The Practice of Preventative Medicine

1
The Practice of Preventative Medicine

• What tests should we check on our patients and
  when should we check them?

2
Outline

• What is preventative medicine and what is its
  importance
• The definition of screening and how we decide
  what a good screening test is
• Who do you look to for the current
  recommendations, and how do they formulate those
  recommendations
• The tests/procedures and current recommendations

3
What is preventative medicine?

• Disease prevention
• Identification of disease at an early stage
• Definition procedures/exams/interventions
  performed on patients without specific
  complaints, to identify and modify risk factors
  to avoid the onset of disease, or to find disease
  early in its course so that by intervening
  patients can remain well.
• Synonyms health maintenance or the periodic
  health examination

4
What is the importance of reviewing this subject.

• 1. It has been shown to be effective at many
  levels (ex. Community wide immunizations,
  colorectal cancer screening, cervical cancer
  screening, breast cancer screening, etc..)
• 2. Your patients are going to ask you questions
  about why you are checking certain tests. (ex.
  Pap smear, mammograms)
• 3. Cost considerations (if a test is not shown to
  be beneficial, then by checking it routinely we
  are wasting resources)

5
(No Transcript)
6
3 levels of preventative medicine

• 1. Primary prevention- prevents disease from
  occuring at all by removing its causes (ex. Use
  of folic acid to prevent neural tube defects,
  immunizations, counseling to adopt healthy
  lifestyles, mandating seat belt use, etc..)
• 2. Secondary prevention- detects disease early
  when it is asymptomatic and when early treatment
  can stop it from progressing (ex. Pap smears,
  mammograms, FOBT, colonoscopy, etc..)
• 3. Tertiary prevention-clinical activities that
  prevent further deterioration or reduce
  complications after a disease has declared itself
  ( ex. Starting BB after MI to decrease risk of
  death, initiation of ACE-I in diabetics to
  prevent nephropathy, etc.)

7
(No Transcript)
8
Screening

• The identification of unrecognized disease using
  various modalities (H/P, lab tests, Xray,
  procedures)
• Sort out well appearing persons with a disease or
  risk factors for disease
• Part of many primary and all secondary prevention
  measures
• Screening tests are typically not diagnostic so
  the clinician must be willing to further
  investigate a test

9
How do we decide what to screen for?

• 1. Must consider the burden of suffering caused
  by the condition (discomfort, disability, death)
• 2. Must consider how good the screening test is
  in terms of sensitivity, specificity, cost, ease
  of use, safety, acceptability, risk of a false
  positive, labeling of pts with conditions.
• 3. Then must consider in the case of primary
  prevention, how good is the intervention in
  preventing the disease, or in secondary
  prevention , how good are the treatments that are
  available for the disease (in terms of
  efficacy/compliance/early treatment as a benefit
  versus late treatment)

10
What does that mean?

• Example Breast cancer screening
• Primarily occurs in women gt50yrs
• In women in their 20s the incidence is very low
  approx 1 in 100,000
• For the women that get breast CA at that age
  there is a substantial morbidity and mortality,
  but given its rarity in that age group screening
  becomes impractical and would probably lead to
  more morbidity and usage of resources from
  further w/u of false positives (biopsies, further
  radiological tests, etc..)

11
The dreaded Statistics

• A good screening test should have a high
  sensitivity to limit the false negatives and a
  high specificity to limit the false positives
  (these are affected by the prevalence of the
  disease in the population).
• For many screening tests the gold standard from
  which sens/spec are determined is based on long
  term follow up with monitoring for occurrence of
  a disease following the screening test -detection
  method. The two problems with this are knowing
  the length of time needed to follow the patient
  to detect the disease, and the assumption that
  the abnormality detected would go on to cause
  disease if left alone (ex. Estimated that
  virtually all men gt90yr have foci of prostate
  CA). A second method to get around this is the
  incidence method which calculates sensitivity
  based on ratio of pts with a disease undergoing
  the screening test over that of the patients with
  the disease who dont have the test (1- pts with
  disease undergoing screening/ pts with disease
  who didnt have screening). This limits counting
  benign conditions found on screening but might
  decrease sensitivity by ignoring pts with
  malignancies with long lead times.

