Seven of Arizona’s fifteen counties currently show two or more major health-risk areas running worse than most of the country at the same time. As a share of the state’s counties, that is 46.7%, the seventh-highest rate in the nation and roughly double the national rate of 22.9%. The ranking is the headline most people would write. The more useful finding is where those seven counties are, and where they are not.
They are not Phoenix and they are not Tucson. Maricopa and Pima counties, home to roughly three of every four Arizonans, do not carry the flag. The seven counties that do (Gila, Mohave, La Paz, Apache, Navajo, Santa Cruz, and Coconino) hold about 662,000 people between them, just under 9% of the state’s 7.4 million residents. Arizona’s converging health risks concentrate almost exactly where its population does not.
For anyone preparing a community health needs assessment in Arizona, that inversion matters. The state’s CHNA conversation is naturally anchored to the metro systems in Phoenix, Tucson, and Flagstaff, because that is where the hospitals and the analysts are. The counties where risks actually pile up are served mostly by critical access hospitals, rural health clinics, Indian Health Service and tribally operated facilities, and small county health departments. Those organizations carry the same federal CHNA obligations with a fraction of the analytical staff.
What the flag measures, applied to Arizona
Banana Analytics scores every U.S. county from 0 to 100 on four health-risk areas: environmental risk, disease burden, provider access, and social and economic conditions. Each score is a national standing, so a 70 means the county sits worse than at least 70% of U.S. counties on that dimension. When two or more areas cross that line at once, the county carries a converging-risk flag. Nationally, 738 of 3,222 counties (22.9%) currently do.
Arizona has seven of fifteen. Two are in the stronger tier, with three of the four areas above the line at once: Gila County and Mohave County. Five carry two areas above the line: La Paz, Apache, Navajo, Santa Cruz, and Coconino. By raw count Arizona ranks 21st among states; by share of counties it ranks 7th, behind the Deep South states and West Virginia that we have written about before.
In the platform’s methodology this pattern is called Multi-Pillar Convergence: two or more of the four pillar scores at or above the 70th national percentile, with tiers at 2 of 4 (Moderate), 3 of 4 (Strong), and 4 of 4 (Extreme). The full scoring documentation is at banana-analytics.com/methodology.
Two caveats before the tour, because Arizona stresses the limits of county-level analysis more than almost any other state.
First, Arizona’s counties are enormous. Coconino County is the second-largest county by land area in the contiguous United States, bigger than nine individual states. Maricopa County alone, with 4.6 million people, would be the 14th-largest state by population if it were one. A single county-level score averages across distances and communities that would be separate regions anywhere east of the Mississippi. Where that averaging changes the story, we say so below.
Second, the disease prevalence figures come from CDC PLACES, which produces model-based small-area estimates rather than direct counts, and the mortality rates from CDC WONDER are crude rates, not age-adjusted. Both points matter for specific claims later in this article, and we flag them where they do.
Environmental exposure is nearly statewide. Convergence is not.
Arizona’s single most distinctive number in our dataset is its average environmental risk score: 67.3, against a national county average of 31.6. No pattern in the state data is louder. Nine of the fifteen counties score above the 70-point line on environmental risk, including Maricopa (79.8), Pinal (82.6), Yavapai (84.8), Cochise (83.8), Yuma (76.0), and Santa Cruz (85.2). Extreme heat, drought, wildfire smoke, and dust-driven air quality land on nearly everyone in the state, metro and rural alike.
But exposure alone does not make risks converge. Five of those environmentally burdened counties (Maricopa, Pinal, Yavapai, Cochise, and Yuma) cross the line on nothing else, so they carry no flag. The seven flagged counties are the places where the exposure lands on top of something else: chronic disease rates that already run high, or social and economic conditions that are already strained. In three of the seven (La Paz, Apache, and Navajo), the county-level environmental score is actually unremarkable, and the convergence is entirely a disease-plus-social-conditions story.
This is the difference between a hazard map and a needs assessment. A hazard map of Arizona is orange almost everywhere. The needs concentrate where the hazard meets a population with less capacity to absorb it.
The driver that is missing: provider access
In the states we have profiled before, provider shortage is the engine of convergence. Mississippi’s and Texas’s flagged counties cross the provider-access line almost universally. Arizona is the opposite. Not one of the seven flagged counties crosses it, and the state’s average provider-access score is 36.2, comfortably better than the national average of 50.
Read that carefully, because it is partly real and partly an artifact of scale. The real part: Arizona’s care footprint is unusually consolidated, with large systems in Phoenix, Tucson, Flagstaff, and Yuma serving very wide catchments, and our provider scoring credits a county when its neighbors have supply residents can plausibly reach. The artifact part: a county-level score cannot see distance inside a county, and Arizona’s counties are the size of small states. A resident of Ganado in Apache County and a resident 90 minutes away contribute to the same county average. A county-grain “no provider gap” in northern Arizona should never be read as “no access problem.”
