Health Disparities

Slide 1 · Health Disparities

Determinants • population genetics • environmental exposures • clinical inequities

Dr Abu Suraih Sakhri

MBBS, MD (Pathology)

www.abusuraihsakhri.com

Slide 2 · Learning objectives

Define incidence, prevalence, severity, and burden disparities.

Explain intersecting social, environmental, and health care determinants.

Distinguish socially defined race, ethnicity, and geographic ancestry.

Interpret the pollution, mortality, CPR, eGFR, and genetic-disease examples.

Slide 3 · Health disparities: What the term measures

Unequal incidence and prevalence across population groups

Differences in disease severity and cumulative burden

Unequal patterns of injury and mortality

A population-level description—not a diagnosis or biological trait of a person

Slide 4 · Incidence, prevalence, severity, burden

Incidence

New cases in a defined period

Prevalence

Existing cases at a given time

Severity

How serious the condition becomes

Burden

Aggregate health impact, including disability or death

Each measure answers a different epidemiologic question.

Slide 5 · Health disparities are comparisons between groups

Useful comparison

Specify the population and comparison groups

Choose an outcome such as mortality or disease burden

State the observation and its time/place

Invalid shortcut

Do not equate group labels with causal mechanisms

Do not infer an individual’s biology from a group average

Do not confuse description with explanation

A disparity prompts causal investigation; It does not identify a cause by itself.

Slide 6 · A multideterminant view of health

Material resources

Employment · income · food security

Living conditions

Housing · neighborhood infrastructure

Clinical access

Health care · insurance status

Social context

Education · ethnicity · partnership status

Policies, behaviors, racism, and sexism interact with these domains.

Slide 7 · Health status is not reducible to one determinant

Social and policy context

Resources and exposures

Access to care

Disease outcomes

Multiple determinants operate together.

The same outcome may reflect several linked pathways.

Slide 8 · Education as a context for disparities

Education level is explicitly named among disparity-associated factors.

It should be examined alongside income, employment, and access to care.

An education gradient does not establish an intrinsic biological difference.

Identify correlated circumstances before drawing causal conclusions.

Slide 9 · Employment and income shape material conditions

Employment / income status

Material resources

Housing / food / care access

Health outcomes

Employment and income among interacting determinants.

This flow depicts a plausible linkage, not measured effects from the textbook.

Slide 10 · Food security is a determinant, not a diagnosis

Food security belongs in an account of population health inequality.

It can intersect with household resources and neighborhood conditions.

Assessment of disparities should include food-related context.

The Health Disparities section provides no specific nutritional prevalence estimates.

Slide 11 · Housing and neighborhood conditions

Relevant circumstances

Housing circumstances vary across groups

Neighborhoods may lack adequate infrastructure

Proximity to industrial activity may alter exposure

Clinicopathologic interpretation

Look for environmental exposure opportunities

Consider access to high-quality services

Do not attribute effects to race as a genotype

The environment experienced by a group can differ from its genetic ancestry.

Slide 12 · Infrastructure influences opportunities for health

Roadways and rail

Sources of traffic-related exposure

Ports and industry

Concentrated surrounding emissions

Water infrastructure

Availability of safe drinking water

Care infrastructure

Access to high-quality health services

These settings enter the disparity pathway through differential exposures or access.

Slide 13 · Health care access is a core determinant

High-quality care may be less accessible to marginalized groups.

Access influences whether disease is recognized and addressed promptly.

Inadequate access to important mortality disparities.

Barriers can also operate outside conventional clinical settings.

Slide 14 · Insurance status is separate from care quality

Coverage dimension

Insurance status is one recorded dimension

Coverage can be considered alongside income and employment

A coverage label alone cannot capture all barriers

Delivery dimension

Actual availability of care also matters

Quality of available care is relevant

Neighborhood and discrimination may overlap

Health insurance and health care access are separately named.

Slide 15 · Partnership and marital status: A contextual factor

Partnership or marital status is among the factors listed.

It belongs in a multidomain characterization of population groups.

