Health Maintenance

Document Sample
 Health Maintenance Powered By Docstoc
					Health Maintenance

      Presented by
John Zweifler, M.D., M.P.H.
Who and What do we screen?
• Significance of condition.
  – Severity
  – Frequency
• Detectable during asymptomatic period.
• Effective intervention available.
Targeting Health Maintenance
• Deaths/year attributable to various conditions.
   – Cigarette smoking - 400,000
   – Diet and exercise - 300,000
   – Excess alcohol - 100,000
   – Breast cancer - 40,000
   – Cervical cancer - 4,000
   – Colo-rectal cancer - 56,000
   – Prostate - 30,000
   – Lung -155,000

       • *Ganiats T Prevenion Strategies in Family Practice. AAFP. 2003
      Assessing screening
• Quality of screening test.
  – Sensitivity, specificity
  – Accuracy
• Acceptability of screening.
  – Cost
  – Convenience
  – Availability
• Potential adverse effects of screening
  and treatment.
    Sensitivity and Specificity
• Condition Present   • Condition Absent

Positive Test     a   Positive Test       b
Negative Test     c   Negative Test      d
Sensitivity=a/(a+c)   Specificity=d/(b+d)
                      a=true positive
Positive Predictive   b=false positive
 Value=a/(a+b)        c=false negative
                      d=true negative
                     Testing Conditions
                      Size of Population = 100,000
                        Sensitivity of Test = 90%
                        Specificity of Test = 90%
• Cancer Prevalence = 1%             • Cancer Prevalence = 0.1%
         Cancer     Cancer                   Cancer      Cancer
         Present    Absent                   Present     Absent

Positive      900           9,900    Positive        90           9,990
 Test                                Test

Negative      100          89,100    Negative        10         89,910
 Test                                 Test

 Positive Predictive Value= 8.3%       Positive Predictive Value = 0.9%
       Cost Effective Analysis
• Considerations in cost effective analysis:
  – Perspective - Patient, payor, society
  – Cost of intervention.
  – Cost of necessary additional tests or monitoring.
  – Cost of complications.
  – Opportunity cost - allocation of resources.
    Cost effective analysis*
   Cost per year of life saved
• Mandating automatic seat-belts: $0-$25,000.
• Influenza vaccination: $500.
• Nicotine gum/smoking cessation: $6,000-
• Statin drugs for men 35-55 years with CHD and
  chol >250mg/dl: $0-$9,000.
• Statin drugs for women 35-45, no CHD,
  cholesterol >300: $1,000,000.

  – *Deyo R. JABFP JAN. - FEB. 2000. Vol. 13 #No. 1 47-54
           Cost Effectiveness of
            Various Screenings
• Annual screening for cervical cancer, women
  21 years or older - $50,000 per life year gained.
• Hypertension screening for asymptomatic men
  20 years and older - $48,000.
• Hypertension screening for asymptomatic
  women 20 years and older - $87,000.
        Types of Prevention
• Primary prevention: prevent or arrest the
  disease process in its earliest stages by
  promoting healthy lifestyles or immunizing
  against infectious disease.
• Secondary prevention: detecting and treating
  asymptomatic risk factors or early
  asymptomatic disease.
• Tertiary prevention: screening for
  complications of known disease.
 United States Preventive Service
      Task Force (USPSTF)
    Guide to Clinical Preventive Services
• Released the first report in 1989.
• Now supported by the Agency for Health Care
  Research and Quality, and the United States Public
  Health Service.
• Relies on evidence based approaches.
• Task force members represent health-care related
  federal organizations and primary care and
  preventive medicine specialties.
        Hierarchy of Research
• I. At least one properly randomized control trial.
• II-1. Well designed control trials without randomization.
• II-2. Well designed cohort or case-control analytic studies.
• II-3. Multiple timed series with or without the intervention
  or dramatic results in uncontrolled experiments.
• III. Opinions of respected authorities, descriptive studies
  and case reports, or reports of expert committees.