The workforce numbers that do resolve at county grain point the same direction. County Health Rankings counts residents per mental health provider, and the split between flagged and metro Arizona is stark: one provider per 1,756 residents in Santa Cruz County, 1,243 in Mohave, 1,114 in La Paz, and 1,080 in Gila, against 468 in Maricopa and 416 in Pima. Several of the flagged counties also post the state’s highest depression prevalence, which we return to below.
The river counties: Mohave and La Paz
The Colorado River corridor holds the two counties where Arizona’s convergence runs hottest, in both senses.
Mohave County (Kingman, Lake Havasu City, Bullhead City) is the largest flagged county at 223,682 people and one of the two in the stronger tier, with environmental risk (79.8), disease burden (75.7), and social and economic stress (83.3) all above the line at once. It is a retirement destination with the disease profile that implies: adult diabetes at 14.2% and COPD at 11.2%, both far above national rates, with adult smoking at 16.1% still doing structural damage.
La Paz County (Parker, Quartzsite) is the sharpest single profile in the state. Population 16,710, median household income about $48,800, and one of the oldest county populations in the United States; more than four in ten residents are 65 or older. Coronary heart disease prevalence is 13.3%, roughly double the national rate, alongside 17.5% diabetes and 12.5% COPD. Its social and economic conditions score worse than roughly 96% of U.S. counties. And its Heat Vulnerability score of 95.9 is among the highest in the nation, which matters more in the month this article is published than in any other.
The high country: Gila County, with Yavapai on watch
Gila County (Globe, Payson, and part of the San Carlos Apache Reservation) posts the highest overall need score in Arizona at 71.4, and it is the state’s other stronger-tier county: environmental risk 83.6, disease burden 77.9, and social and economic stress 94.4, all above the line simultaneously. Twenty-eight percent of its children live in poverty, adult diabetes runs 15.0%, and depression prevalence is 20.7%. The old copper-country economy never fully replaced itself, and the health data reads like it.
Neighboring Yavapai County (Prescott) is worth naming as the borderline case. Today it crosses the line only on environmental risk (84.8), so it carries no flag, but its disease burden (61.3) and social and economic stress (63.7) both sit within ten points of the threshold. Of Arizona’s unflagged counties, it is the one a small data shift could move.
The tribal northeast: Apache and Navajo counties
Apache County, where 71.4% of residents are American Indian or Alaska Native and which includes large portions of the Navajo Nation, posts a social and economic conditions score of 99.5. Read plainly: virtually every county in the United States has less strained social and economic conditions. Median household income is about $43,100, 36% of children live in poverty, adult diabetes prevalence is 17.5%, and adult asthma prevalence (14.3%) is the highest in Arizona. Adult smoking, at 22.4%, is also the state’s highest.
Navajo County next door (Holbrook, Winslow, Show Low, and parts of the Navajo, Hopi, and Fort Apache reservations) runs the same pattern slightly less extreme: social and economic stress of 98.5, diabetes at 15.1%, and the state’s second-highest depression prevalence at 22.2%.
None of this is news to the tribal health organizations, Indian Health Service units, and tribal epidemiology centers that have documented these patterns for decades, and this data does not substitute for their work. What county-level scoring adds is a defensible national benchmark: when a CHNA or a grant application needs to state that a service area’s social and economic conditions sit at the very top of the national distribution, a 99.5 score with published methodology is a citable way to do it.
The border: Santa Cruz County
Santa Cruz County (Nogales, 82.7% Hispanic) is one of two counties that joined the flagged list as our scoring refreshed this spring, and it converges on a different mix than the northeast: environmental risk of 85.2, the highest in the state, plus social and economic stress of 71.7. One in five residents is uninsured, the highest share in Arizona, and the county posts the state’s worst mental health workforce ratio at one provider per 1,756 residents.
What Santa Cruz does not have is a disease problem. Coronary heart disease prevalence (7.5%) and adult smoking (11.6%) are among the best in the state, and its crude mortality is well below the state average. The convergence here is exposure plus economics landing on a comparatively healthy population. The implementation strategies that follow (coverage navigation, environmental health, behavioral health capacity) look nothing like the chronic disease management story in Mohave or La Paz. Same flag, different assessment.
The Flagstaff paradox: Coconino County
Coconino County is the other spring addition and the one that best illustrates the county-grain caveat. Median household income is about $71,500, the highest of the seven flagged counties. Flagstaff is a university town with a regional medical center. And the county still converges: environmental risk of 75.8 plus social and economic stress of 75.3, with the state’s highest depression prevalence at 22.5%.