It should not be used as a deterministic biological explanation.

Interpret its apparent association in the presence of other determinants.

Slide 16 · Racism and sexism can modify other determinants

Beliefs / discriminatory behavior

Unequal opportunities or treatment

Different exposures / access

Unequal outcomes

Robbins explicitly identifies racism and sexism.

Do not treat these as patient-specific genetic properties.

Slide 17 · Political, economic, and health policies matter

Economic policies

Context for employment and income

Political policies

Context for social organization

Health care policies

Context for access and delivery

Personal behavior

May interact with the broader setting

The source emphasizes interaction, not a single universal pathway.

Slide 18 · Marginalization: Interacting disadvantages

Discrimination

Neglected infrastructure

Barriers to quality care

Poor health outcomes

One community may face several disadvantages simultaneously.

The source explicitly combines these examples.

Slide 19 · Social determinants and personal behaviors

Context and opportunity

Education, income, housing, and neighborhood

Health care coverage and access

Discrimination and structural policies

Individual-level factors

Individual beliefs and behaviors also influence health

Behavior occurs within a social setting

Avoid blaming individuals for structural exposures

Behavioral and structural factors as interacting.

Slide 20 · Interpretation: Association does not specify causation

A group-level excess of disease is a finding, not a full mechanism.

Consider exposure, resources, clinical access, and discriminatory processes.

Multiple explanations can coexist and require separate assessment.

Avoid converting an observed association into biological essentialism.

Slide 21 · Why does Robbins interrogate “race”?

Historically, societies sorted people using physical appearance or origin.

These socially applied labels do not define genetic populations.

Population genetics tests variation within and between human groups.

The clinical hazard is mistaking a social category for a genotype.

Slide 22 · Genetic polymorphisms challenge discrete races

Previously defined “races”

Genetic polymorphism studies

Within/between variation assessed

Discrete groups not supported

This evidence has accumulated since the 1960s.

The finding concerns genetic boundaries, not social consequences.

Slide 23 · Within-population variation exceeds between-group variation

Within groups

People assigned the same broad population label are genetically variable

Broad labels do not describe homogeneous genotypes

Between groups

Differences between continental population groupings exist

Their boundaries are not discrete biological “races”

Greater variation within than between broad human populations.

Slide 24 · Gene flow makes human variation continuous

Migration and mixing

Gene flow among populations

Overlapping variation

No discrete racial genotype bins

Genetic variation does not respect rigid social categories.

Geographic ancestry can still associate with allele frequencies.

Slide 25 · A continuum is not a set of genetic boxes

Within group A

Substantial diversity

Within group B

Substantial diversity

Between groups

Overlapping genetic distributions

Across geography

Continuous variation with gene flow

Conceptual schematic: Groups are social labels, not disjoint genetic classes.

Slide 26 · Modern humans do not divide into biological races

Robbins rejects the concept of genetically discrete modern human races.

The conclusion follows from within-group diversity and gene flow.

Socially defined race is not a standalone inherited risk factor.

Ancestry-associated variants must be assessed without rigid exclusions.

Slide 27 · Socially defined race remains consequential for health

Social classification

Social labels can be used to create group hierarchies

Social subordination is historically documented

Groups may face discriminatory treatment

Health consequences

Living environments can differ

Access to timely care can differ

Disease burden and mortality can differ

A social construction can produce biologically important exposures.

Slide 28 · Classification by phenotype or origin has limitations

Skin pigmentation is one historical basis for race labels.

Geographic-origin labels are also socially used.

Terms such as “Caucasian” are identified as outdated.

None provides a precise substitute for an individual’s genotype.

Slide 29 · Race, ethnicity, ancestry: Different constructs

Construct

What Robbins emphasizes

Common error

Socially defined race

Classification by society and history

Treating label as genetic category

Ethnicity

Culture, origin, language, religion

Collapsing distinct communities

Geographic ancestry

Origin associated with some allele frequencies

Assuming an absolute disease boundary

Slide 30 · An environmental example: Residence near pollution

Residential location

Major transport / industrial facilities

Higher air pollution exposure

Greater asthma burden

Higher proximity/exposure for African Americans compared with European Americans.