    – *USPSTF. 2001.
Pelvics and Rectals!?
TOTAL             TOTAL
37 %              63 %
Transverse 11%

                  Descending 7 %
Ascending 9%

                  Sigmoid 24 %
Cecum 11 %
                 Rectosigmoid 9 %
Appendix 6 %
                   Rectum 23 %
          Colorectal Screening
• Selby, NEJM, 1992 - Case control study
  showed 70% reduction in distal CRC in those
  exposed to sigmoidoscopy.
• Selby, Atkins & Sakamoto JFP, 1994 -
  Studies suggest sigmoidoscopic screening q
  10 years may be effective.
• Atkins, NEJM, 1992 - Adenomatous polyps
  <1 cm no benefit to colonoscopic follow up.
Colorectal Cancer and Polyps
•   ~30-50% of Americans 50-75 y.o. have polyps.
•   90% of polyps <one centimeter.
•   If polyp found in sigmoidoscopy -> biopsy
•   If adenomatous -> colonoscopy:
•   Risk of colorectal cancer S/P excision of small
    polyp (<1 cm.) same as general population.
      Colorectal Screening
• Q3 year colonoscopic surveillance
  results in 88-90% reduction in colorectal
  cancer (Family Practice News. 8-1-94)
• Cost - 3 billion/year
             Colorectal Screening

Allison, NEJM, 1996 - sens. spec. +PPV
  Hemoccult            32 98       23
  Hemoccult Sens       71 87        9
  Hemeselect          67 95        20
Mandel, NEJM, 1993 - 1/3 reduction in colorectal cancer
  (CRC) with hemoccults and rehydration.
              Colon cancer
        Fecal occult blood testing
• Newer tests (hemoccult Sensa, and Heme
  Select) are more sensitive.

• Newer tests less specific, resulting in high
  false positive rates.
      Colonoscopy vs. Barium
• BE safer, less costly.
• Colonoscopy diagnostic & curative.
• BE - 44% sensitive, 75% specific (Family Practice
  News. Aug. 1,1994).
Colorectal Screening Recommendations
            USPSTF 2002
• Strongly recommends screen men and women
  50 years of age or older: A
• Screening modalities;
  – FOBT, sigmoidoscopy, or FOBT + sigmoidoscopy
  – Colonoscopy
  – Double contrast barium enema
• Cost effective- <$30,000/year of life saved
  regardless which screening test used
• Interval and upper limits not specified
              Prostate Cancer
• 50% of men >80 y.o. found to have prostate cancer at
• Incidence increased from 90,000 in 1987 -> 317,000 in
• 2nd most common cause of death from cancer in men.
• 21st in years of life lost.
                Prostate Cancer*
• Cost of screening and f/u of local disease in men 50-70
  y.o. - $12-28 billion/year.
• Complications of treatment (impotence, incontinence,
• Screening results in marginal increase in life expectancy,
  decrease in quality of life, and high cost.

