Factors Influencing Low-Acuity Emergency Medical Services Use: An Observational Study Guided by the Andersen Behavioral Model

This observational study of over 41,000 low-acuity emergency department visits in the southeastern United States, guided by the Andersen Behavioral Model, found that predisposing and enabling factors—particularly older age, unemployment or disability, mental health diagnoses, and lack of established primary care—were the primary drivers of EMS utilization, suggesting that expanding access to primary and behavioral health services could reduce unnecessary low-acuity EMS use.

The Gist

Imagine the Emergency Medical Services (EMS) system—the ambulances and paramedics—as a fire department. Its primary job is to rush to burning buildings (life-threatening emergencies) to put out the fire and save lives.

However, in recent years, this "fire department" has been getting overwhelmed. Not because there are more fires, but because people are calling them for things that aren't fires at all—like a small candle that just needs to be blown out, or a question about how to cook a meal. These are "low-acuity" calls: minor health issues that could easily be handled by a regular doctor or a clinic, but instead, people are calling 911 for an ambulance ride.

This study is like a detective investigation trying to figure out why people are calling the "fire department" for a "candle." The researchers didn't just guess; they used a famous map called Andersen's Behavioral Model to organize their clues. Think of this model as a three-bucket system to sort out the reasons people use health services:

1. Predisposing Factors: Who you are (your age, job, personality).
2. Enabling Factors: What helps or stops you (do you have a car? a regular doctor? money?).
3. Need Factors: How sick you actually feel (do you have a mental health struggle? is it the middle of the night?).

The Investigation: What They Found

The researchers looked at a massive pile of data—over 41,000 visits to emergency rooms in the Southeast US where the patients had minor issues (like a mild stomach ache or a small cut) and were sent home. They wanted to see who took the ambulance and who drove themselves.

Here is what they discovered, translated into everyday terms:

🚑 The "High-Risk" Groups (More likely to call an ambulance):

• The "Older & Tired" Crowd: As people get older, they are more likely to call an ambulance. Imagine an elderly person who lives alone; even a small stumble feels scary, and they don't have a family member to drive them, so they call the pros.
• The "Jobless & Retired": People who are unemployed, retired, or disabled were much more likely to take an ambulance. It's like having no car and no one to give you a ride; the ambulance becomes your only taxi.
• The "Night Owls": If you show up at the ER late at night, you are more likely to get an ambulance. Maybe it's because the regular clinics are closed, and you feel too scared to drive yourself in the dark.
• The "Anxious Minds": People with mental health diagnoses were significantly more likely to use EMS. When you are feeling overwhelmed or panicked, the idea of driving yourself to the hospital feels impossible, so you call for help.

🛡️ The "Protective" Groups (Less likely to call an ambulance):

• The "Regulars": If you have a primary care doctor you see regularly, you are much less likely to call an ambulance. It's like having a trusted mechanic; you call them first instead of panicking and calling a tow truck for a flat tire.
• The "Weekenders": Surprisingly, people were less likely to take an ambulance on weekends. The researchers think this is because people might be waiting until Monday to see a doctor, or the system behaves differently on weekends.
• The "Pandemic Era": During and after the COVID-19 pandemic, fewer people took ambulances for minor issues. This is likely because everyone was scared of getting sick in a crowded ambulance or hospital, so they stayed home or used telehealth (video calls with doctors).

❓ The Surprising "No-Connection" Factors:
You might think that where you live (rural vs. city), what language you speak, or what kind of insurance you have would change how you get to the hospital. But in this study, none of those mattered. Whether you lived in a farm town or a big city, or whether you had private insurance or government aid, it didn't change the odds of you taking an ambulance. The main drivers were your social situation (job, age, mental health) and access to a regular doctor.

The Big Takeaway

The study concludes that low-acuity ambulance rides aren't usually about the medical emergency itself. Instead, they are about logistics and life circumstances.

Think of the ambulance as a VIP shuttle bus. If you have a regular doctor (a personal driver), you don't need the shuttle. If you are retired, unemployed, or feeling anxious, and you don't have a personal driver, the shuttle becomes your only option.

What should we do?
The authors suggest we stop blaming the patients and start fixing the system:

1. Give everyone a "Personal Driver": Expand access to regular primary care doctors so people have someone to call before 911.
2. Support the "Anxious Minds": Improve mental health services so people feel less panicked and more capable of making their own way to care.
3. New Rules for the Shuttle: Change how ambulances get paid. Right now, they get paid to drive you to the hospital. We need to pay them to help you stay at home or take you to a clinic if it's safe.

In short: To save the "fire department" for the real fires, we need to make sure people have other ways to handle the "candles."

Technical Summary

1. Problem Statement

Emergency Medical Services (EMS) in the United States are facing increasing strain due to rising call volumes, workforce shortages, and a growing proportion of low-acuity encounters (non-emergent conditions). The current "scoop-and-run" transport model, reinforced by federal reimbursement structures, often directs low-acuity patients to Emergency Departments (EDs), contributing to overcrowding and diverting resources from critical, time-sensitive emergencies.

While prior research suggests that social determinants (e.g., age, unemployment, lack of primary care) influence EMS use, existing literature suffers from:

• Regional sampling limitations (single-agency or municipality studies).
• Inconsistent definitions of "low-acuity."
• Short observation periods.
• A lack of theoretical grounding to explain why these patients utilize EMS.