This is an environmental pathway, not a claim of racial genetic susceptibility.

Slide 31 · Which physical settings does Robbins list?

Major roadways

Traffic-related air exposure

Rail lines

Transport corridor exposure

Ports

Freight and activity zones

Industrial facilities

Industrial air emissions

The common principle is unequal proximity to pollution sources.

Slide 32 · Asthma burden: An exposure pathway

Textbook example

Residential proximity to pollution sources differs

Exposure opportunity is elevated in affected locations

The text describes an increased asthma burden

Clinical reasoning

Race label alone does not reveal an exposure dose

Do not infer a genetic asthma mechanism

Consider the patient’s actual surroundings

Exposure assessment offers a more actionable explanation than a race label.

Slide 33 · Flint and Jackson: Failures of safe-water access

Flint, Michigan, and Jackson, Mississippi.

Both examples concern major water crises in majority African American cities.

Safe drinking-water access was compromised.

Residents were exposed to hazardous environmental contaminants.

Slide 34 · Water crises as a disparity mechanism

Community infrastructure

Unsafe or unreliable water

Environmental contaminant exposure

Potential health harms

The chapter establishes exposure and compromised access.

It does not quantify disease incidence attributable to each water crisis.

Slide 35 · Environmental justice is clinically relevant

Unequal exposure may arise from where people live and work.

Environmental histories can identify mechanisms hidden by group labels.

Discriminatory structures can affect infrastructure and care access.

These connections with pollution and water examples.

Slide 36 · Maternal mortality: An inequality reported

The text reports approximately three-fold higher maternal mortality among African Americans.

Comparison group: Americans of European descent.

Inadequate health care access in the explanation.

Treat this as a textbook-reported estimate, not a current annual surveillance rate.

Slide 37 · Infant mortality: A separate reported disparity

Approximately two-fold higher infant mortality among African Americans.

The comparator is Americans of European descent.

Infant and maternal mortality are distinct clinical endpoints.

Clinical access is one important context named by the source.

Slide 38 · COVID-19 mortality: Disparities in a pandemic

Approximately two-fold higher COVID-19 mortality among African Americans.

The reported comparison is with Americans of European descent.

The paragraph links outcome inequalities partly to care access.

The ratio is historical textbook content, not a claim of present-day rates.

Slide 39 · Three outcomes: Approximate ratios in the text

Outcome

Reported African American / European American rate

Important limitation

Maternal mortality

≈ 3-fold

Not dated in this excerpt

Infant mortality

≈ 2-fold

Not a subgroup-specific estimate

COVID-19 mortality

≈ 2-fold

Not a current live estimate

Slide 40 · Chronic diseases also show uneven prevalence

Type 2 diabetes

More prevalent in the African American population

Hypertension

More prevalent in the African American population

Obesity

More prevalent in the African American population

Clinical inference

Investigate multiple interacting determinants

The source reports direction, not absolute or relative prevalence estimates.

Slide 41 · Prevalence is not an explanation

What is reported

Type 2 diabetes, hypertension, and obesity

Higher prevalence is a group-level observation

The excerpt supplies no numerical mechanism-specific estimates

What it does not prove

Avoid assuming an inherent racial biology

Evaluate environment, access, and social context

Approach each patient individually

Descriptive epidemiology and causal attribution are different tasks.

Slide 42 · Out-of-hospital cardiac arrest exposes another gap

Cardiac arrests in public locations.

African American and Hispanic persons are less likely to receive bystander CPR.

The comparator is persons of European descent.

The disparity lies outside a traditional hospital care encounter.

Slide 43 · Bystander CPR: Point-of-event inequality

Public cardiac arrest

Bystanders present

Decision / provision of CPR

Opportunity for lifesaving care

Unequal provision at the bystander-CPR stage.

No specific explanation for the bystander behavior is established in the excerpt.