12
Consider Simplicity and Cost

• A good screening test should be relatively simple
  to perform (ex. BP) although that is not always
  the case (colonoscopy)
• Cost of not only the test itself, but the cost of
  subsequent workups of positive results and the
  time missed from work for special visits to the
  physician must be considered

13
Acceptability, Labeling

• Acceptability- Do the pt and physician find the
  test acceptable to perform (ex. Many asymptomatic
  pts refuse a colonoscopy, many women refuse
  pelvic exams)
• Labeling- Psychologically a false positive result
  can be devastating for a pt. One study showed
  that almost 50 of women with false mammograms
  read as high suspicion suffered anxiety and
  worries that affected their daily lives (1).

14
Risk of False Positives

• One study showed that on average internists
  selected 54 different tests during periodic
  health examinations (components of CBC, Chem 14
  etc.) (2).

15
Biases

• Lead time bias- Period of time between detection
  of disease by screening versus when it would
  present with symptoms. If the period is very
  short then the screening test will not be very
  useful (ex. Lung cancer). If the lead time is
  long then the test might be useful (ex. Cervical
  CA).
• Length time bias- more slow growing cancers are
  diagnosed during screening than during usual
  medical care when pts are more likely diagnosed
  with a rapid growing tumor causing symptoms, so
  screening may be catching more malignancies with
  a better prognosis and therefore might appear
  more beneficial than it truly is.

16
(No Transcript)
17
Where do you look for the current recommendations?

• The U.S. Preventative Services Task Force
  (USPSTF) was first convened by the U.S. Public
  Health Service in 1984 and its recommendations
  are considered the gold standard for clinical
  preventative services. They review the current
  available literature/information and make
  recommendations based on the available data.
• They have released two print editions and
  portions of the 3rd edition are now available
  online.

18
The tests that I will review

• H/P and Labs
• HP
• CBC
• CHEM 8
• LFTs
• Fasting Lipids
• TSH
• PSA
• UA
• PAP Smear
• HIV Serology
• FOBT

• Radiological/Procedural
• Colonoscopy
• Mammogram
• EKG/Stress testing/Calcium scores
• CXR
• DEXA
• Ankle/Brachial Index (PAD)
• Carotid Doppler

19
How do they rate these tests?

• First they rate the evidence
• Prior to the 3rd edition
• Grade I-at least one properly randomized
  controlled trial
• Grade II-obtained from other controlled trials or
  analytic studies, or from multiple time series
  with dramatic results in uncontrolled
  experiments, has 3 subcategories based on the
  type of study
• Grade III-Opinions of respected authorities based
  upon clinical experience, descriptive studies and
  case reports, or reports of expert committees.
• With the 3rd edition they tried to simplify it
• Good- consistent results obtained from well
  designed, well conducted studies in the
  representative population that directly assess
  effects on health outcomes
• Fair- evidence sufficient to determine effects on
  health outcomes but the strength of evidence is
  limited by the number, quality, or consistency of
  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
  design or concept of studies, gaps in the chain
  of evidence, or lack of information on important
  health outcomes.

20
Next they give a recommendation and grade the
recommendation based on that evidence

• Prior to 3rd edition there were 5 grades
• A-There is good evidence to support the
  recommendation for inclusion of the service for
  the periodic health evaluation.
• B-There is fair evidence to support the
  recommendation for inclusion of the service for
  the periodic health evaluation.
• C-There is insufficient evidence for or against
  the recommendation for inclusion of the service
  for the periodic health evaluation.
• D-There is fair evidence to support the
  recommendation to exclude the service for the
  periodic health evaluation.
• E-There is good evidence to support the
  recommendation to exclude the service for the
  periodic health evaluation.
• For the 3rd edition USPSTF changed grading to
  include recommendations
• A-USPSTF strongly recommends the clinician
  provide the service to eligible pts, because
  there is good evidence it was found to improve
  important health outcomes and concludes that
  benfits outweigh the harms
• B-USPSTF recommends the clinician provide the
  service to eligible pts, because there is fair
  evidence it was found to improve important health
  outcomes and concludes that benfits outweigh the
  harms
• C-USPSTF makes no recommendation for or against
  the service, because there is fair evidence it
  can improve health outcomes but balance of
  benfits and harms is too close for a general
  recommendation
• D-USPSTF recommends against the service for
  asymptomatic pts, because there is fair evidence
  it is ineffective or that harms outweigh benefits
• I-The evidence is insufficient to recommend for
  or against the service

