US Health Statistics at a Glance: Key Data and Trends
The United States generates more health data than almost any nation on earth — and yet making sense of it requires knowing which numbers actually matter and why. This page draws together key statistics on life expectancy, chronic disease burden, health behaviors, and healthcare utilization to give a grounded picture of where American health stands. The figures come from named federal sources including the CDC, NCHS, and CMS, and are placed in context so they function as reference points rather than raw noise.
Definition and scope
Health statistics are the quantified backbone of public health in the US. They are collected, standardized, and published by agencies including the Centers for Disease Control and Prevention (CDC), the National Center for Health Statistics (NCHS), the Agency for Healthcare Research and Quality (AHRQ), and the Centers for Medicare & Medicaid Services (CMS). The scope spans mortality, morbidity, health behaviors, access to care, and healthcare expenditure — each category answering a different kind of question.
The distinction between mortality data and morbidity data is foundational. Mortality statistics count deaths and their causes. Morbidity statistics count illness, disability, and disease prevalence in the living population. A disease can be responsible for enormous morbidity — meaning it affects millions of people's daily functioning — while ranking lower on mortality tables. Diabetes is a precise example: the CDC reports that 38.4 million Americans were living with diabetes as of 2021 (CDC National Diabetes Statistics Report), making it a dominant driver of morbidity even as it rarely appears as the singular cause of death on a death certificate.
How it works
Federal health data flows through a layered infrastructure. The NCHS, a division of the CDC, administers the National Vital Statistics System — the mechanism through which birth and death records from all 50 states are aggregated into national figures. Separately, the National Health Interview Survey (NHIS) and the Behavioral Risk Factor Surveillance System (BRFSS) collect self-reported data on health behaviors and conditions through ongoing household and telephone surveys.
A critical distinction in how these health metrics and indicators are produced:
- Administrative data — derived from insurance claims, hospital records, and billing codes. High volume, but shaped by what gets coded and billed.
- Survey data — drawn from sampled populations using validated questionnaires. Captures behaviors and conditions that may never appear in a clinical record.
- Vital statistics — legally mandated records of births, deaths, and causes of death. Near-complete coverage but subject to coding variation across jurisdictions.
- Surveillance data — condition-specific tracking systems (e.g., cancer registries, HIV surveillance) designed for ongoing monitoring of particular health threats.
Each data type has blind spots. Survey data misses people who are institutionalized or unhoused. Administrative data reflects access to care as much as disease prevalence — a population with low insurance coverage will generate fewer claims even if illness rates are high.
Common scenarios
The major headline figures from federal sources illuminate patterns that repeat across health research and evidence discussions.
Life expectancy dropped to 76.4 years in 2021, the second consecutive year of decline, driven substantially by COVID-19 and drug overdose deaths (NCHS Data Brief No. 456, 2022). That figure sits below the average for peer high-income nations, a gap that has persisted for decades and reflects determinants of health well beyond the clinical system.
Chronic disease burden is the dominant feature of contemporary US morbidity. The CDC estimates that 6 in 10 American adults have at least one chronic disease, and 4 in 10 have two or more (CDC Chronic Disease Overview). Cardiovascular disease remains the leading cause of death, accounting for approximately 1 in every 5 deaths. Respiratory health conditions, including COPD, rank among the top ten causes of mortality.
Healthcare expenditure reached $4.5 trillion in 2022, representing 17.3% of US GDP (CMS National Health Expenditure Data). That proportion is the highest of any wealthy nation by a considerable margin — roughly double the OECD average as a share of GDP — which makes the life expectancy gap particularly striking as a policy reference point.
Mental health data tells its own story. The National Institute of Mental Health estimates that 1 in 5 US adults — approximately 57.8 million people — lived with a mental illness in 2021 (NIMH Mental Illness Data). The gap between prevalence and treatment remains wide; mental health overview resources document the structural barriers that sustain it.
Decision boundaries
Not all statistics carry equal weight for decision-making, and understanding which figure applies to which question prevents common misreadings.
Population-level vs. individual-level data. A statistic showing that 14.5% of US adults smoke cigarettes (CDC BRFSS 2022) describes aggregate prevalence — it does not predict any individual's trajectory. Population statistics inform preventive health policy; they are not clinical risk calculators.
Averages vs. distributions. National averages routinely obscure the deep variation visible when data is broken down by race, income, geography, or age. The national life expectancy figure of 76.4 years masks a gap exceeding 6 years between the highest- and lowest-performing counties in the US, as documented in County Health Rankings data maintained by the University of Wisconsin Population Health Institute. That dispersion is precisely where health equity analysis begins.
Trend vs. snapshot. A single-year figure is a snapshot; the trend line is the story. Overdose mortality, for instance, rose more than 30% between 2019 and 2020 alone (CDC Drug Overdose Surveillance) — a rate of change that a single-year number cannot communicate. Reading statistics across time, against named benchmarks, and stratified by subgroup is what separates informed interpretation from headline-skimming.