*Krahn M, et al., Screening for Prostate Cancer. JAMA Sept. 14,1994
     Prostate Cancer Survival
• Rate of prostatectomy increased 600% from 1984-
• Age adjusted mortality rates - no change.
• 10 year survival with stage A cancer - 85%
• 95% of men with prostate cancer die from other
• 10 times more likely to die from cardiovascular
Prostate Specific Antigen (PSA)
•   Approved by FDA, 1996 - 10% positive.
•   Large overlap between BPH & prostate cancer.
•   PSA 55-75% sensitive, 70% specific.
•   Follow-up with ultrasound, biopsy.
          Prostate Cancer Screening
                USPSTF 2002
• Insufficient evidence to recommend for or against
  routine screening with PSA or digital rectal exam: I
  – PSA can detect early stage prostate cancer.
  – Inconclusive evidence that early detection improves
    health outcomes.
  – Screening associated with important harms including
    false positives, biopsies, and complications of treatment.
  – Uncertain if benefits exceed risk
• 1.3 million osteoporosis-related
  fractures in U.S. each year
• 15% of women have hip fractures
• Strongly associated with low bone
  mineral density(BMD)
• Risk factors - female, age, anglo, low
  body weight, & bilat. oophorectomy
         Value of Screening
• Women >65 years old with low BMD are eight
  times more likely to have hip fracture
• No studies correlating perimenopausal BMD
  with long-term fracture risk
• Other risk factors-age, health,activity,vision
• ?impact on recommendations re calcium,
  hormone replacement therapy, or exercise
             Screening Tests
• Plain films
• C.T.
• Absorptiometry-measures BMD
  – Dual energy x-ray (DXA)
  – Femoral neck measure best predictor of hip fx
• Experimental - Ultrasound and biochemical
•   Calcium, exercise, safety measures
•   Hormone replacement therapy
•   Selective estrogen receptor modulators
•   Biphosphanates
    Osteoporosis Treatment
• Meta analysis of Alendronate showed
  reductions in vertebral and forearm fractures
• Fracture Intervention Trial showed benefit of
  Alendronate in hip (50%) and total fx (30%
  less) in women with low BMD only.
• Raloxifene study showed fewer vertebral fx.
• USPSTF estimates need to screen 731 women
  over 64 years old, or 1,856 women 60-64 to
  prevent one hip fracture.
      Raloxifene To Prevent
• Estrogen-like effect on bones and lipid
  metabolism (decreases total LDL cholesterol
  without changing HDL).
• No estrogen-like effects on breast or uterine
• No post-menopausal bleeding or increase in
  breast CA.
• Patients may experience hot flashes
• Decreases risk of osteoporosis, has not been
  proven to decrease fracture risk.
        USPSTF Osteoporosis
• Screen women aged 65 and older B
• Begin at age 60 for women at increased risk for
  osteoporotic fractures B
• Benefits/harms of screening and treatment too close
  to recommend for other age groups. C
   – Risk for osteoporosis and fracture increases
     with age and other factors
   – BMD measures accurately predict fracture risk
   – Treating asymptomatic women with
     osteoporosis reduces fracture risk.
       Hormone Replacement
•   Can reduce risk of fractures by 25-50%
•   Need to continue indefinitely
•   More likely to continue if have low BMD
•   Decision re HRT hinges on factors
    besides BMD
       Proceed With Caution
    Estrogen Replacement Therapy
• Risk of coronary heart disease exceeds risk of breast
  cancer (230,000 deaths from CHD, 34,000 from
  breast cancer in women older than 55 years).
• Observational studies suggested 40-50% reduction in
  fatal coronary heart disease in post menopausal
  estrogen users. (Grady, et al., Ann Intern Med,
• Observational studies do not establish causal
    Prevention of Coronary Heart
Disease in Post-menopausal Women*
• Randomized trial of estrogen plus progesterone.
  -No differences in cardiovascular outcomes, cancer, or total
  mortality despite lower LDL and higher HDL in HRT group.
  -More thromboembolic events and gallbladder disease in
  HRT group.
  -Trend toward more coronary heart disease in first year, and
  less in later years.

*Hulley, et al., JAMA, 1998;280:6055 & 613.
       Hormone Replacement
• Large RCT’s including women’s Health Initiative and the
  Heart and Estrogen/Progestin Replacement Study (HERS)
  have evaluated HRT.
• HRT beneficial in relieving vasomotor symptoms.
• HRT has beneficial effects on colon cancer and hip fractures.
• Benefits more than offset by increased risk of coronary
  events, stroke, pulmonary embolism, and breast cancer.
• Further analysis of WHI indicates HRT has no significant
  effects on general health, vitality, mental health, depressive
  symptoms, or sexual satisfaction. (Hays et al. NEJM 2003;
  348: 1839-54.)

   – *Grady D NEJM 348; 19. May 8, 2003. 1835-1837.
             Breast Cancer
• 192,000 cases of breast CA & 40,000 deaths
  in 2001
• Breast CA deaths decreased 8-9% in women
  36-59 y/o & 3-5% in women 60-79 from 1989-
• African-American women > 2 times more
  likely to die of breast CA
• More than 40% of years of life lost are from
  women diagnosed < 50 y/o
    Mammography & Breast
• Seven randomized controlled trials in
  women ages 40-74
• The six trials involving women >50
  years old demonstrate decrease in
  mortality from breast cancer of 20-30%
• No difference if screened every 12
  months or every 18-33 months
Randomized Controlled Trials of Breast Screening for
Women Age 40–49: Relative Risk (RR) of Mortality for
    Screened Subjects Versus Control Subjects
                                            # of subjects
 Trial                 Year              Screened Controls          RR
 HIP Study             1963–69                14,423       14,701   0.77
 Malmo                 1976–86                 3,658        3,679   0.51
 Kopparberg            1977–85                9,582         5,031   0.73
 Ostergotland          1977–85                10,262       10,573   1.02
 Edinburgh             1979–88                 5,913        5,810   0.78
 Stockholm             1981–85                14,375        7,103   1.04
 Gothenburg            1982–88                10,600       12,800   0.73
 NBSS-1                1980–87                25,214       25,216   1.36
 HIP—Health Insurance Plan; NBSS—National Breast Screening Study.
 Modified from Smart R, 1995.
Study Design Controversy

