Cardiovascular Risk Factors Identified Among Homeless Adults in San Francisco: Recommendations for Evidence-Based Outreach Services From a Quality Improvement Initiative

A quality improvement initiative assessing unsheltered adults in San Francisco revealed a high prevalence of acute cardiovascular risks, particularly severe hypertension, underscoring the urgent need for targeted outreach, primary care referrals, and health education to mitigate future cardiovascular events.

The Gist

Imagine the city of San Francisco as a giant, bustling ship. Most of the passengers are in comfortable cabins, but there's a group living on the deck, exposed to the wind, rain, and sun. This report is like a medical check-up for those living on the deck, conducted by a crew member named Sarah and her team from the Valliant Foundation.

They didn't just ask, "How are you feeling?" They actually measured the "engine" of these people's bodies to see if they were about to break down. Here is what they found, translated into everyday language:

1. The "Pressure Cooker" Problem (Blood Pressure)

Think of your blood vessels like garden hoses. If you turn the water on too hard, the hose might burst.

• The Finding: The team found that 73% of the people they checked had their "water pressure" turned up way too high. This is called hypertension.
• The Danger Zone: Even scarier, 1 in 10 people had pressure so high it was in the "Hypertensive Crisis" zone. Imagine a hose that is about to snap at any second. Some of these readings were so extreme (like 286/127) that the team had to treat it like an emergency, offering to call 911 immediately.
• The Analogy: It's like driving a car with the engine redlining at 10,000 RPMs while the brakes are failing. Most people on the street didn't even know their "engine" was about to blow.

2. The "Stressed-Out Engine" (Heart Rate)

• The Finding: The average heart rate was 96 beats per minute. A normal resting heart is usually around 60–80.
• The Analogy: Imagine a car idling in neutral. A normal car idles quietly. These people's hearts were idling like they were stuck in traffic, revving high even when they were just sitting still. This suggests their bodies are constantly in "fight or flight" mode, likely due to the stress of not having a home, the cold, or the fear of the next day.

3. The "Heavy Backpack" (Chronic Pain)

• The Finding: Nearly 20% of the people were in severe pain (a 7 to 10 on a scale of 1 to 10).
• The Analogy: Imagine everyone is carrying a backpack. For most people, the backpack is light. For these individuals, it's filled with bricks. Some are carrying a backpack so heavy it's crushing their spine, yet they have to keep walking. This pain isn't just a nuisance; it keeps their hearts racing and makes their bodies tired.

4. The "Silent Saboteurs" (Smoking and Diabetes)

• The Finding: About 13% had diabetes, and 15% were using nicotine.
• The Analogy: If the high blood pressure is a fire, then smoking and diabetes are like pouring gasoline on it. They make the fire burn hotter and spread faster. The report notes that when you mix these together, the risk of a heart attack or stroke goes up dramatically.

5. Who is on the Deck?

The group they checked was mostly men (about 7 out of 10), with an average age of 43. Interestingly, the group was split almost evenly between White and Black individuals. However, the report highlights that Black individuals in San Francisco are overrepresented in the homeless population compared to the general city population, and they face extra barriers because of historical mistrust in the medical system.

What Should We Do? (The Recommendations)

The authors aren't just pointing out problems; they are handing out a repair manual.

1. Stop the Leaks: They need to get these people to doctors fast. If someone's pressure is in the "bursting" zone, they need a ride to the ER, not just a pamphlet.
2. The "Trusted Guide" Approach: Because many people don't trust doctors, the team suggests using "Community Health Workers"—people from the same neighborhoods and backgrounds who can say, "I get it, let's fix this together."
3. Turn Down the Heat: They need to teach people how to manage their diabetes and help them quit smoking, because those are the things making the "engine" rev too high.
4. Listen to the Pain: They need to take the pain seriously. You can't fix a broken heart if the person is in too much physical pain to care.

The Bottom Line

This report is a siren alarm. It tells us that the people living on the streets of San Francisco are in a "silent emergency." Their bodies are under immense stress, their "pressure cookers" are ticking, and they are at high risk of having a heart attack or stroke very soon.

The good news? The team proved that you can check these vital signs on the street and get good data. Now, the city and the community need to act like a pit crew: diagnose the problem, fix the engine, and get these people back on solid ground before the engine blows.

Technical Summary

1. Problem Statement

The unsheltered homeless population faces extreme health disparities, with mortality rates three to four times higher than the general population. Cardiovascular disease (CVD) is a leading cause of death in this group, driven by unmanaged hypertension, diabetes, smoking, and chronic stress. However, existing literature largely relies on self-reported or retrospective data, lacking real-time, field-based clinical measurements of acute cardiovascular risk. There is a critical knowledge gap regarding the immediate physiological status of unsheltered adults, hindering the ability of outreach organizations to prioritize medical referrals, activate emergency protocols, and allocate resources effectively.

2. Methodology

The study was conducted as an Operational Needs Assessment (ONA) by the Valliant Foundation in San Francisco (Inner Mission, Tenderloin, Embarcadero) from November 2024 to September 2025. It was classified as Non-Human Subjects Research (NHSR) by an independent ethics committee, focusing on operational decision-making rather than generalizable scientific knowledge.

