Prevention by liaoqinmei

VIEWS: 4 PAGES: 87

									An Ounce of Prevention
     Duke Family Medicine
 Common Problems in Primary Care
Opportunity knocks
   You are starting a new job in your hometown and the senior
    partners of the group ask you to set up a “prevention protocol”
    for the practice.
   A 16-year old student from the high school track team with a
    stated purpose of “sports physical” is your next patient.
   A couple presents for a pre-marital “blood test” with state form
    in hand.
   You have been asked to be on the committee for a local
    community health fair.
   You are seeing an 11-month-old for his third ear infection.
   Your first patient of the afternoon has a family history of breast
    cancer and she is starting to notice hot flashes and missing
    periods.
   The local paper has an article about the death of a local teenager
    in a corn silo.
   There is a new shirt factory in town and several of your patients
    are joining the work force there.
What we will cover
   Principles of prevention
   Primary prevention
   Secondary
       Screening
            Benefits & Risks of Screening
            Sensitivity, Specificity of a test
            Positive Predictive Value of a test
            How prevalence effects Prediction
   Shared decision making
   Counseling
Purpose

   Reduce incidence of disease
       Example: Immunization
 Reduce mortality and morbidity of
  disease by early detection
 Modifying personal behavior
       Knowledge as power
         Causes of Illness

50.00%
40.00%
30.00%
20.00%
10.00%
 0.00%


           Access


                    Genetics


                               Environment


                                             Behavior
Start at the beginning

   Risk profile
     Family history
     Age

     Gender

     Ethnic origin

     Exposures

     Lifestyle

 Targeted prevention
 Population profile
Opportunity Knocks
  Are you answering the door?
Family Hx

   What 6 target        Heart disease
    adult diseases       Diabetes
    should be            Stroke
    highlighted in
    family history?      Breast cancer
                         Ovarian cancer
                         Colon cancer
Strategies
   Primary
       Prevention of disease occurrence
   Secondary
     Screening asymptomatic population
      for risk
     Preclinical

   Tertiary
     Treatment of symptomatic population
     Minimize complications
Prevention: Components

   Screening
       Test or standardized exam
 Immunizations
 Chemoprevention
       Drugs or biologicals to prevent disease
   Counseling
Does it work?

   Stroke mortality down 50% since
    1972
       Early detection and treatment HTN
 Cervical cancer mortality reduced
  by 80%
 Neonatal screening
       Reduction in cognitive dysfunction
          PKU
          Congenital hypothyroidism
Primary Prevention

   Immunizations
Chemoprophylaxis

   ASA for adults at high risk for CAD
     A Recommendation
     Benefit > risk

 Tamoxifen    or raloxifene for breast
    cancer
     No in average or low risk
     Counseling and shared decision

      making in high risk
Chemoprophylaxis

   Post-exposure prophylaxis
     H. influenza, B
     Hepatitis A

     Hepatitis B

     Meningococcus

     Rabies

     Tetanus
Chemoprophylaxis

 Hormone   replacement
   Patient preference
   Risk > benefit

   Individual risk assessment

   I recommendation

 Travel
   Malaria prophylaxis
   Yellow fever
Screening

   Secondary prevention
Criteria for effective screening

   The burden of illness is high
   Detection is possible before symptoms
    are evident
   Treatment is available, acceptable &
    reduces morbidity and mortality if
    applied early
   Benefits of treatment outweigh harm
    including discomfort & anxiety
       First, do no harm
Recipe for success

 Test is accessible
 Test is acceptable to patient
 Patient willing to participate in
  follow-up and treatment
 Cost effective
Natural History of Disease


                         
            1  2  3  4
1. Biological onset
2. Early diagnosis possible
3. Usual clinical diagnosis
4. Outcome
Natural history of disease

Critical point

Biological             Symptomatic             Outcome
onset                  onset
         Screening          Case Finding

        Critical point: therapy effective prior to this
        point in natural history of disease,
        ineffective afterwards
Critical point and screening

Identify the disease/s illustrates a
   critical point suitable to screening:
1. Colorectal cancer
2. Cervical cancer
3. Pancreatic cancer
Critical point