 •   Non-compliance and Contamination
 •   Study size & statistical significance
 •   Follow-up period
 •   Lead-time & length-time bias
 •   Inclusion of women with breast Ca.
 •   False positives
      Screening for Breast Cancer in
         Women 40-49 Years Old
• Canadian national breast screening study designed to
  answer this question
• No benefit shown -study has been criticized (Miles A. Can
  Med Assoc J 1992;147:1459-1476)
• 3 trials - no benefit, 4 trials - nonsignificant benefit of 22%
  or more
• Meta-analysis of 40-49 y.o.subgroup showed no reduction
  in breast cancer mortality (Elwood J, Online Curr. Clin.
  Trials 1993, Doc. #32)
           Benefits of Screening
         40-49 y/o                  50-69 y/o
 • 10% shift from Stage     • 40% shift from Stage
   II Ca. to Stage I          II Ca. to Stage I
 • No benefit first 9 years • No benefit first 5 years
 • 16% decrease in breast • 27% decrease in breast
   CA mortality 10-14         CA mortality after 5
   years                      years
Peer, et al. Age Specific Effectiveness … J Nat’l Cancer Inst.
Kerlikowske, Efficacy of Screening Mammography. Monogr. Nat’l
   Cancer Inst. 1997;22:79-86
          Cost Effectiveness of
• Breast CA incidence 2-3 x greater in 50-69 y/o
  40-49 age group (Saltzmann et al. Ann Intern
  Med 1997;127:955-965)
• Previous studies showing equal cost effectiveness
  did not account for 10 year lag in benefits
  (Lindfors JAMA 1995;274:881-4 Feig. Cancer
             Cost Effectiveness of
             Mammography (cont.)
• 40-49 y/o (screen q 18 mo)    • 50-69 y/o (screen q 2
• Increase life exectancy 2.5     years)
  d.                            • Increase life expectancy 12
• 4 deaths prevented/10,000       d.
  at 80 y/o                     • 37 deaths
• $105,000 per year of life       prevented/10,000 at 80 y/o
  saved                         • $21.400 per year of life
     Genetic Testing for Breast
• 5-6% of breast cancers associated with
  inherited genetic mutation.
• BRCA1 and BRCA2 among hundreds of
  mutations associated with breast cancer.
• Found in .1% of general population.
• Account for less than 1/5 of familial risk of
  breast cancer.
• Also linked with ovarian cancer.

*Isaacs C, Fletcher SW, Peshkin BN, Up To Date, last updated December 4, 2002
      BRCA 1 and 2 and Cancer*
• Ashkenazi Jews with high incidence of BRCA
  mutations studied.
• 10% of breast cancer associated with BRCA 1 or 2.
• Associated with 82% lifetime risk of breast cancer.
• Associated with 20-40% lifetime risk of ovarian

*King M. Science October 2003
Treatment Options for Breast Cancer
     Genetic Pre-disposition *
      Increased Surveillance
• Cancer Genetic Study Consortium recommends:
   – monthly BSE at age 21,
   – annual CBE beginning at age 25-35,
   – annual mammography beginning at age 25-35,
   – annual or semi-annual ovarian cancer screening with
     ultrasound and CA-125 beginning at ages 25 or 35.
• Efficacy of early and increased surveillance not known.
*Isaacs C, Fletcher SW, Peshkin BN, UpToDate, last updated April 28, 2003
    Treatment Options for Breast Cancer
         Genetic Pre-disposition *
• Prophylactic bilateral mastectomies and oophorectomies.
  - No recurrence after three years in 76 healthy women with
  prophylactic mastectomies, compared to eight cases
  amongst 63 new patient carriers who did not undergo
  - 70% satisfied re decision 14 years later, 25% less
• In one study, bilateral salpingo – oophorectomy reduced
  risk of breast cancer by over 50%.

