 2000).
Laboratory Medicine Point of View
- Test repertoire is increasing.
- Workload is increasing.
- There is uncontrollable demand.
- Unnecessary testing occurs.
- Utilization of tests is varied.
- Continually new evidence needs consideration.
- Translation of evidence to practice is slow.
- There is a perception of limited investment.
Clinical Outcomes from Medical Testing
- Clinical outcomes:
- Death.
- Disease.
- Disability.
- Discomfort.
- Dissatisfaction.
- Testing outcomes:
- Faster decision making.
- Earlier treatment.
- Improved compliance/standardization.
- Reduced complications.
- Reduced empirical/unnecessary treatment.
- Reduced re-admission.
- Improved survival.
- Improved patient satisfaction.
Economic Outcomes
- Cost to the patient.
- Cost to the provider.
- Cost to society.
- Surrogates:
- Number of clinic visits.
- Length of stay.
- Re-admission rate.
- Intervention rate.
- Working days lost.
- Productive years gained.
Outcomes of Testing
- Impact of the test on clinical outcome.
- Analytical performance and impact on outcomes.
- Modality of testing and impact on outcomes.
- Health economics of testing.
Decision Making
- The pyramid of evidence includes:
- Investment decision.
- Cost-effectiveness.
- Technical performance.
- Organizational impact.
- Clinical impact.
- Diagnosis.
- Treatment.
- Health outcome.
- Diagnostic performance.
Implementation of EBLM
- EBLM Principles:
- Define the Question.
- Search for the Evidence.
- Appraise that Evidence.
- Apply the Evidence.
- Audit the Practice.
- Modify the Practice.
- Process Diagram:
- Identify the question → Search for evidence → Critically appraise evidence → Apply to practice → Audit practice → Modify practice.
Step 1: Importance of Good Questions
- Articulates the problem in a clear and structured way.
- Central to reviewing the literature.
- Key to future study design.
Things to Consider
- Patient and disease:
- Background to a condition.
- Aetiology.
- Risk factors.
- Patient-specific factors.
- Test:
- Availability.
- Screening.
- Diagnosis.
- Prognosis.
- Accuracy.
- Clinical outcome.
- P - patient problem or population being addressed.
- I - test or treatment intervention (aka indicator).
- C - comparator or control – to which intervention is compared.
- O - outcome – depends upon the nature of the question.
P - Patient Problem or Population
- Primary problem.
- Patient group.
- Disease or health status.
- Age, race, sex, previous ailments, current medications.
I - Test or Treatment Intervention
- What you plan to do for that patient.
- Specific diagnostic test, treatment, medication.
- Recommend a procedure.
- This is the main consideration for the patient.
C - Comparator or Control
- The alternative.
- Can be omitted.
O - Outcome
- What you plan to accomplish/improve/effect.
- Should be measurable (e.g., eliminate/relieve symptoms, improve function, improve prognosis).
Type of Question
- Therapy/Prevention.
- Diagnosis.
- Aetiology (cause).
- Prognosis.
- Analytical.
- Operational.
- Economic.
PICO Variants
- PPICO
- Population
- Previous tests
- Intervention
- Comparator
- Outcome
- PICOT
- Population
- Intervention
- Comparator
- Outcome
- Time
- PICCO
- Population
- Intervention
- Comparator
- Change management
- Outcome
Examples of Questions in each Category
Q1: Aetiology, Risk Factors and Prevalence
- Case 1: A 34-year-old woman runs 5 miles every day, collapses after 6 weeks with chest pains. Troponin level is 0.02 ng/L (upper reference interval 0.01 ng/L).
- Question: Does sustained physical exercise lead to increased serum troponin level?
- PICO Breakdown:
- P: Women
- I: Sustained physical exercise
- C: Normal lifestyle exercise
- O: Serum troponin level
- Full Question: In women ~35 years old, will sustained physical exercise increase serum troponin levels compared to women with normal lifestyle exercise?
Q2: Diagnosis
- Case 2: Microbiologist concerned about the number of urines sent from primary care for suspected UTI, believes dipstick test might help.
- PICO Breakdown:
- P: Patients presenting to GPs with symptoms suggestive of UTI
- I: Dipstick urinalysis for leucocyte esterase and nitrite
- C: Lab tests for microscopy, culture, and sensitivity
- O: Correct diagnosis of urinary tract infection
- Full Question: In patients presenting to GP with UTI symptoms, can dipstick urinalysis for LE and nitrite be used to rule in/out a UTI diagnosis, reducing the number of urines sent to the lab?
