Evidence Based Practice, Research and Sharing Knowledge Flashcards
Which ONE of the following general practice projects would require approval by a Research Ethics Committee?
A. Face-to-face interviews with new mothers on breastfeeding
B. Policy switch to new cholesterol drug
C. Regional trial comparing acupuncture with ibuprofen
D. Postal questionnaire on online booking
E. CCG survey on alcohol consumption
Answer: C. Regional trial comparing acupuncture with ibuprofen
Justification: Research Ethics Committee review is required for clinical trials involving randomisation.
Resource: RCGP Curriculum Topic Guides; HRA research planning guidance.
Which of the following statements about urgent cancer referrals is TRUE?
A. Higher referral rates result in lower cancer mortality
B. Referral rates have no impact on mortality
C. Lower referral rates have higher conversion
D. Lower referral rates reduce mortality
E. Higher referral rates reduce conversion
Answer: A. Higher referral rates result in lower cancer mortality
Justification: Practices with low urgent referrals showed 7% increased mortality. Conversion rate unaffected.
Resource: BMJ 2015; 351: h5102.
Injecting drug users have a standardised mortality ratio (SMR) ten times that of the general population.
What is the SMR for the IDU group?
A. 90
B. 100
C. 190
D. 900
E. 1000
Answer: E. 1000
Justification: SMR = observed/expected deaths × 100. SMR of 1000 = 10× general population.
Resource: RCGP Curriculum Topic Guides (2019).
How does lead time bias affect screening test interpretation?
A. No effect on morbidity
B. No effect on mortality
C. Decreases morbidity
D. Decreases mortality
E. Increases mortality
Answer: B. No effect on mortality
Justification: Earlier diagnosis appears to increase survival but does not alter disease outcome.
Resource: RCGP Research and Surveillance guidance.
What type of chart is used to display meta-analysis results?
A. Funnel plot
B. Line chart
C. Box and whisker plot
D. Forest plot
E. Histograph
Answer: D. Forest plot
Justification: Displays individual study estimates, confidence intervals, and overall pooled result.
Resources: BMJ 2015; 351: h4028; BMJ 2001; 322: 1479.
The MUST score is used to screen for which condition?
A. Pressure ulcers
B. Malnutrition
C. Dementia
D. Falls
E. Thromboembolism
Answer: B. Malnutrition
Justification: MUST (Malnutrition Universal Screening Tool) identifies risk of malnutrition.
Resource: BAPEN MUST calculator.
Hypertension trial: 12% stroke risk in treatment group vs 20% in placebo. What is the NNT to prevent one stroke?
A. 2
B. 1
C. 8
D. 13
E. 58
Answer: D. 13
Justification: NNT = 1 / ARR = 1 / (0.20 - 0.12) = 12.5 ≈ 13
Resources: BMJ 1995; 310: 452; Bandolier NNTs.
Which symptom has the highest likelihood ratio for oesophagogastric cancer?
A. Reflux
B. Weight loss
C. Nausea/vomiting/bloating
D. Dyspepsia
Answer: B. Weight loss
Justification: LR = 5.46 with narrow CI (3.47–8.60), indicating predictive value.
Resource: RCGP Curriculum Topic Guide (2019).
Which study types are best suited to the following questions?
1. Drop in immunisation rates – explanation?
2. Comparing efficacy of PPIs?
3. Link between childhood smoked fish and UC?
Answer:
1 – Qualitative study
2 – Double-blind RCT
3 – Case-control study
Justification: Appropriate study types depend on whether question is exploratory, interventional, or observational.
Resource: RCGP Research and Surveillance.
RCT on vitamin E in CVD patients: 16.2% events (treatment) vs 15.5% (placebo). What is the relative risk?
A. 1.07
B. 0.93
C. 0.51
D. 1.05
E. 16.2
Answer: D. 1.05
Justification: RR = 0.162 / 0.155 ≈ 1.05; no significant difference.
Resource: RCGP Topic Guides.
FAMCAT case-finding algorithm is used to detect:
A. Depression
B. Myocardial infarction
C. Breast cancer
D. Familial hypercholesterolaemia
E. Pulmonary embolism
Answer: D. Familial hypercholesterolaemia
Justification: FAMCAT helps GPs identify undiagnosed FH based on EHRs.
Resource: InnovAiT 2021; 14(2): 131–136.
What defines Level III evidence?