*Isaacs C, Fletcher SW, Peshkin BN, UpToDate, last updated April 28, 2003
        Prognosis of BRCA
     Associated Breast Cancer*
• Treatment of BRCA initial breast cancer as effective
  as women with sporadic breast CA.
• BRCA women at higher risk for new primary breast
  - 30-40% ten year risk.
• Several hundred possible BRCA related mutations,
  most concerning if specific mutations identified in a
  family member with CA.

*Isaacs C, Fletcher SW, Peshkin BN, Up To Date, last updated December 4
 Treatment Options for Breast Cancer
      Genetic Pre-disposition *
                      Chemo prevention

• Selective estrogen receptor modulators (SERMs)
  such as tamoxifen and raloxifene
• Tamoxifen approved for use in women at high risk for
  breast CA by the FDA.
• No prospective studies demonstrating benefits from
  chemo prevention in BRCA carriers.
• Oral contraceptives: May increase risk of breast CA
  but decrease risk of ovarian cancer.

*Isaacs C, Fletcher SW, Peshkin BN, UpToDate, last updated April 28,
      USPSTF 2002 Breast Cancer
      Screening Recommendations
• Screening mammography with or without clinical breast exam
  every one to two years for women aged 40 years and older. B
    – Evidence strongest for women aged 50-69.
    – For ages 40-49; evidence weaker, benefit smaller, and
       optimal interval uncertain.
• Delay in observed benefit makes it difficult to determine
  incremental benefit of beginning screening at 40 rather than 50.
• Screening recommendations generalizable to age 70 and older
  if life expectancy not compromised by co-morbid disease.
• Evidence insufficient to recommend for or against clinical breast
  exam or breast self examination. I
• Has not assessed efficacy of screening for BRCA mutations.
       Lung Cancer Screening
• 155,000 deaths per year - most related to smoking.
• Screening methods include chest X-ray, spiral CT,
  sputum analysis, and bronchoscopy.
• Five year survival 15%, 60% if tumor stage 1a.
• Spiral CT screening in Japan increased five year
  survivals from 15% to 34%.

*Patty JAMA 10-18-2002, 284: 15. 1977-1980
Lung Cancer Screening-Why not?
• Despite improvements in five year lung ca. survival
  rates, overall mortality in screened populations
  unchanged even after 25 years of follow-up. (Marcus P.
  et. al. J Natl Cancer Inst. 2000; 92 (16): 1308-16.)

• Screening programs pick up more indolent cancers,
  adeno- carcinoma versus squamous cell.
• Spiral CT screening picked up equal numbers of
  cancers in smokers and nonsmokers, despite lethal
  lung cancer being 10 times more common in
  smokers. (Sones Lancet 1998; 351: 1242-45.)
• Lung CA can be asymptomatic - almost half of
  patients assessed for lung reduction surgery have
  lung CA. (Pigula F. Ann Thorac Surg. 1996; 61: 174-76.)
• Lead time and length time bias. (Woloshins Lancet 2002; 359:
            Comparison of Cancer
              Screening Tests*
                            REDUCTION      TO SCREEN
Pap smear for                  >0.80          1,140
cervical cancer
Mammography                         0.23          543
age >50 years
Mammography                         0.08         3,125
age 40-49
Fecal-occult                    0.15 - 0.20    588 - 1,000
blood Colon Ca.

*Gates TJ Am Fam Physician. 2001; 63: 513-22
        Screening for Lipid Disorders
              USPSTF 2001
• Important risk factor for coronary heart disease.
• Coronary heart disease leading cause of
  mortality in U.S. - 500,000 deaths/year.
• 1/2 of men, and 1/3 of women will have coronary
  heart disease event in their lifetime.
• 17% of men, and 20% of women in U.S. have
  total cholesterol >240.
• 27% of coronary heart disease events in men,
  and 34% in women attributable to total
  cholesterol >200mg/dl.
    Screening for Lipid Disorders
• USPSTF recommendations based on four RCTs showing decreases
  in CHD events of 19%-37% and CHD mortality of 20%-28%.
   – Inconclusive regarding total mortality.
• ALLHAT study “no significant impact on mortality*”
   – Treated with pravastatin 40mg daily.
   – Total cholesterol level 17% lower, and LDL cholesterol levels 28%
      lower in pravastatin group.
   – Usual care group had 8% decrease in total cholesterol and 11%
      drop in LDL cholesterol.
   – All cause mortality no different after 4.8 years.