• Participants: 116 total responses were collected.
  • Demographics/Vitals: 72 participants provided demographic data and vital signs.
  • Risk Factors: 116 participants provided data on nicotine use and diabetes history (collected via a secondary survey).
• Inclusion/Exclusion Criteria:
  • Inclusion: Unsheltered individuals who were Alert and Oriented x4 (A&O x4) and provided verbal informed consent.
  • Exclusion: Individuals with altered mental status, acute psychosis, severe intoxication, or those who were stably housed. A "Hard Stop" protocol was enforced to ensure capacity and safety.
• Data Collection Tools:
  • Physiological: FDA-cleared automated wrist sphygmomanometers (G.Lab, Omron) for Blood Pressure (BP) and finger pulse oximeters (CVS) for Heart Rate (HR).
  • Subjective: Numeric Pain Rating Scale (NPRS) 0–10.
  • Self-Report: Structured questionnaires for nicotine use and diabetes history.
• Data Processing & Analysis:
  • Cleaning: Entries with missing values, single-number BP readings, or physiological outliers (e.g., HR > 200, impossible BP ratios) were excluded.
  • Transformation: Pain scores of "0" were converted to "1"; scores >10 were capped at "10".
  • Classification:
    • BP: Categorized per ACC/AHA (2017) guidelines (Normal, Elevated, Stage 1/2 Hypertension, Hypertensive Crisis).
    • HR: Categorized as Bradycardia (<60), Normal (60–100), Tachycardia (>100).
    • Pain: Mild (1–4), Moderate (5–6), Severe (7–10).
  • Statistical Approach: Descriptive statistics (mean, median, frequency) were used. No inferential statistics or hypothesis testing were performed.

3. Key Contributions

• Field-Based Clinical Data: Provides rare, real-time physiological data (BP, HR) from an unsheltered population, moving beyond self-reported estimates to quantify acute risk.
• Operational Framework: Demonstrates a feasible model for integrating clinical screening into street outreach, including a defined "Emergency Action Protocol" for hypertensive crises.
• Ethical Operational Model: Establishes a rigorous ethical framework for non-research health assessments, prioritizing participant autonomy, capacity screening, and immediate safety interventions (e.g., offering 911 calls for critical readings).
• Demographic Insight: Highlights the specific intersection of race, gender, and health burden, noting the overrepresentation of Black/African American individuals in the unsheltered cohort and their specific cardiovascular risks.

4. Key Results

Demographics (N=72 usable for age/gender):

• Age: Mean 42.96 years (Range: 24–70).
• Gender: Predominantly male (73.61% Male, 20.83% Female).
• Race: 41.67% White/Caucasian, 37.50% Black/African American.

Cardiovascular Metrics:

• Blood Pressure (N=59 usable):
  • Total High Blood Pressure (Stage 1+): 72.88% (N=43).
  • Hypertensive Crisis (SYS ≥180 OR DIA ≥120): 10.17% (N=6). Critical readings included extremes such as 286/127 mmHg and 218/154 mmHg.
  • Distribution: 45.76% Stage 2 Hypertension, 16.95% Stage 1, 10.17% Crisis, 8.47% Normal/Elevated, 8.47% Hypotensive.
• Heart Rate (N=65 usable):
  • Mean/Median: 96 bpm.
  • Tachycardia (>100 bpm): 36.92% (N=24).
  • Bradycardia (<60 bpm): 1.54% (N=1).
• Pain Burden (N=66 usable):
  • Mean Pain Score: 3.74.
  • Severe Pain (7–10): 19.70% (N=13).

Comorbidities (Self-Reported):

• Nicotine Use: 15.38% (N=6 of 39 usable).
• Diabetes: 13.16% (N=5 of 38 usable).

5. Significance and Recommendations

Significance:
The study reveals a "silent emergency" within the unsheltered community, where severe, untreated hypertension is pervasive. The prevalence of hypertensive crisis (10.17%) is alarmingly high compared to general population norms, indicating an imminent risk of stroke, myocardial infarction, and renal failure. The combination of high heart rates, severe pain, and unmanaged risk factors suggests a state of chronic physiological stress.

Recommendations for Outreach Services:

1. Emergency Referral Pathways: Immediate establishment of dedicated transport and referral mechanisms for individuals with Hypertensive Crisis readings who decline EMS activation.
2. Targeted Education: Implementation of culturally competent health education, specifically tailored for Black/African American populations to address historical mistrust and improve hypertension management.
3. Chronic Disease Management: Integration of blood glucose screening, diabetes education, and smoking cessation programs into routine outreach.
4. Pain Management: Development of protocols to address severe chronic pain, which acts as a driver for elevated heart rates and cardiovascular instability.
5. Operational Scaling: Adoption of similar multi-metric screening protocols by other outreach organizations to identify at-risk individuals early.

Limitations:
The study is descriptive and non-inferential. Selection bias exists due to the exclusion of individuals with severe intoxication or psychosis (who may have the highest risk). Self-reported data for diabetes and smoking are subject to recall bias. The sample is specific to San Francisco and may not be fully generalizable to all homeless populations.