Clinical epidemiology

1
Clinical epidemiology

• I am not an expert but I will do my best!
• If you know and I dont - speak up!
• This is the basics - there is a major new
  literature to become familiar with and new skills
  to learn.

www.bradfordvts.co.uk
2
EVIDENCE BASED MEDICINE

• EBM is an approach to practicing medicine in
  which the clinician is aware of the evidence in
  support of his / her clinical practice, and the
  strength of that evidence.

3
EVIDENCE BASED HEALTHCARE

• Evidence based health care promotes the
  collection, interpretation, and integration of
  valid, important and applicable patient-reported,
  clinician-observed and research derived evidence.
  The best available evidence, moderated by patient
  circumstances and preferences, is applied to
  improve the quality of clinical judgements and
  facilitate effective healthcare.

4
EVIDENCE BASED MEDICINE
FORMULATE QUESTION
EVALUATE PERFORMANCE
EFFICIENTLY TRACK DOWN BEST AVAILABLE EVIDENCE
IMPLEMENT CHANGES IN CLINICAL PRACTICE
CRITICALLY REVIEW THE VALIDITY AND USEFULNESS OF
THE EVIDENCE
5
AN EXAMPLE OF EBM AND AN INDIVIDUAL PATIENT

• THE PROBLEM
• New patient. 11 year old girl. Generalised tonic
  clonic fits aged 5-8. On sodium valproate, no
  fits 3 years.
• Parents disappointed with progress at school.
  Generally lethargic. Its the tablets - what
  would happen if we stopped them?

6
SOLUTION 1

• Do what the last consultant said.
• Advantage
• Consistency
• Disadvantages
• Getting the notes
• Do the letters say what to do long term?
• Is that advice now out of date?
• Was it correct at the time?

7
SOLUTION 2

• Get a new consultant opinion
• Advantages
• Local secondary care contact made if needed in
  the future
• Advice will be current if the question is asked
  and answered
• Disadvantages
• Time till appointment.
• Getting to hospital
• Is the advice based on up to date knowledge or
  just clinical experience?
• Resources

8
SOLUTION 3

• CAN I FIND THE ANSWER MYSELF?

9

• P EHRHARDT WI FORSYTHE
• LEEDS GENERAL INFIRMARY
• DEVELOPMENTAL MEDICINE AND CHILD NEUROLOGY 31 (5)
  633-9 OCT 1989
• From a total group of 640 children with grand mal
  seizures, 187 who became seizure free for three
  consecutive years on monotherapy which was then
  discontinued have been followed for between 1 and
  14 years. Relapse occured in 22 children (12)
  and was related to age at presentation only four
  of 89 children with primary grand mal seizures
  who had presented after the age of 3 years
  relapsed, compared with 12 of the 45 who had
  presented before their third birthday. Children
  who had had more seizures were at greater risk of
  relapse. EEGs were not useful in predicting
  prognosis, whether taken at presentation or
  before withdrawal of treatment.

10
BARRIERS TO IMPLIMENTING EBM FOR INDIVIDUAL
PATIENTS

• Cost
• Time
• Its new
• What will patients think
• What will my colleagues think
• I dont have the skills to assess the quality of
  the evidence

11
WHY THE MOVE TO EBM?

• RANDOMISED CONTROLLED TRIALS PRE-1960 WERE
  ODDITIES
• REVIEWS AND META-ANALYSES ARE BECOMING AVAILABLE
  AS ACCESSIBLE DIGESTS OF EVIDENCE
• ACCESS TO EVIDENCE VIA I.T.
• METHODOLOGICAL ADVANCEMENTS E.G. NUMBERS NEEDED
  TO TREAT