Slide 44 · Diagnostic algorithms may encode disparity

Patient clinical data

Race-based calculation

Estimated kidney function

Diagnosis / care decision

Race adjustment may change estimated GFR and classification.

A mathematical correction is not automatically biologically justified.

Slide 45 · EGFR: Race correction changes a clinical threshold

With a race correction

A race-based term modifies estimated GFR

The estimate contributes to chronic kidney disease diagnosis

A different reported estimate may alter classification

Without a race correction

Removal of race-based corrections

More African American patients would be identified with CKD

Earlier diagnosis and specialty access may follow

The text does not specify an equation version or diagnostic cut-off.

Slide 46 · The direction of the eGFR effect

Removing race correction is described as increasing CKD diagnoses among African American patients.

Earlier diagnosis creates an opportunity for timely specialty care.

Robbins positions this as a health equity implication of clinical testing.

No precision, sensitivity, or outcome-effect data are provided here.

Slide 47 · Review of race-based decision algorithms

Cardiovascular

Cardiology

Maternal / metabolic

Obstetrics · endocrinology

Cancer / respiratory

Oncology · pulmonology

Procedural systems

Urology · surgery · transplantation

Race-based components across these fields are being reviewed.

Slide 48 · Clinical algorithm audit: Four questions

Input

Is race entered as an adjustment?

Justification

What causal or measurement rationale exists?

Effect

Who crosses diagnostic or referral thresholds?

Equity

Could the adjustment amplify existing inequity?

Illustrative audit framework derived, not a cited formal guideline.

Slide 49 · Ethnicity includes more than a single descriptor

Culture

Shared practices and customs

Country of origin

Geographic and historical links

Language

Communication background

Religion

One possible identity dimension

Robbins treats ethnicity as multidimensional and relevant to outcomes.

Slide 50 · “Hispanic” and “non-Hispanic” are U.S. Categories

The text describes a common U.S. Ethnic classification as Hispanic versus non-Hispanic.

In U.S. Census/research usage, “Hispanic” refers to Spanish origin or descent.

This category is not equivalent to a shared genetic ancestry.

Administrative labels require definition before comparing health outcomes.

Slide 51 · Latin American is broader than Spanish-speaking

Hispanic

Hispanic is framed around Spanish origin or descent

It is often used as a U.S. Census / research category

Latin American

Latin American includes Central and South America

Non-Spanish-speaking Brazilians are included

Ancestry can include African, Amerindian, and European contributions

Neither label describes a uniform genomic group.

Slide 52 · Do not substitute phenotype for ancestry

Social race labels have used skin pigmentation and geographic terms.

Human populations have experienced extensive gene flow.

Population allele frequencies may correlate with regional ancestry.

An individual diagnostic decision should not depend on appearance alone.

Slide 53 · Ancestry-linked risk ≠ race-restricted disease

Appropriate population inference

Allele frequencies can differ by ancestral region

A disease may be more common in one broad ancestry group

Such associations may inform prior probability

Inappropriate patient inference

Disease is not exclusive to one racial label

Individual patients have mixed and diverse ancestries

Do not exclude disease solely on a group label

The distinction with cystic fibrosis and sickle cell disease.

Slide 54 · Cystic fibrosis: All cited groups are affected

Population descriptor in Robbins

Approximate birth frequency reported

Interpretation

Northern European descent

1 in 2,500

Higher frequency in example

African American infants

1 in 17,000

Disease still occurs

Asian American infants

1 in 31,000

Disease still occurs

Slide 55 · Cystic fibrosis does not respect a racial boundary

Ancestry-associated frequency

Different prior likelihoods

All groups retain possible cases

Patient-centered evaluation

The cited figures demonstrate frequency variation, not exclusivity.

Avoid using ancestry labels as the sole diagnostic rule-out.

Slide 56 · Sickle cell disease and historical selection

Falciparum malaria pressure

Protective advantage of sickle trait

Regional allele frequency patterns

Disease across several ancestries

The malaria-selection explanation.