   Cervical cancer
       Critical point occurs prior to onset of
        symptoms
       Effective intervention available in
        asymptomatic phase
       Intervention improves outcome
   Colorectal cancer
   Pancreatic cancer
       Aggressive
       Short history
Screening bias

 Lead-time bias
 Length bias
Screening advances time of diagnosis, not survival time
      50% detected by screening while
      remainder progress to clinical
      symptoms during screening interval
False Security




                              Longer asymptomatic period, more
                              likely to be detected in screening
                              interval
Screening guidelines

   USPSTF
     Review evidence of effectiveness for
      screening in asymptomatic, average
      risk population
     Make recommendations for clinical

      preventive services
     Identify research agenda for clinical

      prevention
Organizations

   USPSTF
       Evidence based
   Specialty organizations
     AAFP
     ACP

     ACPM

     Etc.
USPSTF Ratings: Evidence

A.   Strongly recommends that clinicians provide
     [the service] to eligible patients. (Good evidence)
B.   Provide [this service] to eligible patients. (Fair
     evidence)
C.   No recommendation for or against routine
     provision of [the service]. (Fair evidence, Close
     balance)
D.   Recommends against routinely providing [the
     service] to asymptomatic patients. (Ineffective or
     risk > benefit)
I.   Evidence is insufficient to recommend for or
     against routinely providing [the service].
     (Evidence lacking or poor quality or conflicting,
     cannot determine risk: benefit ratio)
Quality of Evidence
   Good: Evidence includes consistent results from well-
    designed, well-conducted studies in representative
    populations that directly assess effects on health
    outcomes.
   Fair: Evidence is sufficient to determine effects on health
    outcomes, but the strength of the evidence is limited by
    the number, quality, or consistency of the individual
    studies, generalizability to routine practice, or indirect
    nature of the evidence on health outcomes.
   Poor: Evidence is insufficient to assess the effects on
    health outcomes because of limited number or power of
    studies, important flaws in their design or conduct, gaps
    in the chain of evidence, or lack of information on
    important health outcomes.
Example: Diabetes

   USPSTF                         ADA
       Screening does detect          Start age 45
        in pre-symptomatic
        phase                          FPG
            No early                  In a health care
             intervention
             evidence (I)               setting
       Screening approved             Every 3 years
        for hypertension and           Earlier and more
        hyperlipidemia
        patients (B)                    frequent if BMI >
            Modifies                   25 with additional
             management                 risk factors
       The Critical Point             E recommendation
How to judge a Screening Test

   Sensitivity
       In asymptomatic phase
   Specificity
       Avoid false positives
   Positive Predictive Value (PPV)
       When yes means yes
   Negative Predictive Value (NPV)
Indices of accuracy
   Sensitivity:
       Proportion with disease who correctly test +
       Low proportion of false-negatives
   Specificity:
       Proportion without disease who correctly
        test –
       Low proportion of false-positives
   Gold standard
   Sensitivity and Specificity describe test
    accuracy and are
    independent of disease prevalence
Accuracy

   Sensitivity                   Condition   Condition
                                   Present     Absent
       a/a+c
   Specificity
       d/b+d     Positive Test       a           b
   PPV                           (True +)    (False +)
       a/a+b
   NPV            Negative          c           d
                     test         (False -)   (True -)
       d/c+d
Risks

   List potential risks of a poorly
    sensitive test
       False negatives
   List the potential risks of a poorly
    specific test
       False positives
    The potential for harm is real

                            Asymptomatic Adult
                             They are not sick


                            Screening Procedure


             Test normal                           Test abnormal


True negative        False negative      False positive      True Positive


  Reassure             Delay Dx       Significant anxiety    Treat Cancer
                     Possible harm      Possible harm         Prolong life
Accuracy

   Sensitivity                   Condition   Condition
                                   Present     Absent
       a/a+c
   Specificity
       d/b+d     Positive Test       a           b
   PPV                           (True +)    (False +)
       a/a+b
   NPV            Negative          c           d
                     test         (False -)   (True -)
       d/c+d
Prevalence




  The total number of cases of a disease in a given
            population at a specific time.
Predictive Value

 Given a positive results, what is the
  probability that disease is present?
 Positive Predictive Value (PPV)
       Proportion who test + & have disease
   Dependent on disease prevalence in the
    population studied.
Prevalence and PPV