12
THE ALTERNATIVE TO EBM

• UNSYSTEMATIC CLINICAL OBSERVATIONS OVER TIME ARE
  A VALID WAY OF BUILDING AND MAINTAINING KNOWLEDGE
  ABOUT PROGNOSIS, THE VALUE OF TESTS, THE
  EFFICACY OF TREATMENT
• UNDERSTANDING BASIC MECHANISMS OF DISEASE AND
  PATHOPHYSIOLOGY ARE A SUFFICIENT GUIDE FOR
  CLINICAL PRACTICE
• THOROUGH TRADITIONAL MEDICAL TRAINING PLUS COMMON
  SENSE IS SUFFICIENT TO EVALUATE NEW TESTS AND
  TREATMENTS
• CONTENT EXPERTISE AND CLINICAL EXPERIENCE ARE
  SUFFICIENT FROM WHICH TO GENERATE CLINICAL
  GUIDELINES

13
EBM IS ABOUT ...

• CLINICAL EXPERIENCE, DIAGNOSTIC SKILLS AND
  CLINICAL INSTINCT ARE A NECESSARY PART OF A
  COMPENTENT PHYSICIAN.
• HOWEVER, CLINICAL PRACTICE BASED SOLELY UPON
  CLINICAL EXPERIENCE BECOMES TOMORROWS BAD
  JOKE.
• RATIONAL TREATMENT BASED SOLELY UPON BASIC
  PATHOLOGICAL PRINCIPLES MAY IN FACT BE INCORRECT,
  LEADING TO INACCURATE TREATMENT.
• UNDERSTANDING CERTAIN RULESOF EVIDENCE IS
  NECESSARY TO CORRECTLY INTERPRET LITERATURE ON
  CAUSATION, PROGNOSIS, DIAGNOSTIC TESTS AND
  TREATMENT STRATEGY.

14
EBM SKILLS - STATISTICS

• CHANCE - p 1 in 20 (0.05).
• gt 1 in 20 (0.051) not significant
• lt 1 in 20 (0.049) statistically significant
• CONFIDENCE INTERVALS
• what is the range of values between which we
  could be 95 certain that this result would lie
  if this intervention was applied to the general
  population

15
EBM SKILLS - A BASIC INTRODUCTION

• CHANCE, BIAS, CONFOUNDING VARIABLES

16
TYPES OF STUDY - HYPOTHESIS FORMING

• CASE REPORTS / CASE SERIES
• CROSS SECTIONAL / PREVALENCE STUDIES measure
  personal factors disease states - hypothesis
  FORMING - cannot indicate cause effect
• CORRELATIONAL / ECOLOGICAL / GEOGRAPHIC STUDIES.
  prevalence /or incidence measurement in one
  population c/w another pop.

17
TYPES OF STUDY - HYPOTHESIS TESTING

• CASE CONTROL STUDIES

18
CASE CONTROL EXAMPLE -SMOKING LUNG CANCER

• DISEASE
• Cases Controls
• EXPOSURE Yes a b
• EXPOSURE No c d
• Odds Ratio ad/bc (1 no association, gt 1
  possible association, lt 1 protective
  effect)
• DISEASE
• Cases Controls
• (lung cancer)
• EXPOSURE Yes 56 230
• (smoking) No 7 246
• The odds ratio would therefore be 56 x 246
  13776 8.6.
• 
  7 x 230 1610

19
TYPES OF STUDY - HYPOTHESIS TESTING

• COHORT STUDIES

20
COHORT STUDIES

• 
  OUTCOME
• Yes No
• Exposed a b
• Not exposed c d
• Attributable risk (absolute risk or risk
  difference)
• "What is the incidence of disease attributable to
  exposure"
• Answer a - c.
• Relative risk "How many times are exposed persons
  more likely to develop the disease, relative to
  non-exposed persons?" i.e. the incidence in the
  exposed divided by the incidence in the
  non-exposed.
• This is expressed as a divided by
  c .
• ab
  cd

21
COHORT STUDY EXAMPLE

• Deep vein thromboses (DVT) in oral contraceptive
  users. (Hypothetical results).
• OUTCOME (DVT)
• Yes No
• Exposed ( on oral contraceptive ) 41
  9996
• Not exposed (not on o.c.) 7
  10009
• These results would give an attributable risk of
  34 and a relative risk of 6 - significantly large
  enough numbers to indicate the possibility of a
  real association between exposure and outcome.
  However, the possibility of biases very often
  arises.