Q3: Prognosis
- Case 3: Oncologist reads about a new gene panel test to predict breast cancer recurrence and consults a pathologist.
- PICO Breakdown:
- P: Women diagnosed with breast cancer
- I: Tissue expression of a panel of genes
- C: Grading of cancer based on histology, tumor size, biomarkers, etc.
- O: Recurrence of disease
- T: 5-year follow-up period
- Full Question: How well can the expression of a panel of genes in primary tumor at diagnosis in women with breast cancer predict the likelihood of recurrence?
Q4: Treatment: Guiding Therapy
- Case 4: Primary care physician questions self-monitoring of blood glucose in type 2 diabetes management and consults a clinical biochemist.
- PICO Breakdown:
- P: Patients with type 2 diabetes
- I: Self-monitoring of blood glucose (SMBG) as part of an integrated package
- C: No self-monitoring of blood glucose
- C: Change management: Adjust dosage according to blood glucose
- O: Control of blood glucose concentration (reduction in complications)
- Full Question: Does including SMBG as part of an integrated package in managing type 2 diabetes lead to better glycemic control and a reduction in complications?
- Case 5: A prostate risk management program advocates using a PSA assay meeting specific quality criteria. The current assay doesn't meet the criteria.
- PICO Breakdown:
- P: Men receiving PSA tests as part of a prostate cancer risk management program
- I: Serum PSA testing meeting the agreed quality criteria with respect to accuracy
- C: Serum PSA testing not meeting the agreed quality criteria with respect to accuracy
- O: Unnecessary biopsies avoided and cancers detected
- Full Question: Does the use of a PSA test that meets agreed quality criteria with respect to accuracy lead to improved test performance (through avoidance of unnecessary biopsies) as well as detecting more cancers?
Q6: Operational
- Case 6: The health management organization wants to investigate the costs and benefits of transferring more diabetes management into the community setting and wants to assess point-of-care HbA1c testing.
- PICO Breakdown:
- P: Patients with diabetes
- I: Point of care testing (POCT) for HbA1C
- C: Laboratory testing for HbA1c
- C: Clinical decision on result at the time of intervention
- O: Glycaemic control
- Question: Will the use of POCT for HbA1C as part of an integrated package of care for the management of diabetes improve glycaemic control?
Q7: Economic
- Case 7: The clinical biochemist wants to introduce brain natriuretic peptide (BNP) testing to rule out heart failure and reduce the need for echocardiography.
- PICO Breakdown:
- P: Patients presenting to GPs with symptoms suggestive of heart failure
- I: BNP testing
- C: No BNP testing
- O: Total cost of diagnosing heart failure
- T: Over a period of one year
- Question: Does the introduction of BNP as a first-line test in the diagnostic triage pathway for patients with suspected heart failure attending their GP reduce the total cost involved in making the diagnosis of heart failure?
Step 2: Searching for the Evidence
- Internet (good websites).
- PubMed.
- Literature:
- Peer-reviewed journals.
- Journal quality.
- Study quality.
- Clinical trials.
- Research.
Searching Tips
- OR: One of a choice of words.
- AND: All words requested.
- NOT: Excludes certain words.
- ADJ or NEAR: Words must be next to each other or within about 5 words.
- *: Truncate words with variable endings.
- (): Group words.
- Limits: Restrict language, publication date, humans, etc.
- [ti] or :ti: The word must be in the title.
- MeSH: Medical Subject Headings (controlled vocabulary in PubMed or Cochrane).
- Presenting critically appraised data.
- Systematic review: A summary of the literature that uses defined methods for the performance of a comprehensive literature search and critical appraisal of the studies identified from the search.
- Meta-analysis: A set of statistical techniques for combining similar information from different studies to derive an overall estimate of a treatment effect or characteristics of a test.
Hierarchy of Strength of Evidence
- EBM Levels of Evidence Pyramid (from top to bottom):
Meta-analysis → Systematic Reviews → Randomized Controlled Trials → Cohort Studies → Case Control Studies → Case Series & Case Reports → Animal Studies/Laboratory Studies (Guyatt et al. 2000). - RCTs and controlled clinical trials: help answer treatment and diagnosis questions.