This study aims to fill these gaps by identifying predisposing, enabling, and need-based factors associated with EMS transport for low-acuity ED patients using a robust theoretical framework.

2. Methodology

Study Design & Setting:

• Type: Secondary, retrospective observational study.
• Data Source: A 10% random sample of electronic health record (EHR) data from 21 emergency departments across a multi-hospital regional referral network in the Southeastern US (North Carolina, South Carolina, Georgia, Virginia), serving >7 million people.
• Timeframe: January 1, 2016, to April 29, 2025 (covering pre-pandemic, pandemic, and post-pandemic periods).

Participants & Inclusion Criteria:

• Target Population: Patients aged >17 years.
• Triage Level: Emergency Severity Index (ESI) 4 or 5 (non-urgent).
• Disposition: Discharged from the ED (excluded patients admitted or requiring acute hospitalization).
• Exclusions: Pregnant patients, suicidal ideation/self-harm, involuntary commitment, and non-standard arrivals (e.g., air medical, law enforcement).

Theoretical Framework:
The study utilized Andersen's Behavioral Model (ABM) of Health Services Use to categorize independent variables into three domains:

1. Predisposing Factors: Demographics and social characteristics (Age, Sex, Race, Primary Language, Employment Status).
2. Enabling Factors: Logistical and structural resources (Insurance type, Rurality/Urban status, Established Primary Care Clinician, Time of Arrival).
3. Need-Based Factors: Perceived or evaluated illness severity (Mental Health Diagnosis, Pandemic Era).

Statistical Analysis:

• Primary Outcome: Mode of arrival (1 = EMS/Ambulance; 0 = Other modes).
• Analysis: Multivariable logistic regression to estimate Adjusted Odds Ratios (aOR) with 95% Confidence Intervals (CI).
• Controls: Variables were harmonized across EHR systems; multicollinearity was checked (VIF < 5). Listwise deletion was used for missing data.

3. Key Contributions

• Theoretical Application: This is one of the first studies to explicitly apply Andersen's Behavioral Model to EMS utilization, moving beyond simple demographic correlations to a structured analysis of how social and structural factors interact.
• Scale and Scope: It utilizes a large, multi-state dataset (21 EDs) with a 9-year longitudinal span, overcoming the limitations of single-agency studies.
• Pandemic Context: It uniquely analyzes EMS utilization trends across three distinct eras: Pre-COVID, COVID, and Post-COVID, providing insight into how public health crises alter care-seeking behavior.
• Definition Clarity: It strictly defines low-acuity as ESI 4/5 and discharge, excluding patients who were eventually admitted, ensuring the focus remains on truly non-emergent cases.

4. Results

Sample Demographics:

• Total Encounters: 41,772 low-acuity ED visits.
• EMS Utilization: 3,233 (7.7%) arrived via EMS; 92.3% arrived via other modes.

Significant Predictors of EMS Use (Increased Odds):

• Age: Older age was a strong predictor (aOR 1.30 per 10-year increase).
• Employment Status: Being retired/disabled (aOR 3.60) or unemployed (aOR 2.26) significantly increased odds compared to employed patients.
• Sex: Male patients had higher odds (aOR 1.20).
• Time of Day: Nighttime presentation (7 PM – 6:59 AM) increased odds (aOR 1.63).
• Clinical Need: Presence of a mental health diagnosis (ICD-10 "F" codes) increased odds (aOR 1.76).

Significant Predictors of Reduced EMS Use (Protective Factors):

• Primary Care: Having an established primary care clinician significantly reduced odds (aOR 0.57).
• Timing: Weekend presentation (aOR 0.91) and visits during the COVID (aOR 0.63) or Post-COVID (aOR 0.54) periods were associated with lower EMS use.
• Race: White race was associated with slightly lower odds (aOR 0.89) compared to non-White patients.

Non-Significant Factors:

• Rurality (Urban vs. Rural), Insurance type (Commercial, Medicaid/Medicare, Self-pay), and Primary Language showed no statistically significant association with EMS use in the multivariable model.

5. Significance and Implications

Systemic Impact:
The study confirms that predisposing and enabling factors (social determinants and access to care) are the primary drivers of low-acuity EMS use, rather than clinical acuity alone. The 7.7% utilization rate aligns with national estimates that 5–15% of EMS transports are non-emergent, representing a significant drain on system capacity.

Policy and Clinical Recommendations:

• Primary Care Expansion: The strong protective effect of established primary care suggests that expanding access to outpatient and behavioral health services, particularly for older adults and those with mental health conditions, could divert low-acuity patients from EMS.
• Reform Reimbursement Models: Since insurance type was not a driver, the authors suggest that funding models (like the CMS ET3 Model) should decouple payment from hospital transport, allowing for "treat-in-place" or alternative destination pathways.
• Targeted Interventions: Community Paramedicine and nurse triage lines should be prioritized for high-risk groups identified in the study: older adults, the unemployed/disabled, and those with mental health diagnoses.
• Mental Health Integration: The high correlation between mental health diagnoses and EMS use highlights a critical gap in community behavioral health support.

Conclusion:
The study concludes that addressing low-acuity EMS utilization requires a shift from a purely clinical response to a social and behavioral health approach. By targeting the underlying social drivers (employment status, access to primary care, and mental health support), health systems can preserve EMS capacity for true emergencies while improving care continuity for vulnerable populations.