The resolution of the paradox is that Coconino County is not one place. It spans 18,600 square miles and includes parts of the Navajo Nation and the Havasupai Reservation alongside Flagstaff; about a quarter of residents are American Indian or Alaska Native. A county average that blends a college town with remote communities hours from a hospital understates both the stability of the former and the strain on the latter. For a CHNA covering northern Arizona, the county score is the starting flag, and the within-county split is the actual finding.
Six of the seven are heat-vulnerable, and it is July
We publish state analyses when they are seasonally relevant, and this is the Arizona season. Our Heat Vulnerability signal blends extreme heat exposure with the population factors that turn heat into hospitalizations: chronic cardiovascular and metabolic disease, and thin specialty care. Six of the seven flagged counties trigger it at high confidence: La Paz (95.9), Apache (92.5), Gila (90.6), Mohave (89.5), Navajo (87.1), and Santa Cruz (81.7). The exception is Coconino, at 56.6, where 7,000 feet of elevation does for Flagstaff what no cooling center can do for Parker.
The clinical mechanism, and the specialty-care shortfall that compounds it, is the subject of our national analysis of heat risk and cardiologist supply. The Arizona-specific point is narrower: the counties where Arizona’s risks converge are, with one explainable exception, also the counties where July and August do their worst work. Heat preparedness in these seven counties is not a checklist item in a CHNA. It is a primary finding.
Heat Vulnerability is one of eight named Compound Signals in the platform: extreme heat exposure × heat-vulnerable population. Signal scores and confidence levels appear on each county page and in the platform’s Compound Signals tab.
What mortality does and does not confirm
Across the seven flagged counties, 2024 all-cause mortality averages 1,484 deaths per 100,000 residents against 1,084 in the eight unflagged counties, a 37% gap. Heart disease mortality runs 36% higher and chronic lower respiratory disease mortality 43% higher. Those are crude rates from CDC WONDER, and we want to be direct about what they can and cannot say.
The gap is not uniform, and it is partly age structure. La Paz (2,178), Mohave (1,872), and Gila (1,745) carry some of the highest crude rates in the state in significant part because their populations are old. Meanwhile two flagged counties sit near the bottom of the state’s mortality table: Coconino (774) and Santa Cruz (818), whose populations are young and, in Santa Cruz’s case, notably healthy on behavior measures. The honest summary is that mortality strongly corroborates the convergence flag where the flag is driven by disease burden (the river counties, the high country, the tribal northeast), and it is the wrong validation instrument where the flag is driven by exposure and economics (the border, Flagstaff). A CHNA should cite these rates county-by-county, not as a single blended claim.
The behavioral health thread
One finding cuts across every group in this article. Depression prevalence in the flagged counties runs from 16.0% to 22.5%, and the three highest figures in the state (Coconino 22.5%, Navajo 22.2%, Gila 20.7%) are all in flagged counties. Set that against the workforce ratios cited earlier, where most flagged counties have one mental health provider for every 1,000 to 1,800 residents versus one per 400 to 500 in the metros, and the shape of the need is clear: the counties with the most measured distress have a half to a quarter of the metro workforce density.
For CHNA purposes this is the finding with the clearest implementation pathway. Telebehavioral health partnerships, integrated behavioral health in primary care and IHS facilities, and school-based services all have evidence bases, and all become more defensible in a needs assessment when the workforce gap is quantified against national benchmarks rather than asserted.
For practitioners
If you write CHNAs for Arizona hospital facilities, the grouping in this article is a usable organizing structure for a secondary data review: the river retirement corridor (chronic disease and heat), the central high country (economic decline plus disease), the tribal northeast (social and economic conditions at the top of the national distribution), the border (exposure and coverage, not disease), and the Flagstaff paradox (county averages hiding within-county splits). Five patterns, five different implementation conversations, inside one state with fifteen counties.
If you serve a critical access hospital, rural health clinic, FQHC, or tribal health program in one of the seven counties, the metro-anchored CHNA framing you will find in the state’s published assessments mostly does not describe your service area. The data above, and the free county pages behind every link in this article, are intended to be directly citable in your assessment with source attribution intact.
If you work in population health or community benefit at one of the metro systems: your service areas may not carry the flag, but Maricopa, Pinal, Yuma, and Yavapai all cross the environmental risk line, and the referral catchments of Phoenix and Tucson facilities reach deep into the seven flagged counties. The Compound Signals tab in the platform can score any service area you define, whether that is a county list, a ZIP list, or a drive-time boundary.
The numbers in this article come from the same production dataset behind every county page on this site, pulled July 8, 2026, and include CDC PLACES disease prevalence, Census ACS five-year social and economic measures, CDC WONDER 2024 mortality, County Health Rankings workforce ratios, and CMS NPPES provider data. Scores refresh with each pipeline release, so live pages may drift slightly from the figures here; where they do, the county page is the current number.