- Cohort Studies: help answer prognosis and etiology/harm questions.
- Case Control Studies.
- Case Series & Case Reports.
- Animal Studies/Laboratory Studies.
Step 3: Appraisal of the Evidence
- Critical Appraisal: The process of carefully and systematically examining research to judge its trustworthiness and its value and relevance in a particular context.
- 'Weighing up the evidence to assess its validity (closeness to the truth) and usefulness (clinical applicability)' (Buris A. What is critical appraisal? http://www.medicine.ox.ac.uk).
- Sackett DL, Haynes RB. On the need for evidence-based medicine. Evid Based Med 1995;1:4-5.
Appraisal of the Evidence: Key Questions
- Was there a clear question for the study to address (PICO)?
- Are the results valid?
- Was there a comparison with an appropriate reference standard?
- Did all patients get the diagnostic test and the reference standard (bias)?
- Could the results of the test of interest have been influenced by the results of the reference standard (bias)?
- Is the disease status of the tested population clearly described?
- Were the methods for performing the test described in sufficient detail?
- What are the results?
- How sure are we about these results? (statistical analyses).
- Can the results be applied to your patients/ the population of interest?
- Can the test be applied to your patient or population of interest?
- Were all outcomes important to the individual or population considered?
- What would be the impact of using this test on your patients/population?
What is Critical Appraisal? - Considerations
- Not a negative dismissal of any piece of research; but a balanced assessment of benefits and strengths of research against its flaws and weaknesses.
- Not assessment of results alone; but assessment of research process and results.
- Considers both quantitative and qualitative aspects of research.
- To be undertaken by all health professionals as part of their work, not just expert researchers/statisticians.
Why Should We Critically Appraise?
- Published research is not always reliable – incorrect conclusions can be drawn from poor evidence.
- Published research is not always relevant.
- We need a systematic framework to interpret research.
Confidence in Results: Key Questions
- Could they have occurred by chance?
- Confidence intervals?
- Statistical significance?
- Adequate description of the data collection methods?
- Appropriate, clearly described, and justified methods of analysis?
Published Research Findings - May Be False (Ioannidis 2005)
| Scenario | % Time Eventually True |
|---|
| Confirmatory meta-analysis of good quality RCTs | 85 % |
| Adequately powered RCT with little bias and pre-study 50 % chance effective | 85 % |
| Meta-analysis of small inconclusive studies | 41 % |
| Underpowered but well-performed phase I/II RCT | 23 % |
| Underpowered, but poorly performed phase I/II RCT | 17 % |
Step 4: Apply the Evidence
- Can follow the PICO framework again:
- P: Does the evidence apply to the patient/population and setting in the original question?
- I: Does my method have the required performance, and am I using the right decision (cut-off) values?
- C: Is the current standard of care similar to the control groups featured in the evidence?
- C: What is the change in practice required, and is it achievable?
- O: What were the outcomes shown in the evidence, and are they appropriate and achievable?
Can the Test Be Applied to Your Patient or Population of Interest?
- Age, sex, ethnicity, and spectrum bias.
- Equipment, reagent, and staff resources.
- Level and availability of expertise required to interpret the tests.
- Current practice and availability of services.
- Timings of blood collection and test measurement.
- Training requirements.
What Would Be the Impact of Using This Test on Your Patients/Population?
- Benefits vs. cost / harm.
- Consider clinical needs.
- Quality specifications.
- Does the laboratory service meet the established standard of care?
- Is the process of care optimal and make best use of the laboratory service, including process change?
- Are the best health outcomes being achieved?
- Economic criteria.
Step 5: Audit the Practice
- The audit assesses whether the application of new evidence and the introduction of a new test has led to improved outcome.
- Also serves purposes of:
- Training.
- Effective management.
- (Continuous) quality improvement.
- Maintenance of good practice.
Step 6: Modify the Practice
- EBLM is a continual process.
- Changes in technology.
- Research discoveries.
- Changes in health priorities.
- Comparison (validation) of a new diagnostic test to current gold standard –
- Specificity, sensitivity, cost, turn-around time, intervention, sample type.
Evidence Based Laboratory Medicine - Summary
- Identify the question → Search for evidence → Critically appraise evidence → Apply to practice → Audit practice → Modify practice.
Guidelines for the Process of EBLM
- There is a need for guidelines.
- Summarized in Oosterhuis et al. 2004