Inequalities in Children’s Educational Outcomes:

Using Administrative Data to Gain a Population-Based Perspective on

Health

Marni Brownell, Noralou Roos, Randy Fransoo, Leslie Roos, Anne Guèvremont,

Leonard MacWilliam, Lauren Yallop,

Ben Levin & Beth Edwards

Questions to be Addressed:

1. Why focus on educational outcomes when studying population health?

2. What are administrative databases?

3. What do administrative databases tell us about child outcomes that is different from other data sources?

1. Why focus on educational outcomes when studying population health?

First, what is health?

WHO originally defined it as:

A state of complete physical, mental and social well-being and not merely the absence of disease or infirmity. (1948)

In 1986 the WHO stated that at every level of planning health promotion should emphasize “a social, economic and ecological, rather than a purely physical and mental

perspective on health.” (3)

Research on the social determinants of health – much more to health than health care

Examples:• Marmot et al. (1991) - Whitehall study of British Civil

Servants (6)

• Evans et al. (1994) – Why Are Some People Healthy and Others Not? (2)

• Hertzman et al. (2002) – Child development as a determinant of later health outcomes (4)

• Education levels strongly related to health outcomes

• Education outcomes have inter-generational effects

• Education levels may be more readily changed by policy initiatives than other components of socioeconomic status (SES)

2. What are administrative databases?

• Data originally collected to provide and manage services• extremely useful for research

• Linkages made across data sets using anonymized identifiers make cross-service and longitudinal research possible

• Manitoba Centre for Health Policy – model for linked administrative databases

Manitoba Population Health Research Data Repository

Population-Based Research Registry

Hospital

Home Care

Pharmaceuticals

ImmunizeVital

Statistics

Costs

Nursing Home

Physician

Family Services Education

Key health databases start in 1970

Census Data at area level

National surveys

Healthy Child Program Data

- Birth weight, gestation

- Injuries

- Chronic diseases (asthma, diabetes)

-Meds

-dosage

-diagnosis

-income assistance

-in care

-standards tests

-high school marks

-graduation

-retention

-residence

-marital status

-family size

3. What do administrative databases tell us about child outcomes that is different from other

data sources?• Combined with information from a research registry they provide a

population-based perspective that might not be available otherwise

• We can combine information to examine underlying causes of inequalities in child health:

- Area level socioeconomic status- Age of mother- Birth weight- Apgar scores- Educational Outcomes

The following slides will demonstrate how educational outcomes vary with socioeconomic status

SES Groups, Winnipeg 2001 Census

Neighbourhood Socioeconomic Status

High SES (Most Advantaged) Pop = 48,789; Child pop: 13,087

Middle Class Pop = 354,712; Child pop: 90,272

Low-Mid Pop = 140,469; Child pop: 32,803

Low SES (Most Disadvantaged) Pop = 104,989; Child pop: 28,202

HighMiddleLow-MidLow

Assessed by: High school education, Unemployment rate, Single parent families, Female singe parent families, Female labour force participation

Grade 12 (S4) Performance by SES Group Language Arts Standards Test 2001/02

Pass/Fail rates of test writers 17/18 year olds who should have written

N=221

121

98276

31

75%83%

92%87%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SES

27%

52%

65%

77%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SES

Withdrawn

In Grade11 (S3) orlower

In Grade12 (S4),but no LATest Mark

DropCourse,Absent,Exempt, IncompleteFail

Pass

Recovery: What happens to the retained students in 2 yrs after 01/02? (Percent Graduated)

0%

10%

20%

30%

40%

50%

SES

Low High

High School Completion by SES & Mother’s AgeGrade 9 (S1) students in 1997/98: What happens in next 5 years?

37%

60%

74%81%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SES

Withdrawn

Continuing

Near Grad

Graduate

Grade 3 Performance by SES and Mother’s Age(Language Arts Standards Test 1998/99)

83%91% 94%93%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SES

Pass/Fail Rate of Test Writers

50%

70%78%

84%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SESPassed FailedAbsent ExemptIncomplete Grade 2 or lower

Eight year olds who should have written

Healthiness of Children at Birth(1984) by Winnipeg SES Group

% of Children Normal Birthweight

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SES

% of Children with "Good" Apgar Scores

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

Low Low-Mid Middle High

SES

Infant Hospitalization Rate First Year of Life: (Children Born 1998/99-2000/01)

0

20

40

60

80

100

120

140

160

180

Low Low-Mid Middle High

SES

pe

r 1

,00

0 o

ne

-ye

ar-

old

s

Percent Winnipeg Children Enrolled in Reading Recovery Program, Grade One, 2001

0%

2%

4%

6%

8%

10%

12%

14%

Low Low-Mid Middle High

SES

Child Care Spaces for 0 to 12 year olds, by Winnipeg Neighbourhoods, 2001 (per 1000 children)

0 50 100 150 200 250 300

HighSES

Middle

Low-Mid

LowSES

What can be done?• Enriched early childhood environments

– Quality child care, especially for kids in low SES families– Research shows that quality ECD helps all, but makes largest

difference for those at highest risk

• Enhanced programs in school years (e.g. early literacy programs)

• Engage with health authorities, community groups, and parents to make programs aimed at enhancing childhood development universal and needs-based

• Appreciate the short and long term health implications of educational outcomes.

Conclusions• Differences in outcomes across SES may be

dramatically underestimated without a population-based approach

– Disadvantaged groups are at very high risk for poor outcomes

• Not all disadvantaged children do poorly

• Of the total number of kids with poor academic outcomes, the majority are not in the most disadvantaged groups

– But low SES kids are much less likely to recover from a setback

This research has centered on school achievement, but the focus of policies aimed at changing the trajectories of disadvantaged children should not be limited to the

school system. Our analyses and work by others (e.g. Hertzman et al. 2002)(4) reveal that, while the vast

majority of children at every socioeconomic level show remarkable similarities at birth, inequalities in

achievement are evident early in childhood, prior to school entry. Children who are already behind their peers when they begin school will likely fall further

behind; engaging them in the educational process may be difficult. This makes it imperative for governments to provide effective early childhood programs (starting in the first few years of life) to improve the experiences of children at risk, and to improve the physical, mental and

social well-being of all children.

Final Thoughts…