   90% Sensitivity, 95% Specificity
                      Size of population = 100,000
      Sensitivity of test = 90%          Specificity of test = 90%

                Cancer Prevalence = 1%           Cancer Prevalence=0.1%
Prevalence
                 Cancer    Cancer                    Cancer    Cancer
                 Present   Absent                    Present   Absent
Positive test    900       9,900     Positive test   90        9,900
Negative test    100       89,100    Negative test   10        89,910
                    PPV = 8.3%                            PPV = 0.9%

                 12 worried for                      111 worried for
                 every case                          every case
Reliability

 Reproducible results
 Valid
 Reliable
A Clinical Question

   A 30 y.o. woman presents to your
    office for a check-up. She has no
    history of illness.
     What preventive measures would you
      recommend on the basis of age and sex
      alone?
     What data do you need to decide

      about additional screening?
Routine recommendations

   Alcohol Misuse                   Alcohol: B
   ASA for prevention of            ASA: high risk
    CVD
   Blood pressure screening         BP screening: A, 18 and
   BCRA mutation                     older
    screening                        BCRA: High risk
       Genetic risk assessment
        for ovarian and breast
        cancer susceptability
   Breast cancer prevention         Breast Ca prevention:
    medication                        High risk
   Cervical cancer screening        Cervical cancer screening:
                                      A, sexually active with a
                                      cervix
30 y/o female

   Chlamydia screening      Chlamydia: A, through
                              age 25 and high risk
   Dental and               Dental: B, dental care,
    periodontal disease       flossing, fluoride, oral
                              cavity exam
   Depression
                             Depression: B, screening
   Diabetes                  in appropriate setting
   Diet                     Diabetes: B, high risk;
                              HTN, Hyperlipidemia
   G.C.                     Diet: Hyperlipidemia or
   HIV                       other risk factors for
                              chronic disease, B
                             G.C.: high risk, age < 25
                             HIV: Risk based: A
30 y/o old female

   Lipids           Lipids: high risk: men age
   Obesity           35, women 45: treat if
   Syphilis          high and at increase risk
                      for CHD; younger if risk
   Tobacco use       high (20-30 men, 20-45
                      women)
                     Obesity: B, offer intensive
                      counseling and
                      behavioral interventions.
                     Syphilis: A for high risk
                     Tobacco: A, screen and
                      offer intervention
                     TB screening: high risk
Level D
But, doc!

   “Shouldn’t I have a mammogram?”
     Second leading cause of cancer death
      in US women
     1/8.2 women in US diagnosed over

      lifetime
     1/30 will die because of the disease

     > 50% without identified risk factors
Cases of Breast Cancer in 2000

 45000

 40000

 35000

 30000

 25000
 20000

 15000

 10000

  5000

     0
         <30   30-39   40-49   50-59   60-69   70-79   80 +

ACS Surveillance Research, 2000
Breast Cancer

   Prevalence Varies by Age
     20 cases/100,000 women @ age 30
     180 cases/100,000 women @ age 50

   Risk Factors
       Age, Genetic, Hormones, Radiation,
        Prior Cancer, High Fat Diet, Alcohol
Cumulative Risk of Breast CA
Age             Chance
By age 30       1 out of 2,525
By age 40       1 out of 217
By age 50       1 out of 50
By age 60       1 out of 24
By age 70       1 out of 14
By age 80       1 out of 10
Lifetime        1 out of 8
ACS data 2000
    Mammogram @ age 30
    Sensitivity                 No
    70%                                                Prevalence: 20/100000
    PPV: .003     Disease     Disease      Total
       Test +       14         4,999       5,013       Sensitivity: 70%


       Test -        6        94,981      94,987       Specificity: 95%


                                                       PPV: 0.003
       Total        20        99,980      100,000

0




    For every case, 358 undergo needless evaluation, cost and anxiety
    [5013/14=358]
   Mammogram @ age 60
Sensitivity:                  No
70%                                                      Prevalence:
PPV: 0.026     Disease      Disease        Total         180/100000

   Test +        128         4,991        5,119
                                                         Same sensitivity and
                                                         specificity

   Test -        52         94,829        94,881
                                                         PPV: 0.026


    Total        180        99,820       100,000



For every case, 39 get worked up needlessly [5119/128]
What about outcomes?
Outcome: Risk reduction

