US Health Statistics: Mortality, Morbidity, and Trends

The United States collects health data at a scale that would have seemed implausible a century ago — deaths coded to 70,000+ diagnostic categories, disease surveillance running through all 50 states, and chronic condition registries that track millions of people over decades. This page covers the major frameworks through which that data is organized, what drives the patterns it reveals, where the numbers get contested, and what the most significant current trends actually show.

Definition and scope

Mortality refers to death rates within a defined population — the simplest, hardest endpoint in health statistics, because the outcome is unambiguous. Morbidity is broader and messier: it captures illness, injury, disability, and disease burden, categories that require diagnosis, reporting infrastructure, and definitional consensus to count at all.

Together, these two domains form the backbone of US population health surveillance. The Centers for Disease Control and Prevention (CDC) coordinates the primary national surveillance systems, including the National Vital Statistics System (NVSS), which processes roughly 3.4 million death certificates annually (CDC NVSS). The National Center for Health Statistics (NCHS), a division of the CDC, produces the authoritative annual reports that anchor most national comparisons.

Health trends, the third element, describe how mortality and morbidity rates change across time — whether a condition is rising, falling, plateauing, or shifting across demographic subgroups. Trends are the interpretive layer: raw counts mean little without a denominator, a time series, and an age-adjustment calculation that controls for the fact that an aging population will generate more deaths even if individual risk is falling.

The scope of US health statistics extends from national aggregates down to county-level data in systems like the CDC's PLACES database, which estimates prevalence for 36 chronic disease measures across all US counties. The Health and Human Services (HHS) framework, particularly the Healthy People initiative, sets 10-year measurable targets that give trend data a benchmark to measure against.

Core mechanics or structure

US health statistics flow through three interlocking collection systems, each with a different methodology and time lag.

Vital statistics come from birth and death certificates filed at the state level and compiled federally by the NCHS. Death certificates require a cause-of-death determination — often a chain of conditions — coded using the International Classification of Diseases, 10th Revision (ICD-10). The accuracy of this system depends on physician coding practices, which vary.

Survey-based surveillance uses probability samples to estimate conditions that wouldn't otherwise be systematically reported. The National Health Interview Survey (NHIS), conducted by the NCHS, is a continuous household survey that has tracked self-reported health conditions since 1957. The Behavioral Risk Factor Surveillance System (BRFSS), run by state health departments in partnership with the CDC, is the world's largest telephone health survey — covering more than 400,000 respondents annually (CDC BRFSS).

Registry and administrative data include disease-specific registries like the CDC's National Program of Cancer Registries and the National Cancer Institute's SEER program, plus insurance claims data processed through Medicare and Medicaid. Administrative data have the advantage of scale but reflect only insured, treated populations — a structural selection bias.

Age-adjustment is the statistical correction that makes trend comparisons meaningful. Crude death rates simply count deaths per 100,000 population. Age-adjusted rates apply a standard age distribution (the US 2000 Standard Population is the current reference) to remove the effect of demographic aging. The age-adjusted rate is the number almost always cited when comparing mortality across states, racial groups, or decades.

The broader framework of health metrics describes how these systems are evaluated for reliability and policy relevance.

Causal relationships or drivers

Behind every mortality and morbidity trend is a web of causes that resist simple attribution. The social determinants of health — income, education, housing, neighborhood environment — account for an estimated 30–55% of health outcomes according to the World Health Organization, a range that itself reflects methodological variation across studies.

At the disease level, cardiovascular disease remains the leading cause of death in the US, accounting for approximately 695,000 deaths in 2021 (CDC, Leading Causes of Death 2021). Cancer follows at roughly 605,000 deaths in the same year. Together, these two conditions account for just under half of all US deaths — a concentration that shapes where surveillance resources flow.

Behavioral drivers — tobacco use, physical inactivity, excess alcohol consumption, poor diet — operate upstream of both cardiovascular disease and cancer. The tobacco and health connection alone is estimated by the CDC to cause more than 480,000 deaths annually in the US, including deaths from secondhand smoke exposure (CDC, Smoking and Tobacco Use).

The opioid crisis represents a more recent and sharply accelerating driver of mortality. Drug overdose deaths surpassed 100,000 for the first time in the 12-month period ending April 2021, with synthetic opioids — primarily illicitly manufactured fentanyl — driving the increase (CDC, Drug Overdose Data). This signal moved overdose deaths into the top 5 causes of death for adults under 65, a category-level shift that took statisticians and policymakers somewhat by surprise.

Classification boundaries

The boundary between mortality and morbidity data is sharper than it might seem. Mortality data capture a definitive event with legal documentation. Morbidity data capture a condition, which may go undiagnosed, underreported, or differently defined across clinical settings.

Disability-adjusted life years (DALYs) and quality-adjusted life years (QALYs) attempt to bridge this gap by combining mortality (years of life lost) with morbidity (years lived with disability). The Global Burden of Disease study, coordinated by the Institute for Health Metrics and Evaluation (IHME), uses DALYs as its primary metric — which produces a different ranking of health burdens than crude mortality statistics alone. Mental health conditions, for example, rank far higher in DALY analyses than in mortality statistics, because they cause enormous disability without always appearing on death certificates.