22
RANDOMISED CONTROLLED TRIALS
23
RANDOMISED CONTROLLED TRIALS

• 
  OUTCOME
• Yes No
• Comparison intervention a b
• Experimental intervention c d
• Relative risk reduction How many fewer
  patients will get the outcome measured if they
  get active treatment versus comparison
  intervention
• a /ab - c/cd
• a/ab
• Absolute risk reduction What is the size of
  this effect in the population
• a/ab - c/cd

24
RCT EXAMPLE - 4S STUDY

• STABLE ANGINA OR MYOCARDIAL INFARCTION MORE THAN
  6 MONTHS PREVIOUSLY
• SERUM CHOLESTEROL gt 6.2mmol/l
• EXCLUDED PATIENTS WITH ARYHTHMIAS AND HEART
  FAILURE
• ALL PATIENTS GIVEN 8 WEEKS OF DIETARY THERAPY
• IF CHOLESTEROL STILL RAISED (gt5.5) RANDOMISED TO
  RECEIVE SIMVASTATIN (20mg gt 40mg) OR PLACEBO
• OUTCOME DEATH OR MYOCARDIAL INFARCTION (LENGTH OF
  TREATMENT 5.4 YEARS ) WERE THE OUTCOMES

25
RCT EXAMPLE - 4S STUDY

• OUTCOME (death)
• Yes No
• Comparison intervention (placebo) 256
  1967 2223
• Experimental intervention (simvastatin) 182
  2039 2221
• The ARR is (256/2223) - (182/2221) 0.115 -
  0.082 0.033.
• The RRR is 0.033/0.115 0.29 or expressed as a
  percentage 29.
• 1/ARR NUMBER NEEDED TO TREAT.
• 1/0.033 30.
• i.e. if we treat 30 patients with IHD with
  simvastatin as per 4S study, in 5.4 years we will
  have prevented 1 death.

26
NNT EXAMPLES

• Intervention Outcome NNT

27
Why are RCTs the gold standardBreast cancer
mortality in studies of screening with
mammography women aged 50 and over (55 in Malmo
study, 45 in UK)

28
SCREENING - WILSON JUNGEN (WHO, 1968)

• IS THE DISORDER COMMON / IMPORTANT
• ARE THERE TREATMENTS FOR THE DISORDER
• IS THERE A KNOWN NATURAL HISTORY WINDOW OF
  OPPORTUNITY WHERE SCREENING CAN DETECT DISEASE
  EARLY WITH IMPROVED CHANCE OF CURE
• IS THE TEST ACCEPTABLE TO PATIENTS
• SENSITIVE AND SPECIFIC
• GENERALISABLE
• CHEAP / COST EFFECTIVE
• APPLY TO GROUP AT HIGH RISK

29
SCREENING

• DISEASE
• PRESENT ABSENT
• TEST POSITIVE A B
• NEGATIVE C D
• Sensitivity a/ac Specificity d/bd
• positive predicitive value a/ab
• negative predicitve value d/cd.

30
Value of exercise ECG in coronary artery stenosis

• DISEASE
• PRESENT ABSENT
• TEST POSITIVE 137 11
• NEGATIVE 90 112
• Sensitivity a/ac 60 Specificity d/bd
  91
• positive predicitive value a/ab 93
• negative predicitve value d/cd 55.

31
Sensitivities and Specificities for different
testsAlcohol dependency or abuse(as defined by
extensive investigations in medical and
orthopaedic in patients)

• SENS SPEC
• GGT 54 76
• MCV 63 64
• LFTs 37 81
• Yes to 1 or gt of CAGE ?s 85 81
• Yes to 3 or gt of CAGE ?s 51 100

32
MAKING SENSE OF THE EVIDENCE - ARE THESE RESULTS
VALID -i.e. should I believe them?

• Randomised (where appropriate)?
• Drop outs and withdrawals?
• Followup complete?
• Analysed in the groups to which randomised?-
• Intention to treat.

33
MAKING SENSE OF THE EVIDENCE- ARE THESE RESULTS
USEFUL?-i.e. should I be impressed by them, are
they relevant to my patients (GENERALISABLE)

• How large was the treatment effect?
• How precise was the estimate of treatment effect
• Were all important clinical outcomes considered?
• Do benefits outweigh risks?