A. Meta-analysis of RCTs
B. Controlled study without randomisation
C. Qualitative study
D. Case-control study
Answer: B. Controlled study without randomisation
Justification: Level III = quasi-experimental studies without randomisation.
Resource: Oxford CEBM 2011 Levels of Evidence.
Which feature is desirable in a cancer screening programme?
A. High sensitivity
B. Prioritise screening over treatment
C. Ignore cost if benefit exists
D. Low specificity
E. Late treatment
Answer: A. High sensitivity
Justification: Early detection requires sensitive tests to reduce false negatives.
Resource: Wilson–Jungner criteria; WHO 2003 criteria.
‘Jumping to conclusions and then looking for facts to support themʼ represents which bias?
A. Publication bias
B. Courtesy bias
C. Congruence bias
D. Outcome bias
E. Confirmation bias
Answer: E. Confirmation bias
Justification: Selecting evidence to support existing belief.
Resource: RCGP Evidence & Research Guide.
In a study of 800 children, 20 had UTI. What is the prevalence?
A. 2.2%
B. 5%
C. 0.25
D. 2.5%
E. 4%
Answer: D. 2.5%
Justification: Prevalence = 20/800 = 0.025 or 2.5%.
Resource: Basic epidemiology principles.
A 68-year-old refuses statins. You find a 10-year longitudinal study on statin users. What study type is this?
A. Case-control
B. Cohort
C. Network meta-analysis
D. RCT
E. Cross-sectional
Answer: B. Cohort
Justification: Follows a group over time, examining incidence in exposed vs non-exposed.
Resource: RCGP Topic Guide on Evidence-Based Practice.
BP readings (excluding day 1) yield an average systolic BP of:
A. 147 mmHg
B. 149 mmHg
C. 144 mmHg
D. 138 mmHg
E. 162 mmHg
Answer: B. 149 mmHg
Justification: NICE recommends discarding day 1 readings, average remainder = 149 mmHg.
Resource: NICE Hypertension guidelines.
Odds ratio of lung cancer in smokers vs non-smokers in factory workers. OR is:
A. 5
B. 15
C. 10
D. 0
E. 20
Answer: C. 10
Justification: OR = (1000×16,000) / (20,000×80) = 10
Resource: BMJ 2000; 320:1468.
Trainer writes experiences to reflect on patient care. What adjective best describes this writing?
A. Historical
B. Creative
C. Analytical
D. Biographical
E. Reflective
Answer: E. Reflective
Justification: Reflective writing explores emotions, experiences, and critical learning.
Resource: University of Nottingham – Reflective Writing.
Patient shows online articles and asks which method gives best evidence.
A. Cohort
B. RCT
C. Expert opinion
D. Case-control
E. Meta-analysis
Answer: E. Meta-analysis
Justification: Meta-analyses combine multiple RCTs – highest level of evidence.
Resource: RCGP Curriculum Topic Guide.
Which ONE of the following general practice projects would require approval by a Research Ethics Committee?
A. Patient satisfaction survey
B. Policy switch to new statin
C. Trial comparing acupuncture and ibuprofen
D. Postal survey on booking system
E. CCG alcohol survey
Answer: C. Trial comparing acupuncture and ibuprofen
Justification: Randomised controlled trials require Research Ethics Committee (REC) approval.
Resource: RCGP Curriculum Topic Guide; HRA Research Planning Guidance.
In a NORMAL DISTRIBUTION, what proportion of values lie within 2 standard deviations of the mean value?
A. 75%
B. 95%
C. 68%
D. 99%
E. 50%
Answer: B. 95%
Justification: In a normal distribution, ~68% fall within 1 SD, ~95% within 2 SD, ~99% within 3 SD.
Resource: RCGP Curriculum Topic Guide 2019.
Which ONE of the following research types would you regard MOST HIGHLY?
A. Meta-analysis
B. Case control study
C. Cohort study
D. Expert opinion
E. Randomised control trial
Answer: A. Meta-analysis
Justification: Meta-analysis systematically reviews RCTs and offers the highest level of evidence.
Resource: Centre for Evidence Based Medicine.
What is the PREVALENCE of bacteriuria in this group of 800 children (20 cases)?
A. 2.2%
B. 4%
C. 5%
D. 2.5%
E. 0.25
Answer: D. 2.5%
Justification: Prevalence = 20/800 = 0.025 = 2.5%
Resource: RCGP Curriculum Topic Guide 2019.
