Hospital Admission Guidelines for Community-Acquired Pneumonia in Adults
Use validated severity-assessment tools—specifically the Pneumonia Severity Index (PSI) or CURB-65 score—combined with clinical judgment to determine whether a patient requires hospital admission; patients in PSI risk classes IV–V or with CURB-65 ≥2 should be hospitalized, while those in PSI classes I–III may be managed as outpatients unless specific high-risk features are present. 1
Severity Assessment Tools
Pneumonia Severity Index (PSI)
PSI stratifies patients into five mortality risk classes based on age, comorbidities, vital signs, and laboratory findings; it was derived and validated in over 50,000 patients and reliably predicts 30-day mortality. 1
PSI class I and II patients (lowest risk) should be treated as outpatients, PSI class III patients may be observed or briefly hospitalized, and PSI classes IV and V require inpatient care due to significantly elevated mortality risk. 1
The PSI has been validated across multiple settings and consistently identifies low-risk patients who can safely avoid hospitalization, reducing unnecessary admissions by 20–30% when applied systematically. 1, 2
CURB-65 Criteria
CURB-65 assigns one point each for Confusion, Uremia (BUN >20 mg/dL), Respiratory rate ≥30/min, low Blood pressure (SBP <90 or DBP ≤60 mmHg), and age ≥65 years; a score ≥2 indicates need for hospitalization. 1
CURB-65 is simpler than PSI and can be calculated rapidly at the bedside, making it particularly useful in emergency departments and primary-care settings. 1
A CURB-65 score of 0–1 identifies patients suitable for outpatient management, while scores ≥2 warrant admission due to increased mortality and complication rates. 1
Absolute Indications for Hospital Admission
Even when severity scores suggest low risk, hospitalization is mandatory if any of the following are present 1, 3:
- Hypoxemia: Arterial oxygen tension (PaO₂) <60 mmHg or oxygen saturation <92% on room air
- Hemodynamic instability: Systolic blood pressure <90 mmHg or diastolic blood pressure ≤60 mmHg, or need for vasopressor support
- Altered mental status or confusion (independent of age)
- Inability to maintain oral intake due to vomiting, severe dysphagia, or gastrointestinal dysfunction
- Severe laboratory abnormalities: White blood cell count <4,000/µL, absolute neutrophil count <1,000/µL, hematocrit <30%, or platelet count <100,000/µL
- Acute kidney injury: Serum creatinine >1.2 mg/dL or BUN >20 mg/dL (7 mmol/L)
- Metabolic acidosis (arterial pH <7.35) or evidence of sepsis/organ dysfunction
- Radiographic complications: Multilobar infiltrates, cavitation, rapid radiographic progression, or pleural effusion
- Decompensated comorbid illness requiring inpatient management (e.g., heart failure exacerbation, COPD exacerbation, uncontrolled diabetes)
- Social factors: Absence of a responsible caregiver or unstable home environment
ICU Admission Criteria
ICU-level care is indicated when one major criterion OR three or more minor criteria are met 1, 4:
Major Criteria (any one requires ICU)
- Septic shock requiring vasopressor support
- Respiratory failure requiring mechanical ventilation (invasive or non-invasive)
Minor Criteria (≥3 require ICU)
- Respiratory rate ≥30 breaths/min
- PaO₂/FiO₂ ratio <250
- Multilobar infiltrates on chest imaging
- Confusion or altered mental status
- Uremia (BUN ≥20 mg/dL)
- Leukopenia (WBC <4,000/µL)
- Thrombocytopenia (platelets <100,000/µL)
- Hypothermia (core temperature <36°C)
- Hypotension requiring aggressive fluid resuscitation
Clinical Judgment and the "Art of Medicine"
Severity scores have high negative predictive value (>95%) but modest positive predictive value, meaning they reliably identify patients who can be treated as outpatients but may miss some who require admission. 1, 5
Physicians should not rely solely on PSI or CURB-65; clinical gestalt remains essential, and approximately 30% of "low-risk" patients are appropriately hospitalized based on factors not captured by scoring systems. 1
When the overall clinical appearance is concerning—even if objective criteria are not fully met—place the patient under observation status for 24–48 hours until stability is confirmed. 1
Studies demonstrate that physicians often overestimate severity and hospitalize low-risk patients unnecessarily, but they also occasionally underestimate severity at initial evaluation, underscoring the need for structured assessment tools. 1
Additional Risk Factors That Lower the Threshold for Admission
The following features increase the likelihood of a complicated course and should prompt strong consideration of hospitalization, even in patients with low PSI or CURB-65 scores 1, 5, 6:
- Age >65 years (independent predictor of mortality)
- Chronic comorbidities: COPD, bronchiectasis, diabetes mellitus, chronic renal failure, congestive heart failure, chronic liver disease, malignancy, cerebrovascular disease, chronic alcohol abuse, malnutrition, or immunosuppression
- Recent hospitalization within the past year
- Aspiration risk factors: Impaired swallowing, neurologic disease, altered consciousness
- High-risk pathogens: Suspected Legionella, Staphylococcus aureus, or Gram-negative organisms
- Symptom duration <7 days or >28 days (atypical time course suggests complicated infection)
Outpatient Management Criteria
Patients may be safely managed as outpatients if ALL of the following are met 1, 2:
- PSI class I–III or CURB-65 score 0–1
- No absolute indications for admission (see above)
- Oxygen saturation ≥92% on room air
- Hemodynamically stable (SBP ≥90 mmHg, HR ≤100 bpm)
- Able to tolerate oral medications and maintain hydration
- Reliable social support and safe home environment
- Access to follow-up care within 48 hours
Implementation and Outcomes
Implementation of structured admission decision support reduces unnecessary hospitalizations by 30–50% without increasing adverse outcomes, as demonstrated in multiple prospective studies. 1, 2
Outpatient treatment is preferred when safe because it allows faster return to normal activity, is strongly preferred by patients (80% prefer outpatient care), costs 25 times less than inpatient care, and avoids hospital-acquired complications such as thromboembolic events and nosocomial infections. 1
Mortality in appropriately selected outpatients is <1%, while hospitalized patients have mortality rates of 5–15% depending on severity, validating the safety of outpatient management in low-risk cases. 1, 2
Common Pitfalls to Avoid
Do not admit patients based solely on age or single abnormal vital sign; use validated severity scores to guide decisions and avoid unnecessary hospitalizations. 1
Do not discharge patients with hypoxemia, hemodynamic instability, or inability to maintain oral intake, even if PSI or CURB-65 scores are low; these features override scoring systems. 1, 3
Do not ignore social factors; lack of caregiver support or unstable housing is a legitimate reason for admission even in low-risk patients. 1
Reassess patients who do not improve within 48–72 hours of outpatient therapy; failure to respond warrants hospital admission and further evaluation for complications or resistant organisms. 1