   Relative risk and RR reduction
       Probability of event in active
        group/probability in control group
   RR reduction
       Absolute risk/ probability in control
        group
Screening effectiveness

 Gain in life expectancy
 Cost per case detected
 Cost per life saved
 Gain in quality-adjusted years
 Number needed to screen
       How many need to be screen impact
        one person
Mammography

   Does screening reduce disease?
For Breast Cancer

   The burden of illness is high
       Most common CA in women with 182,000
        new cases & 44,000 deaths in 2000
   Detection is possible before symptoms
       Self Exam, Physician Exam, Mammogram
   Treatment is available
       Surgery, Chemotherapy, Radiation
   Treatment value exceeds the discomfort &
    anxiety
Mammography Benefit
   Swedish study
       4 randomized studies
       247,010 women
       15.8 years follow-up
       21% reduction in mortality
            RR 0.79 in screened group
            Statistically significant in 55-69 y/o
       Benefit evident at 4 years out
       Increased to 10 years
   20-30% reduction in mortality realized
    after a delay of 6-10 years
USPSTF
   2002 USPSTF recommendations
       No evidence of value for ages <40
            Strongest evidence age 50-69
       Every 1-2 years beginning at age 40
          Most studies indicate mortality benefit
          Absolute benefit smaller in 40’s due to
           incidence in the age group vs. >50
       No proof of value > 70 but likely is a
        good idea
            In absence of co-morbidities that reduce
             life expectancy
What if she were a he?

   Alcohol Misuse           Diabetes
   ASA for prevention       Diet
    of CVD                   G.C.
   Blood pressure           HIV
    screening                Lipids
   Chlamydia screening      Obesity
   Dental and               Syphilis
    periodontal disease
                             Tobacco use
   Depression
He?
When you get older…
Older


        The younger woman was D.
Older
Older
Colorectal cancer

   Long latency: 10 years
   Large Adenomatous Polyps
   Risk assessment
   Fecal Occult Blood Test for Colon Cancer
       Sensitivity 26-92%
       Specificity 90-99%
       PPV 2-11%
   Sigmoidoscopy
       50% reduction in overall CRC mortality
   Colonoscopy
       Awaiting data on outcome comparison
Shared decision making

   Serum PSA for Prostate Cancer
       Sensitivity and Specificity
            Varies with cutoff value, age and
             comorbidity
     PPV 20%
     2nd leading cause of CA death in men

   Does screening improve outcome?
Food for thought
Shared decisions
Shared decisions

   Would you feel better knowing?
       Not knowing?
 What happens if the PSA is
  elevated?
 What happens if you have cancer?
 What difference will it make for you
  to know?
Shared decisions
Shared decisions: pros and cons
Guidelines

   USPSTF
       Recommendation level: I
   Major medical organizations
     Discuss potential benefits and possible
      harms
     Consider patient preferences.

     Individualize the decision to screen.

     No endorsement of universal
      screening
Selective screening

   Most likely to benefit
     Age over 50
     Younger men at high risk (45)
            African American men
                 Younger age of onset
                 More aggressive disease
            First-degree relative
   Unlikely to benefit
       Men with less than 10 year life
        expectancy
Knowledge as power
   Family history
       CHD
            Risk increase x 2-6
       Cancer
            Colon: 2-5 x increase
            Breast: 2-6 x increase
            Prostate
       Nature vs. nurture
   Targeted screening ?
   Targeted intervention
   BEHAVIOR CHANGE?
Motivation

 Identify stage of change
 Tailor made messages
     Patient preference
     Consider

         Knowledge and literacy
         Beliefs

         Culture

         Support

   Reflective listening
Resources

   www.myhq.com
       ID: pickens
       Password: marshall
   USPSTF
       “The Bible” of prevention
            Asymptomatic population
            Office based
            Primary and secondary prevention
            Evidence based
   http://fmclerkship.mc.duke.edu
Primum non nocere

   FINIS
Five R’s

 Relevance
 Risks
 Rewards
 Roadblocks
 Repetition
Behavioral counseling

 Assess
 Advice
 Agree
 Assist
 Arrange
Encourage self-management

 Realistic goals
 Education
 Support
 Write it down
 Gold star treatment

								
To top