21
History and Physical

• No general recommendation for or against general
  H/P and the only thing it costs is time
• Obviously a thorough H/P on initial consultation
  and then periodically is the best screening tool
  that we have since the majority of our diagnoses
  are made through H/P
• Certain aspects that do have recommendations

22
Blood Pressure

• For pts age 18 Grade A with good evidence for
  regular screening for HTN (at least every one to
  two years)

23
Clinical Breast Exam

• Grade-I no study comparing CBE to no exam, and
  studies show that breast cancer mortality is
  similar in groups screened with mammography /-
  CBE, most organizations including the AMA
  recommend to begin yearly exams starting at age
  40 (if not earlier based on family history)

24
Digital Rectal Exam

• Grade I Although DRE has some utility as a
  screening tool for prostate CA (although
  sensitivity is still quite low) it is
  inconclusive whether early detection improves
  outcomes and screening is associated with
  important harms (frequent false , anxiety,
  biopsies etc.). None of the major medical
  societies endorses universal or mass screening of
  any particular group, but they conclude that men
  most likely to benefit are age 50 (or younger if
  they have a strong family history) and have a
  life expectancy of gt10yrs.

25
CBC

• No general recommendation on a complete blood
  count
• For H/H- Grade C (2nd edition) with level I/II
  evidence which is inconclusive, so no rec. for or
  against general use as screening tool. For
  Pregnant women Grade A with level I/II evidence
  for screening

26
Chemistry Panel

• No general recommendation on total chem panel
• Glucose-
• In general population-Grade I, not enough
  evidence available to recommend for or against
  screening
• In patients with HTN or hyperlipidemia- Grade B
  with good evidence for annual screening since in
  pts with HTN and hyperlipidemia there is a higher
  level of DM II and that those pts benefit from
  more aggressive treatment of their
  HTN/hyperlipidemia

27
Liver Function Tests

• No general recommendations to use as a screening
  tool
• There are recommendation for Hep viral screening
• Hep B surface antigen
• In pregnant women at first prenatal visit- Grade
  A with good evidence for screening all pregnant
  women
• In general asymptomatic pts- Grade D with no
  evidence of benefit and possible harms (cost,
  labeling, etc.)
• In high risk pts- Grade C, no recommendation for
  or against, might be beneficial to identify pts
  that might benefit from vaccination
• Hep C serology
• In general population- Grade D, with no evidence
  of benefit and possible harms
• In high risk pts- Grade I, In high risk pts
  screening would have higher yield and tx with
  current therapies appear beneficial but no data
  as to whether early tx improves mortality
  (although older therapies are suggestive that it
  does) so they were unable to give a general
  recommendation for or against, but other groups
  including the NIH/CDC do recommend screening high
  risk groups

28
Lipids

• For Men age 35 and Women age 45- Grade A with
  good evidence for screening every 5 years
• For men/women less than those ages with other
  risk factors for CAD- Grade B with good evidence
  for screening every 5 years
• For men/women less than those ages without risk
  factors for CAD- Grade C with no good evidence
  for or against screening although potential
  benefit in this pt population is likely less

29
(No Transcript)
30
TSH

• In Asymptomatic pts- Grade I, no recommendation
  for or against regular screening since there is
  no substantial data showing benefit to treating
  subclinical screening detected thyroid disorder

31
PSA

• Pts of any age- Grade I, no recommendation for or
  against use as a regular screening test for
  similar reasons to DRE. Must discuss
  risks/benefits of the test with pt and decide
  together whether to proceed.

