Measurement and Finance

H Edu 5100

Quantitative perspectives

The size and demographic characteristics of the population

The direct measures of health and ill health in a population

Use of services

Census data

Every 10 years Textbook numbers

1996 = 79.9% of lived in MSAs1980 = 78%1960 = 63%

Census data

Textbook numbers1998 = 48.9% male (49.1% in 2000 census)

87.3% non-Black (87.7% in 2000 census)

35.2 = median age (35.3 in 2000 census)

12.7% was 65 and over (12.4% in 2000)

(80.2% lived in MSA from 2000 census)

Why?

What do these data tell us? Why do we need them?

Epidemiology

The study of how and why diseases are distributed in the population…

Why some get sick and some don’t

Some key words…

Study Distribution Determinants Health-related states Populations Control

1/2

1 = numerator 2= denominator

Rate

Denominator = 1,000 or 10,000 Base Comparisons

Rate

Denominator = population at risk Numerator = the events The population at risk has to be all of the

people or things that the numerator could have happened to

Remember…

The numerator has to have all the events you’re measuring

The denominator must include exactly the number that the numerator’s event could have happened to

Rates with the same base can be compared

Definitions

Risk = the probability of an unfavorable event.

Risk factors = factors associated with an increased risk of acquiring disease

Try this…

There are more drownings in Los Angeles each year than in Las Vegas.

Los Angeles must have a less adequate swimming instruction program.

Los Angeles

What’s the population at risk? Los Angeles has a much larger population

at risk. Las Vegas is in the desert, reduces the

population exposed to the risk of drowning.

Do this on your own…

A new and puzzling disease has become epidemic in the Midwest. It is significant that 80 percent of those affected live within a mile of a railroad track. It is therefore obvious that railroads must be somehow related to the disease…

What sounds best?

We reduced the rate from 2/1000 to 1/1000

We reduced the rate from 20/10,000 to 10/10,000

We reduced the rate from 2,000/2,000,000 to 1,000/1,000,000

Check-off list

Establish a problem exists Confirm the homogeneity of the events Collect all the events Characterize the events

Characterizing events

Person Place Time

Sex Residence Year

Occupation Occupation Season

Age Events Day, etc.

Economic group

Anatomic site Onset

Immunization status

Geographic site

Duration of an activity

Check-off list (cont)

Look for patterns and trends Formulate an hypothesis Test the hypothesis Write up results

Epidemic Curves

Endemic = whenever diseases are regularly and continuously present

Epidemic = the occurrence of disease that clearly exceeds normal expectancy.

Pandemic = epidemic that has gone international

Pneumonia and influenza

Respiratory Insuficiency

Viral Hepatitis

Malaria in Bakersfield, CA

You figure it out

Incidence and prevalence

Prevalence = how many people have a disease at any one moment.

Incidence = how many incidents of the disease are just starting. The new cases.

Illness Among 20 People

Figure V-1

20 More

The Anteater

Anteater

At dinner 3,126 red and 5,210 black ants 3 red to 5 black ratio (not rate) Prevailing ratio in the area is 9 red and 10

black. For every 10 black ants along Random

Path, 9 red ants could come to dinner, but only 6 do.

Ant Queens

What are the odds of each ant being dinner?

Does this mean we’ll have to move? Can I handle the loss with a few more

eggs and if so, how many more eggs?

Ant Queens

Both queens kept track of the next 1,000 ants to graduate from the nursery.

By the end of the season, the ultimate fate of each was known.

Red Ant Queen

Grim Reaper visited in forms such as birds, careless feet, being sat upon, etc.

Anteater accounted for 400

Black Ant Queen

500 ended up as dinner

Direct comparison

Red ants have 80 percent of the risk that black ants have…

(400/1000 vs 500/1000) The anteater could know this…

Retrospective and Prospective

Retrospective (case control) = begin after the fact and look back

Prospective (cohort) = start with a condition in question and watch it over a period of time to see what happens

See if you understand…

The smoking histories of all patients entering a hospital with cold sores who enter the same hospital

Retrospective

See if you understand…

The physical examination records of all incoming freshmen class of 1928 at U.C. are examined to see if their recorded height and weight are related to their chance of developing coronary heart disease by 1968.

Prospective

See if you understand…

All the mothers of Grass Valley who delivered a baby in 1970 are interviewed to see if the number of colds they had during their pregnancy is related to any defects apparent at birth in their infant.

Retrospective

See if you understand…

A whole town is examined and all who are judged to be well are questioned extensively about their diet to see whether or not their eating habits will predict their risk of developing subsequent bowel cancer.

Prospective

See if you understand…

All the patients being discharged from one hospital in NYC with a diagnosis of myocardial infarction (heart attack) are characterized by certain demographic variables and compared to NYC population to see if any of these characteristics are determinants of survival from a heart attack there.

Retrospective

Cause and Association

Association = relationships which may exist between the occurrence of one thing (like a risk factor) and the occurrence of another (like a disease).

Causality …

Causality

If you have “it” and the “disease” is more likely to occur, and

When you take “it” away, the “disease” is less likely to occur, then

“it” is considered a cause of the “disease.”

Five basic criteria…

Consistency Strength Specificity Time relationship Coherence

Practice 1

During the previous year, nine residents of a community died from the same type of cancer. List some of the reasons that might justify an investigation.