What is the correct description of this study: follows night-shift workers over 10 years for CVD?
A. Cohort study
B. Meta-analysis
C. Systematic review
D. Randomised control trial
E. Crossover trial
Answer: A. Cohort study
Justification: Cohort studies follow individuals over time to observe outcomes and risk exposures.
Resource: RCGP Curriculum Topic Guide 2019.
What best describes POSITIVE PREDICTIVE VALUE?
A. Proportion of true positives correctly identified
B. Proportion who test negative who do not have disease
C. Proportion of true negatives
D. Proportion of test positives who have disease
E. Number needed to treat
Answer: D. Proportion of test positives who have disease
Justification: PPV = TP / (TP + FP)
Resource: Cochrane UK. Sensitivity and specificity blog. 2019.
TRUE statement about reference intervals?
A. Not normally distributed
B. 95% fall outside range
C. Always fixed
D. Mean = 0.3 pmol/l
E. Normal range = 0.12–0.48 pmol/l
Answer: D. Mean = 0.3 pmol/l
Justification: Normal distribution defined by mean and SD; 95% within ±2 SD.
Resource: Local lab and statistical norms guidance.
Match the study to its aim:
1. Follow-up of headache → SAH risk
2. Barriers to angina revascularisation
3. Effectiveness of nasal steroids in OME
Answer:
1 – Cohort study
2 – Qualitative study
3 – Randomised controlled trial
Justification: Appropriate study designs vary with aim: observational, exploratory, interventional.
Resource: Greenhalgh T. BMJ 1997; 315:305.
What is MOST desirable in a cancer screening programme?
A. Prioritise screening over treatment
B. Ignore cost
C. Low specificity
D. High sensitivity
E. Treat late
Answer: D. High sensitivity
Justification: Sensitivity ensures early disease detection in screening.
Resource: Wilson–Jungner & WHO criteria.
Aspirin reduces stroke from 15% to 11%. What is NNT?
A. 15
B. 11
C. 26
D. 4
E. 25
Answer: E. 25
Justification: ARR = 0.15 – 0.11 = 0.04; NNT = 1/0.04 = 25
Resource: Standard NNT formula.
RCT: dementia rate 4% control vs 2% treatment. What is the NNT?
A. 200
B. 2
C. 50
D. 98
E. 25
Answer: C. 50
Justification: ARR = 0.04 – 0.02 = 0.02; NNT = 1 / 0.02 = 50
Resource: RCGP Curriculum Topic Guides.
Injecting drug users have 10× mortality. What is the SMR?
A. 90
B. 100
C. 190
D. 900
E. 1000
Answer: E. 1000
Justification: SMR = (observed / expected) × 100; 1000 = 10× standard mortality
Resource: RCGP Curriculum Topic Guide.
Cholesterol 5.6 mmol/l reduced by 25% using ‘Redchol’. New value?
A. 4.8
B. 4.2
C. 3.2
D. 3.8
E. 4.0
Answer: B. 4.2
Justification: 25% of 5.6 = 1.4; 5.6 – 1.4 = 4.2 mmol/l
Resource: Simple % reduction calculation.
Research data:
Test +: 80 (cancer), 40 (no cancer)
Test –: 20 (cancer), 60 (no cancer)
What is specificity?
A. 20%
B. 33%
C. 60%
D. 66%
E. 75%
Answer: C. 60%
Justification: Specificity = TN / (TN + FP) = 60 / (60 + 40) = 0.6
Resource: RCGP Topic Guide.
TUM1 test:
+ cancer: 32; + no cancer: 20
– cancer: 8; – no cancer: 180
What is the specificity?
A. 50%
B. 96%
C. 62%
D. 90%
E. 80%
Answer: D. 90%
Justification: Specificity = TN / (TN + FP) = 180 / (180 + 20) = 0.9
Resource: RCGP Curriculum Guide.
Follow 40–50-year-olds with night shifts for 10 years. Study type?
A. Cohort
B. Meta-analysis
C. Systematic review
D. Crossover trial
E. RCT
Answer: A. Cohort
Justification: Follows population over time with exposure and outcome
Resource: RCGP Curriculum Topic Guide.
Which term best describes test identifying those WITH disease?
A. NPV
B. Sensitivity
C. Odds Ratio
D. Specificity
E. Absolute risk
Answer: B. Sensitivity
Justification: Sensitivity = TP / (TP + FN), ability to detect true positives
Resource: BMJ 2003; 327:716.