34
READING CRITICALLY

• GENERAL
• The aims of the study are not stated
• The study is not original in concept
• The study is not particularly useful or relevant
  to general practice
• There could be ethical objections to the design
  or reporting of the study

35
READING CRITICALLY

• METHOD
• The design of the study is not consistent with
  the aims
• The sample is not representative of the whole
  population in question
• Controls are needed and not used
• Controls used are not appropriate
• Method(s) used for selecting cases / controls not
  clearly described
• Other method details (e.g. numbers, time periods,
  statistical methods used) are not clear and
  consistent
• Questionaires and proformas are not thoroughly
  tested or are not relevant.

36
READING CRITICALLY

• RESULTS
• There is missing data. e.g. drop-out rates,
  non-responders
• Other details e.g. numbers, percentages, p values
  are inaccurate / unclear
• Statistical methods would be useful but are not
  used
• Statistical testing is used but is inappropriate
• The tests of significance used do not meet the
  conditions for the application of these tests
• The sample size is so small that potentially
  clinically significant findings do not achieve
  statistical significance
• The sample size is so large that statistically
  significant findings have little clinical
  significance

37
READING CRITICALLY

• DISCUSSION
• The study is not discussed critically
• The results are not discussed in relation to
  other important literature in the field
• The discussions and conclusions speculate too far
  beyond what has been shown in this study

38
THE COMMONEST ERRORS ARE-

• ERRORS IN SAMPLE GROUPS OR QUESTIONAIRE DESIGN
• FAILURE TO DESCRIBE THE METHOD CLEARLY
• PROBLEMS WITH ALTERNATIVE RISK FACTORS,
  EXCLUSIONS AND WITHDRAWLS
• END POINTS AND DIAGNOSTIC DEFINITIONS UNCLEAR
• POPULATION IS NOT TYPICAL OF MINE

39
GUIDELINES - SELECTING DISEASE OR CONDITION

• HIGH MORBIDITY OR MORTALITY
• VARIATION FROM LOCAL OR NATIONAL PATTERN
• MAJOR SERVICE USER e.g.. stroke, arthritis,
  mental health
• CURRENT SERVICES OF QUESTIONABLE EFFECTIVENESS OR
  EFFICIENCY
• PRIORITIES - LOCAL OR NATIONAL NOISE - CLINICAL
  DEVELOPMENT OR RECOGNITION THAT IMPROVEMENT
  POSSIBLE
• BE REALISTIC - e.g.. review DCs, not the whole
  of gynae services.

40
The evidence isnt there

• Bandolier
• Evidence Based Medicine
• DTB, MeReC
• CRD
• Effective Healthcare Bulletins
• Effectiveness Matters
• Database of Abstracts of Reviews of Effectiveness
  
• Cochrane libraray

41
GPs will never find the time to track the
evidence down

• A lot has already been tracked down, critically
  appraised and packaged
• A practice that collectively can find an hour a
  week can make huge strides
• GPs dont have to DO the work personally! -
  others (in the PHCT, at universities and at
  health authorities) can make a valid contribution

42
GPs dont have the skills and experience to
critically appraise the evidence and determine
its applicability to their locality.

• A great deal of evidence has already been
  appraised
• Its not difficult!
• Critical appraisal skills have been shown to work
  (in randomised controlled trials!)
• Who else is in a position to determine
  applicability other than the local PHCT!

43
The relevant evidence cannot be recalled during
the consultation when the answers are required

• COMPUTERS!
• Most consultations do not require individual
  literature searching for those that do a further
  appointment is no real hardship
• Major areas of health gain can be handled with
  electronic reminders

44
EBM IN THE FUTURE

45
LIMITATIONS

• EBM WHAT IS BEST FOR AN INDIVIDUAL PATIENT
  (patient utility)
• EVIDENCE BASED PURCHASING BEST USE OF HEALTH
  CARE RESOURCES FOR THE LOCAL POPULATION (cost
  utility). i.e. knowledge of local needs,
  priorities and constraints
• WHAT IF THESE CONFLICT?