The geographic distribution is broader than West Africa alone.

Slide 57 · The sickle mutation spans the malaria belt

Equatorial Africa

Higher-frequency region

Southern Europe

Higher-frequency region

Middle East

Higher-frequency region

Parts of Asia

Higher-frequency region

These are examples of regions named by Robbins—not prevalence measurements.

Slide 58 · A patient’s identity is not a diagnostic exclusion

Ancestry-associated conditions may occur outside commonly associated groups.

Distinguish geographic allele-frequency patterns from social race labels.

Keep phenotype, family history, and appropriate testing conceptually separate.

Robbins explicitly warns against biasing the approach to individual patients.

Slide 59 · Applied case: Asthma and the built environment

HYPOTHETICAL CASE

A patient with recurrent asthma symptoms lives near an industrial transport corridor. A group-level report shows greater pollution exposure in similar neighborhoods.

Interpretive questions

Which exposure pathway is plausible from the Robbins example?

What environmental history would help characterize this patient?

Why would a race label alone be a weak causal explanation?

Slide 60 · Case discussion: Follow the exposure pathway

Residence near corridor

Potential ambient air exposure

Asthma burden / symptoms

Individual evaluation

Environmental history tests a plausible mechanism.

Group associations cannot substitute for patient-level assessment.

Slide 61 · Applied case: Interpreting kidney function

HYPOTHETICAL CASE

Two reports use different eGFR approaches for the same hypothetical patient; One includes a race adjustment and the other does not.

Interpretive questions

Which reported value could affect CKD classification?

What downstream access-to-care consequence does Robbins identify?

Why should the equation’s adjustment be critically examined?

Slide 62 · Case discussion: Why an equation matters

Calculation rule

Reported eGFR

Disease classification

Specialty care access

Removing race correction with additional CKD diagnoses.

Changes in classification create opportunities for earlier referral.

Slide 63 · Applied case: Ancestry and cystic fibrosis

HYPOTHETICAL CASE

A clinician considers cystic fibrosis but is tempted to dismiss it because a patient is not described as having Northern European ancestry.

Interpretive questions

What do the three birth-frequency examples demonstrate?

Why does a lower population frequency not imply impossibility?

How can anchoring on a social group label bias care?

Slide 64 · Case discussion: Probability is not impossibility

Population information

Different population birth frequencies

Northern European ancestry has the highest cited frequency

A group association can shape population expectations

Individual care

Affected infants occur in each cited group

Labels cannot establish individual genotype

Avoid withholding evaluation due to an assumed exclusion

The correct lesson is graded prior likelihood—not an absolute racial boundary.

Slide 65 · How to read Robbins Figure 9.1 responsibly

Figure 9.1 concerns county-level differences in life expectancy by racial/ethnic group.

Comparisons are made relative to a White population in U.S. Counties for 2019.

County and group must both be considered when interpreting the displays.

The source figure itself is not reproduced: No values or map patterns are inferred.

Slide 66 · Integrating observation, pathway, and response

Document an unequal outcome

Investigate determinants

Identify plausible pathways

Avoid biased clinical rules

An observed disparity is the beginning of explanation, not its endpoint.

Source examples span environmental exposures, access, and algorithms.

Slide 67 · Five claims that should remain distinct

Health disparities concern unequal disease outcomes across groups.

Intersecting social, environmental, policy, and health care factors matter.

Socially defined race does not establish genetically discrete human groups.

Clinical algorithms and bystander actions can amplify unequal access.

Ancestry-linked disease associations are probabilistic, not absolute.

Slide 68 · Primary source and figure provenance

Robbins, Cotran & Kumar. Pathologic Basis of Disease. 11th ed. (2026). “Health Disparities.”

Dwyer-Lindgren K, et al. Life expectancy by county, race, and ethnicity in the USA, 2000–19: a systematic analysis of health disparities. Lancet. 2022;400(10345):25–38. (Credited in Robbins Fig. 9.1 caption.)

Figure 9.1 was not copied; only its supplied textbook caption informed the lecture.