Which condition shows BIMODAL age distribution?
A. Pityriasis rosea
B. Hodgkin’s lymphoma
C. Osteoarthritis
D. Bronchiolitis
E. Prostate cancer
Answer: B. Hodgkin’s lymphoma
Justification: Bimodal peaks in young adults and elderly
Resource: Epidemiology of lymphoma data.
Which figure best shows flow of patients in RCTs?
A. Funnel plot
B. Histogram
C. Run chart
D. CONSORT diagram
E. Population pyramid
Answer: D. CONSORT diagram
Justification: CONSORT shows participant flow and attrition in RCTs
Resource: www.consort-statement.org
Which figure is best described as a funnel plot?
A. Normal distribution
B. Forest plot
C. Funnel plot
D. Box and whisker
E. Population pyramid
Answer: C. Funnel plot
Justification: Used in meta-analysis to assess heterogeneity and bias
Resource: BMJ 2011; 343: d4002.
Which SINGLE condition does the FAMCAT clinical case-finding algorithm identify?
A. Pulmonary embolism
B. Myocardial infarction
C. Depression
D. Familial hypercholesterolaemia
E. Breast cancer
Answer: D. Familial hypercholesterolaemia
Justification: The Familial Hypercholesterolaemia Case Identification Tool (FAMCAT) identifies patients with familial hypercholesterolaemia.
Resource: Evans PH, Rafi I, Hayward J. InnovAiT 2021; 14(2): 131–136.
A study shows 10% cardiovascular mortality in control vs 5% in treatment group. What is the NNT?
A. 2
B. 5
C. 10
D. 20
E. 50
Answer: D. 20
Justification: Absolute risk reduction = 10% - 5% = 5%. NNT = 1 / 0.05 = 20.
Resource: TheNNT.com – TheNNT, explained.
Forest plot comparing pulmonary rehabilitation to control in COPD shows which result?
A. Both Man and Behnke’s studies show significant benefit
B. Each study shows significant benefit
C. Murphy’s study shows significant benefit
D. No studies show significant benefit
E. Only Behnke’s study shows significant benefit
Answer: A. Both Man and Behnke’s studies show significant benefit
Justification: Significant results occur when confidence intervals do not cross the midline in a forest plot.
Resource: Cochrane UK. How to read a forest plot?
Which programme offers international placements outside Europe for GPs?
A. Doboj Conference exchange
B. Family Medicine-360
C. Erasmus
D. The Nordic GP congress exchange
E. Hippokrates
Answer: B. Family Medicine-360
Justification: This programme offers up to 4-week placements globally for early-career GPs.
Resource: Morris DCB, Pettigrew DLM. InnovAiT 2024; 17(4): 166–170.
Which p-value indicates statistical significance?
A. p > 1
B. p < 0.05
C. p > 0.05
D. p < 1
E. p = 1
Answer: B. p < 0.05
Justification: A p-value < 0.05 indicates a significant difference between groups.
Resource: Greenhalgh T. BMJ 1997; 315: 422.
Group A stroke risk = 12%, Group B = 20%. NNT?
A. 2
B. 1
C. 8
D. 13
E. 58
Answer: D. 13
Justification: ARR = 0.08; NNT = 1 / 0.08 = 12.5 ≈ 13.
Resource: Cook R, Sackett D. BMJ 1995; 310: 452.
HIV prophylaxis study: OR for seroconversion?
A. Prophylaxis increases risk
B. OR can’t be calculated
C. RR is better than OR
D. OR = 0.5
E. OR = 2
Answer: D. OR = 0.5
Justification: OR = (8 × 200) / (20 × 160) = 0.5, indicating reduced risk.
Resource: Bland J, Altman D. BMJ 2000; 320: 1468.
Best study design to assess if PPIs cause osteoporosis?
A. Cohort study
B. Crossover study
C. Case-control study
D. Meta-analysis
E. Cross-sectional study
Answer: D. Meta-analysis
Justification: Meta-analyses offer the highest level of evidence.
Resource: Oxford CEBM. Levels of Evidence.
Best audit criterion for repeat PPI prescribing?
A. Count monthly repeat prescriptions
B. Max 4 PPI prescriptions/month
C. Count patients on PPI > 6 months
D. Refresher course for all GPs
E. Record reason for starting PPI
Answer: E. Record reason for starting PPI
Justification: Good audit criteria relate to quality of care processes.
