Teaching Introduction to Public Policy

1
Teaching Introduction toPublic Policy Research
Methods

• Rick Mayes, Assistant Professor
• Department of Political Science

2
Background Objectives

• Public Policy 101 Incentives structure modern
  life as we know it.
• e.g., IRS and tax audits, HOV lanes and toll
  roads, Deans List and
• academic probation, parenting, teaching,
  dating, sales, Amway, etc.
• Incentives come in 3 basic flavors or varieties
  (e.g., smoking)
• (a.) moral U.S. govt asserts that terrorists
  raise money from black-market sales of cigarettes
• (b.) social banning of cigarettes in restaurants
  and bars
• (c.) economic 3-per-pack sin tax ( but not
  in Virginia obviously)

3
Moral/Social Incentives and Modern Life

• The Chicago Police Department in conjunction
  with the Mayor's office have now made
  prostitution solicitors' information available
  online. By using this website, you will be able
  to view public records on individuals who have
  been arrested for soliciting prostitutes or other
  related arrests. The following individuals were
  arrested and charged for either patronizing or
  soliciting for prostitution. It is not a
  comprehensive list of all individuals arrested by
  the Chicago Police Department for patronizing or
  soliciting for prostitution. The names,
  identities and citations appear here as they were
  provided to police officers in the field at the
  time of arrests.

                       DOE/SMITH, CARLOS M/31
165XX BRENDEN LN. OAKPARK 1102 N CICERO AVE
2005/10/01 720 ILCS 5.0/11-15-A-1
DOE/SMITH, JOSE M/37 54XX S ROCKWELL ST
CHICAGO 1102 N CICERO AVE 2005/10/02 720 ILCS
5.0/11-15-A-1
DOE/SMITH, JOHN M/54 28XX W 38TH PL CHICAGO
2500 S CALIFORNIA BLVD 2005/09/06 720 ILCS
5.0/11-15-A-1
DOE/SMITH, ALEX M/28 22XX MAGNOLIA CT WEST
BUFFALO GROVE 1102 N CICERO AVE 2005/10/02
720 ILCS 5.0/11-15-A-1
4
Economic Incentives and Modern Life

• - Australian prison ships in the early 1900s
• - April 15, 1987 and the disappearance of
• of 7 million American children
• - frequent flyer miles (loyalty programs)

5
Incentives Driving Medicares Pay for
Performance

• Institute of Medicine reports
• - To Err is Human (1999)
• - Crossing the Quality Chasm (2001)
• (2) John Wennberg Small-Area Large-Variation
  studies
• - tonsillectomy rates (1977)
• - cesarean section rates (1996)
• - variation in Medicare spending/per
  beneficiary

6
(No Transcript)
7
(No Transcript)
8
Number of Acute Care Hospital Beds/per 1,000
Residents
Source Dartmouth Atlas of Virginia
9
Number of Hospital Discharges for all Medical
Conditions (DRGs)
Source Dartmouth Atlas of Virginia
10
Average number of physician visits per patient
during last six months of life who received most
of their care in one of 77 best US hospitals
Source John Wennberg (2005)
11
Incentives Driving Medicares Pay for
Performance

• Researchers and Insurers Conclusions
• (1.) Physician practice styles vary considerably,
  especially regarding
• diagnoses for which treatment decisions are
  not driven by consensus
• on appropriate care and it is not possible to
  obtain evidence-based
• guidelines from reading journals or consulting
  textbooks.
• e.g., back surgery rates (the /per 1,000
  Medicare beneficiaries)
• - 7/per 1,000 in Naples, FL
• - 2/per 1,000 in Hanover, NH
• - 4.5/per 1,000 national average
• (2.) In medicine, supply generally creates its
  own demand

12
Rates of Surgery for Back Pain/per 1,000 Medicare
Enrollees
Source Dartmouth Atlas of Virginia
13
Rates of four orthopedic procedures among
Medicare enrollees in 306 Hospital Referral
Regions (2000-01)
Source John Wennberg (2005)
14
Association between cardiologists and visits per
person to cardiologists among Medicare enrollees
(1996) 306 HRRs
Source John Wennberg (2005)
15
The Association between Hospital Beds and
Hospitalization for Hip Fracture and Congestive
Heart Failure
Source Dartmouth Atlas of Virginia
16
Interview w/Tom Scully, former CMS Administrator
(2002)