32
UA

• For Asymptomatic men/women (not pregnant)- Grade
  D with fair evidence that there is no benefit to
  routine screening
• Pregnant women- Grade A, recommends routine
  UA/Culture at 12-15 wks to r/o asymptomatic
  bacteruria
• As a screen for bladder CA- Grade D with fair
  evidence that screening is not beneficial due to
  the low prevelance of asymptomatic disease that
  progresses to clinically significant disease

33
PAP Smear/Pelvic

• Women who are sexually active and have a cervix-
  Grade A with good evidence for beginning
  screening pap smears at least every 3 yrs within
  3yrs of onset of sexual activity or by age 21.
  There is no direct evidence that yearly screening
  is more effective, but several organizations have
  made recs. regarding yearly screening till 2 to 3
  normals followed by screening every 3 yrs, or
  yearly testing in pts with higher risk sexual
  activity, although there is no data to support
  either of these recs..
• Women 65 with adequate normal pap smears
  recently and at not at high risk- Grade D with
  limited evidence for benefit with cont screening
  in this population
• Women who had hysterectomy for benign disease-
  Grade D with fair evidence that there is little
  benefit to cont screening in the population
• HPV testing- Grade I with poor evidence to
  determine the risks/benefits of HPV testing in
  conjunction with pap smear as a screening tool.
  Trials are ongoing looking at question, but
  currently no recommendation for or against its
  use.

34
HIV Serology

• History is very important on deciding this issue
• In non high risk groups- Grade C recommendation
  with fair evidence , no recommendation for or
  against screening
• In high risk groups (IVDU, homosexual men)- Grade
  A with good evidence for periodic screening
• For Pregnant women- Grade A with good evidence
  for routine screening

35
FOBT/Colonoscopy

• Recommends screening for CRC in general with
  Grade A, good evidence that screening beginning
  at age 50 for average risk individuals is
  beneficial. They do not make direct
  recommendations on which screening modality to
  use although they suggest multiple possibilities
  including yearly FOBT flexible sigmoidoscopy
  every 5 yrs, or colonoscopy every 10 yrs
  (considered the gold standard) that vary in their
  cost, complications, availability,
  sensitivity/specificity etc..
• For pts with a FH of colon CA before age 60 it is
  reasonable to begin screening at an earlier age

36
Mammogram

• For women age 40- Grade B recommendation with
  fair evidence of benefit for screening with
  mammogram every 1-2 yrs with or without CBE/SBE

37
EKG/Stress testing/Calcium Scores

• In pts at low risk for CHD events- Grade D
  recommendation with fair evidence that although
  testing will detect a small percentage of pts
  with disease, the harm from screening the
  population including unnecessary procedures/over
  treatment/labeling outweigh any benefits
• In pts at increased risk for CHD- Grade I with
  insufficient evidence to recommend for or against
  screening in this population. It is unclear in a
  asymptomatic pt in this population whether these
  tests add more information than risk
  stratification based on conventional CHD risk
  factors, and whether the benefits outweigh the
  harms. (the AHA suggests stress testing might be
  beneficial for diabetic pts if they are going to
  start a vigorous exercise program).

38
CXR

• To screen for Lung CA- Grade I with insufficient
  evidence to recommend for or against routine
  screening (there is no good study designed to
  look at screening versus not screening, but
  observational studies suggest that CXR screening
  does not decrease mortality from lung CA which
  makes since because by the time it is visible on
  a CXR it has usually metastasized on a cellular
  level and we have essentially no effective
  treatment options for metastatic disease).

39
DEXA (Dual-energy x-ray absorptiometry )

• Women age 65 or women between 60-64 who are at
  increased risk (low wt, postmenopausal not on
  estrogen replacement)- Grade B recommendation
  with good evidence that screening is beneficial
  in these groups of women. There is no specific
  recommendation on the screening interval, but it
  is thought that 2 yrs might be needed to see a
  substantial difference in BMD.
• Women age 60-64 who are not at increased risk or
  women younger than 60- Grade C with fair evidence
  that it would prevent a small number of fractures
  but the benefits and harms were similar so no
  recommendation for or against screening in this
  population
• Women/Men on long term corticosteroids- Not
  specifically reviewed. A summary statement by
  the Amer College of Rheum in 2001 recommended
  that for Pt on Prednisone 5mg daily for time of
  3months that they be screened annually to
  biannually (and tx preventatively)(3).
• Screening in Men- has not been reviewed to date