Resource: RCGP audit resources.
FC test: 124 TP, 12 FP, 8 FN, 260 TN. Sensitivity & PPV?
A. 96% (260/272), 91% (124/136)
B. 94% (124/132), 91% (124/136)
C. 96% (260/272), 97% (260/268)
D. 91% (124/136), 97% (260/268)
E. 94% (124/132), 96% (260/272)
Answer: B. 94% (124/132), 91% (124/136)
Justification: Sensitivity = 124/132; PPV = 124/136
Resource: Standard diagnostic test metrics.
Antipsychotic use linked to 30% higher VTE risk. Most accurate interpretation?
A. 32 of 100 on antipsychotics get VTE
B. Risk is 32× higher
C. Odds of VTE are 1:3
D. Risk is 1.3× that of non-users
E. None of the above
Answer: D. Risk is 1.3× that of non-users
Justification: Relative risk of 1.3 = 30% increased risk compared to non-users.
Resource: Parker P, Coupland C, Hippisley-Cox J. BMJ 2010; 341: c3245.
Oesophagogastric cancer LR data: most predictive symptom?
A. Reflux
B. Need more information
C. Nausea/vomiting/bloating
D. Weight loss
E. Dyspepsia
Answer: D. Weight loss
Justification: Highest LR (5.46) with CI not crossing 1 indicates predictive value.
Resource: RCGP Curriculum Topic Guides 2019.
What defines statistical power?
A. 1 – type 2 error
B. 1 – type 1 error
C. 1 – SD
D. √ SD
E. Variance
Answer: A. 1 – type 2 error
Justification: Power = ability to detect true effect = 1 – β (type 2 error).
Resource: Standard statistical definitions.
Test: a = 30, b = 5, c = 20, d = 45.
Which parameter is:
30 / (30 + 20)?
30 / (30 + 5)?
45 / (45 + 5)?
Answer:
Sensitivity = 30 / (30 + 20)
PPV = 30 / (30 + 5)
Specificity = 45 / (45 + 5)
Justification: Standard diagnostic test formulas for a 2×2 table.
Smoking cohort study:
Smokers: 3000 lung cancer, 24,000 other deaths
Non-smokers: 100 lung cancer, 20,000 other deaths
NNH for smoking and lung cancer?
A. 23
B. 2
C. 9
D. 6
E. 7
Answer: C. 9
Justification: NNH = 1 / (0.111 – 0.005) = ~9.4
Resource: Standard NNH calculation.
What is ‘middle-ground researchʼ?
A. Quantitative by specialists
B. Patient-centred care
C. RCT-based medicine
D. Local evaluations
E. Theoretical academic research
Answer: B. Patient-centred care
Justification: Focuses on timely, realistic, patient-focused outcomes.
Resource: Mercer S et al. InnovAiT 2018; 11(9): 495–499.
Forest plot: DOACs vs warfarin in AF stroke prevention.
Which are TRUE?
A. Warfarin more effective
B. Heterogeneity is significant
C. 3/4 trials show no significant difference
D. DOACs worse for bleeding
E. DOACs more effective overall
Answer: C and E
Justification: 3/4 trials not significant, but meta-analysis favours DOACs overall.
Resource: BMJ 2015; 351: h4028; BMJ 2001; 322: 1479.
Study: oligospermia vs control group, both questioned on phone use. What is study type?
A. RCT
B. Quasi-experimental
C. Cohort
D. Case-control
E. Crossover
Answer: D. Case-control
Justification: Retrospective comparison of exposure in cases vs controls.
Resource: Standard epidemiological study types.
Trial of new topical anaesthetic vs placebo, blinded to subject and researcher. What type?
A. RCT
B. Crossover
C. Case series
D. Case-control
E. Cohort
Answer: A. RCT
Justification: Randomised, double-blind interventional trial.
Resource: Oxford CEBM – Study designs.
USS screening for ovarian cancer:
TP = 161, FP = 1634, FN = 221, TN = 48607
Which are CORRECT?
A. Sensitivity = 42%
B. Specificity = 97%
C. PPV = 9%
D. NPV = 0.5%
E. 1634 had unnecessary surgery
Answer: A, B, C, E
Justification: Sensitivity = 161/382; Specificity = 48607/50241; PPV = 161/1795
NPV is 99.5%, not 0.5%. 1634 false positives underwent surgery.
Resource: Diagnostic accuracy metrics.