Practice 1 ideas

Determine how many cases to expect Research Control and prevention Why people died Training Politics Legal concerns

Practice 2

During August, a county health department received reports of 12 new cases of tuberculosis and 12 new cases of aseptic meningitis. Tuberculosis does not have a seasonal distribution; however, aseptic meningitis, which is caused primarily by an infection, is highly seasonal, and peaks from August-October. What additional information is needed to determine whether either of these groups of cases is an outbreak?

Practice 2 ideas

How many diseases usually occur

Practice 3

You are called to help investigate a cluster of 17 men who developed leukemia in a community. Some of them worked as electrical repair men, and others were ham operators. Which study design would you choose to investigate a possible association between exposure to electromagnetic fields and leukemia?

Practice 3 ideas

Case-control Compare the 17 people already identified

with the control Were they exposed to the same

electromagnetic fields

Practice 4

The manager of a grocery store has reported a rash illness among the store’s workers. What type of study would you use to determine the source of the outbreak? Why? What is the appropriate measure of association? Review the table on exposure to celery and calculate the measure of association and interpret your results.

Practice 4 table

Total

workers

Cases

Exposed 56 25

Non-exposed

70 5

Practice 4 table

Total

workers

Cases Attack

rate

Exposed 56 25 25/56 = 44.6%

Non-exposed

70 5 5/70 = 7.1%

Practice 4 table

Total

workers

Cases Attack

rate

Exposed 56 25 25/56 = 44.6%

Non-exposed

70 5 5/70 = 7.1%

Relative risk

44.6/7.1 = 6.3

Practice 4 ideas

Cohort study Relative risk … attack rate for workers exposed to

celery divided by the attack rate for those who were not exposed.

Attack rate for exposed = 25/56 or 44.6% Attack rate for non-exposed = 5/70 or 7.1% Relative risk = 44.6/7.1=6.3 Compare other vegetables and fruits Look for significance

Vital Statistics

Births Deaths Fetal deaths Marriages Divorces

Infant Mortality

Infant Mortality Rates

Country Rate per 1,000 live births

Sweden 2.77

Japan 3.28

Finland 3.59

Germany 4.20

Australia 4.76

Canada 4.82

United Kingdom 5.22

United States 6.63

Afghanistan 165.96Source: http://www.geographyiq.com/ranking/ranking_Infant_Mortality_Rate_aall.htm

Infant Mortality Rate

Utah = 5.3 deaths per 1,000 live births

Morbidity = sickness & disease

State health department data

Health status & health-related behaviors

Causes of death in the U.S. 2001Rank Number Percent of

total deathsRate

1

2

3

4

5

6

7

8

9

10

11

12

13

14

All Causes

Diseases of heart

Malignant neoplasm

Cerebrovascular diseases

Chronic lower respiratory diseases

Accidents

Diabetes mellitus

Influenza and pneumonia

Alzheimer’s disease

Nephritis

Septicemia

Intentional self-harm (suicide)

Chronic liver disease

Assault (homicide)

Essential hypertension

2,416,425

700,142

553,768

163,538

123,013

101,537

71,372

62,034

53,852

39,480

32,238

30,622

27,035

20,308

19,250

100.0

29.0

22.9

6.8

5.1

4.2

3.0

2.6

2.2

1.6

1.3

1.3

1.1

0.8

0.8

848.5

245.8

194.4

57.4

43.2

35.7

25.1

21.8

18.9

13.9

11.3

10.8

9.5

7.1

6.8

Causes of death in the U.S. 1900Rank Number Percent of

total deaths

Rate

1

2

3

4

5

6

7

8

9

10

All Causes

Pneumonia

Tuberculosis

Diarrhea, enteritis, ulceration of the intestines

Diseases of the heart

Intracranial lesions of vascular origin

Nephritis

All accidents

Cancer and other malignant tumors

Senility

Diphtheria

343,217

40,362

38,820

28,491

27,427

21,353

17,699

14,429

12,769

10,015

8,056

100.0

11.8

11.3

8.3

8.0

6.2

5.2

4.2

3.7

2.9

2.3

72.3

64.0

50.2

40.3

38.3

31.7

25.9

22.9

17.9

14.4

Healthy People 2010

Use of health care services

Hospital discharge data Health department data

Sources and Uses of Health Care Funds

1929 1965 2002

Total health spending (millions) $3,656 $41,012 $1,545,900

Adjusted for inflation (2002 $$) 32,400 192,300 1,545,900

Per capita (adjusted) 305 962 5,427

As a % of GDP 3.5% 5.7% 14.7%

% Paid by

Self (out-of-pocket) 81% 44% 15%

Third parties 19% 56% 85%

Government 13% 25% 45%

Private insurance < 1% 25% 35%

Philanthropy, other 6% 6% 5%

Flow of funds

Exchange Two-party transaction Derived demand

Role of Government

Sources

Shift to third-party payment Payments are made from…

Income and ethnicity as determinants of mortality (1996 data)

Mortality rate

Income Category White Black

< $9,999 0.918% 1.234%

$10,000 - $14,999 0.840% 1.123%

$15,000 - $19,999 0.706% 0.899%

$20,000 – $24,999 0.660% 0.867%

$25,000 - $29,999 0.591% 0.603%

$30,000 + 0.542% ****

Use of funds Intensity of services Hospitals Nursing homes Doctors, nurses and other health care workers Drugs Administration Public health Research