Surveillance case definitions — the diagnostic criteria that determine what counts as a reportable condition — are set by the Council of State and Territorial Epidemiologists (CSTE) in collaboration with the CDC. A change in case definition can create an apparent trend that is entirely definitional. The 2012 expansion of the Lyme disease surveillance case definition, for instance, contributed to apparent increases in reported cases that partly reflected broadened criteria rather than pure disease expansion.

Tradeoffs and tensions

The decision to age-adjust death rates resolves one problem and creates another. Age-adjustment makes temporal and geographic comparisons fair — but it obscures the raw burden on a community. An aging rural county with a low age-adjusted death rate may still have hospitals overwhelmed by the sheer volume of elderly patients dying from chronic disease. Both numbers are true; they answer different questions.

Race and ethnicity classification in vital statistics data is another live tension. Death certificates record race based on information provided by funeral directors, who may observe or infer race rather than record self-identification. This introduces systematic misclassification — particularly for American Indian and Alaska Native populations, where undercounting in mortality data has been documented (NCHS, Racial and Ethnic Disparities in Mortality).

The health equity implications are direct: if deaths in a population are systematically undercounted, that population's health burden is understated, and resource allocation follows the incomplete signal rather than the real one.

Privacy protections — particularly HIPAA's constraints on data sharing — also create friction. Richer individual-level data would improve causal inference, but linkage across datasets (vital records, insurance claims, electronic health records) requires navigating disclosure rules that legitimately protect patients. The result is a surveillance system that is wide but sometimes surprisingly shallow.

Common misconceptions

Misconception: The US has the highest life expectancy among developed nations.
The US life expectancy at birth was 76.4 years in 2021 — a figure that placed it below 40 other countries (NCHS, United States Life Tables 2021). This is a persistent source of confusion, partly because the US performs well on disease-specific survival rates (notably cancer five-year survival) while performing poorly on population-level longevity indicators that capture the full distribution of health outcomes.

Misconception: Higher healthcare spending produces better mortality outcomes.
The US spends approximately $12,500 per person annually on healthcare — more than any other high-income nation (CMS, National Health Expenditure Data) — yet ranks near the bottom of OECD nations on preventable mortality and amenable mortality measures (OECD Health Statistics). Spending and outcomes are not synonymous, a gap explained partly by care access disparities, administrative overhead, and variation in preventive health utilization.

Misconception: Infectious diseases are no longer significant drivers of US mortality.
Before 2020, pneumonia and influenza combined typically ranked among the top 10 causes of US death. COVID-19 became the third-leading cause of death in 2020 and 2021, demonstrating that infectious disease mortality risk never disappeared — it was simply suppressed during a period of relative epidemiological stability.

Checklist or steps

How US health statistics flow from event to published estimate:

  1. A vital event (birth, death) or health encounter occurs and is documented at the point of care.
  2. State vital records offices collect and process certificates; disease registries receive mandated reports from providers.
  3. State data are transmitted to the CDC/NCHS under federally standardized formats and coding protocols (ICD-10 for causes of death).
  4. NCHS applies quality checks, reconciles inconsistencies, and computes crude and age-adjusted rates using the 2000 US Standard Population.
  5. Preliminary national estimates are released (typically within 12 months of the reference year); final data follow 18–24 months after the reference period.
  6. NCHS publishes data in the National Vital Statistics Reports series, accessible through the CDC NCHS website.
  7. Researchers, state health departments, and federal agencies access disaggregated data through tools like CDC WONDER for stratified analysis by age, sex, race, and geography.
  8. Healthy People targets are compared against current-year estimates to assess progress; trend analyses are updated annually.

Reference table or matrix

Key US Health Statistics Data Sources

System Operator Primary Metric Update Frequency Geographic Resolution
National Vital Statistics System (NVSS) CDC/NCHS Mortality, birth rates Annual (preliminary + final) National → county
National Health Interview Survey (NHIS) CDC/NCHS Self-reported conditions, health behaviors Annual National, select state
Behavioral Risk Factor Surveillance System (BRFSS) CDC + state HDs Behavioral risk factors, chronic conditions Annual State → metro
National Cancer Registry (NPCR + SEER) CDC + NCI Cancer incidence, survival Annual State → county
CDC PLACES CDC Chronic disease prevalence estimates Annual County → census tract
National Health and Nutrition Examination Survey (NHANES) CDC/NCHS Biomarkers, clinical measures Continuous 2-year cycles National
Medicare Claims Data CMS Utilization, diagnoses in 65+ population Ongoing National → county

The human health reference hub provides entry points across the full spectrum of health topics documented through these systems.

For context on how individual risk factors feed into aggregate statistics, the health risk factors and chronic disease overview pages describe the pathways that make statistical trends clinically legible.

References