The American Psychological Association released its annual Stress in America survey in November. The sample was 3,199 adults, nationally representative, conducted over three weeks in late summer. Politics ranked second on the list of significant stressors — cited by 65 percent of respondents. Only the economy ranked higher.

That number is worth sitting with. Not because it is surprising, but because of what it has become over time. In 2016, the presidential election caused significant stress for 52 percent of adults. The 2025 survey was not conducted during an election. It was conducted in August, between cycles, in a relatively quiet news month.

The stress did not recede. It became the baseline.

The APA survey’s most striking finding is not the political stress number. It is what that stress is doing to the body. Among adults who reported high levels of loneliness — itself strongly linked to perceived societal division — 94 percent reported at least one physical symptom of stress: headaches, fatigue, muscle tension. Eighty percent of adults with high loneliness reported chronic illness. These are not self-reports of unhappiness. They are reports of physiological consequence.

The human stress response is not designed for abstract disagreement. It is designed for perceived threat. The body does not cleanly distinguish between physical danger and persistent social threat. When the informational environment continuously signals that the world is divided into the trustworthy and the dangerous, the body responds accordingly — elevated cortisol, disrupted sleep, heightened vigilance, reduced capacity for deliberation. A 2022 study published in PLOS ONE tracked these effects across the first Trump administration and found that political stress exacted measurable health costs from large numbers of American adults between 2017 and 2020, with physical health deterioration detectably worse in the aftermath of the 2020 election.

The mechanism is chronic, not acute. A short, bounded stressor resolves. A low-grade alarm running continuously does not. The system never clears.

What produced this condition is not difficult to trace. Modern political operation has shifted away from persuasion and toward activation. The goal is not to move the undecided. It is to keep the already-mobilized in a sustained state of moral emergency.

That calculation was put into writing in October 1971. Pat Buchanan, then a senior aide and speechwriter for President Nixon, sent a strategy memo advising that the administration should “cut the Democratic Party and country in half” — and that, in his confident assessment, “we would have far the larger half.” The strategy was not to persuade. It was to polarize deliberately, take the larger fragment, and govern from division. Nixon won 49 states in 1972. The logic proved correct, and it was noted.

Polling and behavioral modeling identify the emotional levers — resentment, fear, status threat, tribal loyalty — and messaging is optimized against them. Digital platforms accelerate the process by rewarding content that produces reaction and suppressing content that produces reflection. The optimization function does not ask whether a message is accurate or stabilizing. It asks whether it generates engagement.

Engagement, for the human nervous system, means activation. And activation, sustained without resolution, is the definition of chronic stress.

What makes the current moment distinct is not the existence of political conflict. It is the geographic expansion of that conflict into nearly every domain of ordinary life.

Medicine became a loyalty test during the pandemic years and has not recovered. The professional societies that issue clinical guidance now carry political valence — their recommendations filtered through identity before they reach patients or providers. Food has been sorted: organic purchasing patterns align with political identity at rates that would have seemed implausible twenty years ago. Parenting decisions, educational choices, entertainment consumption, word selection in professional communication — each has been assigned to a faction, and each carries an implicit cost for those who fail the sorting test.

The process follows a consistent pattern. Identify a domain with wide exposure. Attach moral or existential stakes to opposing positions. Assign those positions to opposing groups. Amplify until neutrality becomes socially untenable. Once that threshold is crossed, the original subject no longer matters. It has become an instrument for sorting.

A society can absorb genuine disagreement. It cannot indefinitely absorb identity sorting conducted across the full surface of daily life.

The physiological cost distributes broadly, but it does not require active political engagement. The APA data make this clear: the stress is ambient. Scrolling a feed, navigating a workplace that has incorporated political signals into its culture, sitting in a waiting room with a television tuned to news — exposure is sufficient. Participation is not required.

The accumulating consequence is documented. Chronic low-level stress is associated with impaired immune function, metabolic disruption, cardiovascular strain, sleep degradation, and a narrowing of cognitive flexibility — specifically, the capacity to evaluate evidence and tolerate ambiguity. These outcomes are not political. They are biological. And they fall on people who never sought any of this, who would prefer to eat dinner without a signal check and take their children to the doctor without navigating a loyalty matrix.

The incentive structure that sustains this environment is not hidden. Political organizations benefit from sustained mobilization. Media organizations benefit from engagement. Platforms benefit from time on device. Advocacy organizations benefit from urgency. Each of these actors absorbs the gain. The individual absorbs the cost — in sleep, in relationships, in physiological wear, in the slow erosion of the cognitive tools needed to make sound decisions about health, money, and risk.

That asymmetry does not self-correct. It is stable precisely because the costs are diffuse and the benefits are concentrated.

Partial withdrawal helps at the margins. Reducing screen time lowers exposure. But it does not address the structural condition. Many institutions now reproduce the same dynamics regardless of whether any individual is consuming political media. The ambient condition persists.

What would actually interrupt the pattern is a reconstruction of neutral territory — domains where people can exist, interact, and make decisions without continuous implicit demands for allegiance. Some of that work is individual: treating local reality as more authoritative than national narrative, maintaining relationships that cross identity lines, resisting the reflex to convert every preference into a signal. Some of it is institutional: reducing the incentive alignment between outrage-based engagement and revenue, restoring epistemic standards that distinguish evidence from interpretation, recognizing that the current condition carries costs that are not being assigned to those who produce them.

The APA titled its 2025 report A Crisis of Connection. The framing is accurate, but it understates the mechanism. Connection is not being lost by accident or neglect. It is being displaced by a system optimized to keep people activated, separated, and attentive to threat.

That system is producing a measurable health outcome. The 2025 survey data, the longitudinal tracking from 2017 through 2020, the clinical reporting from practitioners who describe politics as the unavoidable backdrop to nearly every session — these are not separate observations. They are a pattern.

A population under permanent mobilization is not a healthy population. Recognizing that cost — naming it as a health question and not merely a political one — is where a serious response begins.