• Mayes Others Ive interviewed have said that
  hospitals will cry, cry, cry about their
  finances and level of Medicare reimbursement,
  but that sometimes you have take it with a grain
  of salt.
• Scully Oh, theyre doing great!  Ill tell you,
  go find me a hospital that hasnt built a giant
  new bed-tower in the last few years.  Theyve
  actually slowed down, because the government has
  phased out Medicare capital (reimbursement) We
  used to pay for capital in Medicare it was a DRG
  add-on for capital expenditures.  Well, if youre
  getting 40 percent of your revenues from Medicare
  and you want to build a new building and Medicare
  will pay for 40 percent of it, right?  Then why
  not? 
• So what you were getting all through the 1980s
  was a massive building spree up into the
  early 1990s and even through the 90s, because it
  was a 10-year phase out of the DRG add-on for
  capital.  If you wanted to build a new hospital
  wing in 1990even if you didnt have any patients
  for itif you budgeted 100 million, Medicare
  would write you a check for 40 million!  So what
  do you get?  You got a hell of a lot of big new
  hospital wings, need them or not. This is one of
  the reasons weve had such massive over-capacity
  
• Youd have to be an idiot not to put up a new
  building every couple of years, because Medicare
  paid for such a big part of it.  That is slowing
  down now and youre starting to see the demand
  catch up on capacity in a lot of markets.
• Roemers Law A hospital bed built is a
  hospital bed filled. (behavior is unconscious)

17
Association between of hospital beds per 1,000
residents (1996) and discharges per 1,000
(1995-96) among Medicare enrollees in 306 HRRs
Source John Wennberg (2005)
18
(No Transcript)
19
(No Transcript)
20

• relationship between
• Abortion Crime

21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
Early Legalizers New YorkCalifornia
Washington Alaska Hawaii
27
(No Transcript)
28

• relationship between
• Psychostimulant Use
• and
• Educational, Socioeconomic, Demographic and
• Health System Characteristics

29
Background Significance

• So what? What is the significance of ADHD and
  psychostimulants?
• ADHD is the most commonly diagnosed behavioral
  disorder in children, making up more than 50 of
  all child psychiatric diagnoses
• 75-85 of children diagnosed with ADHD are
  prescribed psychostimulant drugs (e.g.,
  Ritalin, Adderall, Dexedrine, Concerta)
• the U.S. population consumes 90 of the worlds
  production of psychostimulants
• school-age children in the U.S. consume 4 times
  more psychiatric medication than children in the
  rest of the world combined

30
Background Significance

• So what? What is the significance of ADHD and
  psychostimulants? contd . . .
• an estimated 3-5 of school-age children have the
  disorder (NIH Consensus Statement, 1998) other
  estimates 4-12 (Vanderbilt, MUSC)
• (1.5-to-6 million kids, 1-to-3 students in
  every classroom in the U.S.)
• ADHD prevalence estimates from actual community
  studies range from as low as 1.7 (Calif. Bay
  Area) to as high as 20 (military bases near
  Virginia Beach)
• rates of psychostimulant use vary as much as
  3-fold between states and 10-fold within them
  (Rappley et al., 1995 Wennberg Wennberg, 2000
  Zito et al., 1997)
• At the peak age for psychostimulant use, 11,
  almost 1 in every 10 boys in the U.S. uses these
  drugs (Cox et al., Journal of Pediatrics,
  February 2003).
• big approximately 1.2 billion spent in 2001
  on ADHD related drugs
• (new potential growth markets pre-schoolers
  and adults)

31
Diagnostic and Prescribing Trends

• 400-500 increase in ADHD diagnoses since 1991
• 1991 800,000 to 950,000 children diagnosed with
  ADHD
• 2001 4 to 4.25 million children diagnosed with
  ADHD
• 800-900 increase in psychostimulant use since
  1991
• 1991 2 million psychostimulant prescriptions
• 2001 21 million psychostimulant prescriptions

32
Diagnostic and Prescribing Trends American
Exceptionalism
33
Mayes, Bokhari, Scheffler (2005) Conditional
Probability of a Child Being Diagnosed with ADHD
34
Primary Research Question

• What explains the enormous variation
• in psychostimulant consumption across the U.S.?

35
Methylphenidate and Amphetamine Distribution,
2000 (DEA data)(average 4,150 grams/100,000
individuals)
0 to 1,600 Low
(4.6)
1,600 to 3,150 Below Average (25.5)
3,150 to 5,150 Average (43.5)
5,150 to 6,750 Above Average (19.6)
6,750 to 8,350 High
(4.9)
8,350 to 11,000 Extremely High (1.8)
36
Characteristics of Counties with High and Low Use
of Psychostimulants, DEA ARCOS data (2000)
difference in means significant at the .10 level,
at the .05 level, at the .01 level
37
Characteristics of Counties with High and Low Use
of Psychostimulants, DEA ARCOS data (2000)
difference in means significant at the .10 level,
at the .05 level, at the .01 level
38
Selected Regression Coefficients for Market
Model indicates that the coefficient is
significant at the 10 level, at the 5 level,
at the 1 level
39
Selected Regression Coefficients for Market
Model indicates that the coefficient is
significant at the 10 level, at the 5 level,
at the 1 level