40
Ankle/Brachial Index

• In Asymptomatic Individuals- Grade D with fair
  evidence that pts in this category would not
  benefit from screening/treatment

41
Carotid Doppler

• In Asymptomatic Individuals- Grade I with
  insufficient evidence for screening with carotid
  doppler in this population

42
Summary What you should definitely do

• Men
• Annual HP starting at age 18 including BP
  measurement
• Fasting Lipids every 5 yrs starting at age 35 for
  average risk pts and younger if pt has
  significant risk factors
• CRC screening starting at age 50 with either
  annual FOBT with flex sig every 5 yrs or
  colonoscopy every 10 yrs. If Pt has FH of early
  CRC then it is acceptable to begin screening at a
  younger age (dictated by their history)
• If pt has HTN or Hyperlipidemia then check annual
  fasting glucose
• HIV Serology annually in high risk populations

• Women
• Annual HP starting at age 18 including BP
  measurement
• Fasting Lipids every 5 yrs starting at age 45
  for average risk pts and younger if pt has
  significant risk factors
• CRC screening starting at age 50 with either
  annual FOBT with flex sig every 5 yrs or
  colonoscopy every 10 yrs. If Pt has FH of early
  CRC then it is acceptable to begin screening at a
  younger age (dictated by their history)
• If pt has HTN or Hyperlipidemia then check annual
  fasting glucose
• HIV Serology annually in high risk populations
• Pap Smear every 1-3 yrs (no good evidence to
  support one way over another) starting within 3
  yrs of initiation of sexual activity or by age
  21.
• Mammogram annually starting at age 40
• DEXA biannually starting at age 65 for normal
  risk pts or age 60 for higher risk pts
• H/H, Glucose, Hep B sur Ag, UA/Culture, and HIV
  serology all recommended in pregnant women

43
Screening tests that the evidence is inconclusive
or not available

• Men
• DRE /- PSA annually starting at age 50 in men
  with a 10 yr life expectancy
• Annual fasting glucose in pts without risk
  factors
• Lipids in pt less than 35 yrs old without CAD
  risk factors
• TSH
• HIV screening in low risk individuals
• Hep B/C screening in high risk individuals
• CXR in long term smokers to screen for lung CA
• DEXA (no recommendation to date)
• EKG /- Stress testing for asymptomatic pts with
  multiple risk factors for CAD (including DM)

• Women
• Annual fasting glucose in pts without risk
  factors
• Lipids in pt less than 45 yrs old without CAD
  risk factors
• TSH
• HIV screening in low risk individuals
• Hep B/C screening in high risk individuals
• CXR in long term smokers to screen for lung CA
• DEXA in pts less than 60 yrs old
• EKG /- Stress testing for asymptomatic pts with
  multiple risk factors for CAD (including DM)
• HPV testing in addition to PAP Smear

44
Tests Not recommended for screening

• CBC/CHEM8/LFTs
• ABI
• PAP/Pelvic for pts who had hysterectomy secondary
  to benign disease, or for women 65yrs or older
  who have had negative pap smears on regular
  routine screening
• Carotid Dopplers
• Hep B/C serologies
• UA (for bladder CA or other etiologies)

45
Bibliography

• 1. Lerman, C, Trock, B, Rimer, BK, et al.
  Psychological and behavioral implications of
  abnormal mammograms. Ann Intern Med 1991
  114657.
• 2. Romm, FJ, Fletcher, SW, Hulka, BS. The
  periodic health examination Comparison of
  recommendations and internists' performance.
  South Med J 1981 74265.
• 3. Recommendations for the prevention and
  treatment of glucocorticoid-induced osteoporosis
  2001 update. American College of Rheumatology Ad
  Hoc Committee on Glucocorticoid-Induced
  Osteoporosis. Arthritis Rheum 2001
  Jul44(7)1496-503. 54 references
• 4. www.ahrq.gov/clinic/prevnew.htm.
• 5. www.uptodate.com, Preventive services
  recommendations for periodic health evaluation,
  F. Daniel Duffy, MD