   – *JAMA 288 [23]: 2998-3007, 2002.
   Screening for Lipid Disorders
   USPSTF Recommendations
• Routinely screen men 35 and women 45 y.o. for lipid
  disorders and treat if at increased risk for CHD: A
• Routinely screen men age 20-35 and women age 20-
  45 if other risk factors present: B
• Screen with total cholesterol and high density
  lipoprotein levels: B
    – Can be measured with non-fasting sample.
• Insufficient evidence for or against triglyceride
  screening: I
• Interval (5 years?) and upper age limit (65?) not
               Type II Diabetes
• Screening recommended by ADA after age 45.
• Cost of screening on all persons aged 25 or older
  estimated at $236,000 per life year gained ($57,000
  per quality adjusted life year gained).*
• Based on single screening only.
• Reduces lifetime cumulative incidence of end stage
  renal disease, blindness, and lower extremity
  amputation by 26%, 35%, and 22% respectively.
• More cost effective in younger individuals and

*CDC diabetes cost effectiveness study group, JAMA, November 25, 1998:280, No.
   20, 1757-1763.
Screening for Microalbuminuria
• 3-8% of diabetics have macroalbuminuria
• 20-30% of diabetics develop nephropathy.
• Over half of all dialysis patients are diabetic.
• Diabetes Control and Complications Trial (DCCT) with
  Type 1 diabetics demonstrated benefit of enalapril on
  blood pressure, serum creatinine, and albumin
  excretion (N Engl J Med, 1993;9:977-86).
• Screening for microalbuminuria recommended by ADA,
  NIH,and WHO, all consensus-based.
• No RCTs have evaluated efficacy of screening
  diabetics for microalbuminuria in reducing renal failure.
• Control of BP and lipids more important in reducing
  microvascular complications than tight glucose control.
    USPSTF Diabetes Screening
• Insufficient evidence to recommend routine
  screening in asymptomatic adults: I
   -Tight control of glucose does not significantly affect
     macrovascular complications
   -Tight control benefits microvascular complications
     but takes years to manifest, uncertain benefit of
     early detection
• Screen adults with HTN or hyperlipidemia for
  diabetes : B
• Tight glycemic and BP control reduce albuminuria
  but uncertain if important impact on renal failure.
          Serum Tumor Markers*
• Prostate Specific Antigen (PSA)
• Cancer antigen (CA) 27.29-monitor response in metastatic
  breast CA patients.
• Carcinoembryonic antigen (CEA) – detect colorectal
• CA 125 – used to evaluate pelvic masses in post-
  menopausal women, therapy for ovarian CA, and detect
• Alphafetoprotein (AFP) – marker for hepatocellular CA.
• With the exception of PSA, not sensitive or specific enough
  to be used in screening.
• “No tumor marker has demonstrated survival benefit in
  randomized control trials of screening in the general
*Perkins GL, 2003;68:1075-82, AFP
           Proceed With Caution
           Cerebral aneurysms*
• 15,000,000 Americans may develop aneurysms.
• Ruptured aneurysms account for 20% of the 3,000,000
  strokes annually in the USA, and 80% of stroke deaths.
• More and more detected as incidental findings on MRI.
• Cerebral bleeding or stroke in asymptomatic individuals
  with aneurysms less than 10 mm in diameter-.05% per
• Complications or deaths from corrective surgery-13% in
  first year.

*Wiebers, et al., N Engl J Med 1998;339:1725-33.
          Medicare Coverage of
           Preventive Services
• Expanded with Budget Reconciliation Act, August 1997.
• Estimated cost - 2 billion/year.
• Annual mammos - 40 y.o. and older.
• Pelvic exam & pap smear - q 3 years.
• Annual prostate screening in men >50 y.o. with Digital
  Rectal Exam and PSA beginning in year 2000.
• Colorectal screening >50 y.o. with Fecal Occult Blood q
  year, Flexible sigmoidoscopy q4 years, Colonoscopy
  and barium enemas q 2 years in high risk groups.
• Bone mass measurements in high risk groups.