In hospitalised patients who develop respiratory symptoms and fever, one should consider the diagnosis. The likelihood increases when upon investigation symptoms are found of respiratory insufficiency, purulent secretions, newly developed infiltrate on the chest X-Ray, and increasing leucocyte count. If pneumonia is suspected material from sputum or tracheal aspirates are sent to the microbiology department for cultures. In case of pleural effusion thoracentesis is performed for examination of pleural fluid. In suspected ventilator-associated pneumonia it has been suggested that bronchoscopy(BAL) is necessary because of the known risks surrounding clinical diagnoses.

Patients with symptoms of CAP require evaluation. Diagnosis of pneumonia is made clinically, rather than on the basis of a particular test. Evaluation begins with a physical examination by a health provider, which may reveal fever, an increased respiratory rate (tachypnea), low blood pressure (hypotension), a fast heart rate (tachycardia) and changes in the amount of oxygen in the blood. Palpating the chest as it expands and tapping the chest wall (percussion) to identify dull, non-resonant areas can identify stiffness and fluid, signs of CAP. Listening to the lungs with a stethoscope (auscultation) can also reveal signs associated with CAP. A lack of normal breath sounds or the presence of crackles can indicate fluid consolidation. Increased vibration of the chest when speaking, known as tactile fremitus, and increased volume of whispered speech during auscultation can also indicate fluid.

When signs of pneumonia are discovered during evaluation, chest X-rays, are performed to support a diagnosis of CAP, and examination of the blood and sputum for infectious microorganisms and blood tests may be used to support a diagnosis of CAP. Diagnostic tools depend on the severity of illness, local practices and concern about complications of the infection. All patients with CAP should have their blood oxygen monitored with pulse oximetry. In some cases, arterial blood gas analysis may be required to determine the amount of oxygen in the blood. A complete blood count (CBC) may reveal extra white blood cells, indicating infection.

Chest X-rays and X-ray computed tomography (CT) can reveal areas of opacity (seen as white), indicating consolidation. CAP does not always appear on x-rays, because the disease is in its initial stages or involves a part of the lung an x-ray does not see well. In some cases, chest CT can reveal pneumonia not seen on x-rays. However, congestive heart failure or other types of lung damage can mimic CAP on x-rays.

Several tests can identify the cause of CAP. Blood cultures can isolate bacteria or fungi in the bloodstream. Sputum Gram staining and culture can also reveal the causative microorganism. In severe cases, bronchoscopy can collect fluid for culture. Special tests can be performed if an uncommon microorganism is suspected, such as urinalysis for Legionella antigen in Legionnaires' disease.

Clinical prediction rules have been developed to more objectively predict outcomes of pneumonia. These rules are often used in deciding whether or not to hospitalize the person.

- Pneumonia severity index (or "PSI Score")

- CURB-65 score, which takes into account the severity of symptoms, any underlying diseases, and age

Bacteremia is most commonly diagnosed by blood culture, in which a sample of blood drawn from the vein by needle puncture is allowed to incubate with a medium that promotes bacterial growth. If bacteria are present in the bloodstream at the time the sample is obtained, the bacteria will multiply and can thereby be detected.

Any bacteria that incidentally find their way to the culture medium will also multiply. For example, if the skin is not adequately cleaned before needle puncture, contamination of the blood sample with normal bacteria that live on the surface of the skin can occur. For this reason, blood cultures must be drawn with great attention to sterile process. The presence of certain bacteria in the blood culture, such as S"taphylococcus aureus", "Streptococcus pneumoniae", and "Escherichia coli" almost never represent a contamination of the sample. On the other hand, contamination may be more highly suspected if organisms like "Staphylococcus epidermidis" or "Propionibacterium acnes" grow in the blood culture.

Two blood cultures drawn from separate sites of the body are often sufficient to diagnose bacteremia. Two out of two cultures growing the same type of bacteria usually represents a real bacteremia, particularly if the organism that grows is not a common contaminant. One out of two positive cultures will usually prompt a repeat set of blood cultures to be drawn to confirm whether a contaminant or a real bacteremia is present. The patient's skin is typically cleaned with an alcohol-based product prior to drawing blood to prevent contamination. Blood cultures may be repeated at intervals to determine if persistent — rather than transient — bacteremia is present.

Prior to drawing blood cultures, a thorough patient history should be taken with particular regard to presence of both fevers and chills, other focal signs of infection such as in the skin or soft tissue, a state of immunosuppression, or any recent invasive procedures.

Ultrasound of the heart is recommended in all those with bacteremia due to "Staphylococcus aureus" to rule out infectious endocarditis.

In pneumonia, a collection of fluid may form in the space that surrounds the lung. Occasionally, microorganisms will infect this fluid, causing an empyema. To distinguish an empyema from the more common simple parapneumonic effusion, the fluid may be collected with a needle (thoracentesis), and examined. If this shows evidence of empyema, complete drainage of the fluid is necessary, often requiring a drainage catheter. In severe cases of empyema, surgery may be needed. If the infected fluid is not drained, the infection may persist, because antibiotics do not penetrate well into the pleural cavity. If the fluid is sterile, it must be drained only if it is causing symptoms or remains unresolved.

In rare circumstances, bacteria in the lung will form a pocket of infected fluid called a lung abscess. Lung abscesses can usually be seen with a chest X-ray but frequently require a chest CT scan to confirm the diagnosis. Abscesses typically occur in aspiration pneumonia, and often contain several types of bacteria. Long-term antibiotics are usually adequate to treat a lung abscess, but sometimes the abscess must be drained by a surgeon or radiologist.

Some CAP patients require intensive care, with clinical prediction rules such as the pneumonia severity index and CURB-65 guiding the decision to hospitalize. Factors increasing the need for hospitalization include:

- Age greater than 65

- Underlying chronic illnesses

- Respiratory rate greater than 30 per minute

- Systolic blood pressure less than 90 mmHg

- Heart rate greater than 125 per minute

- Temperature below 35 or over 40 °C

- Confusion

- Evidence of infection outside the lung

Laboratory results indicating hospitalization include:

- Arterial oxygen tension less than 60 mm Hg

- Carbon dioxide over 50 mmHg or pH under 7.35 while breathing room air

- Hematocrit under 30 percent

- Creatinine over 1.2 mg/dl or blood urea nitrogen over 20 mg/dl

- White-blood-cell count under 4 × 10^9/L or over 30 × 10^9/L

- Neutrophil count under 1 x 10^9/L

X-ray findings indicating hospitalization include:

- Involvement of more than one lobe of the lung

- Presence of a cavity

- Pleural effusion

Patients with HCAP are more likely than those with community-acquired pneumonia to receive inappropriate antibiotics that do not target the bacteria causing their disease.

In 2002, an expert panel made recommendations about the evaluation and treatment of probable nursing home-acquired pneumonia. They defined probably pneumonia, emphasized expedite antibiotic treatment (which is known to improve survival) and drafted criteria for the hospitalization of willing patients.

For initial treatment in the nursing home, a fluoroquinolone antibiotic suitable for respiratory infections (moxifloxacin, for example), or amoxicillin with clavulanic acid plus a macrolide has been suggested. In a hospital setting, injected (parenteral) fluoroquinolones or a second- or third-generation cephalosporin plus a macrolide could be used. Other factors that need to be taken into account are recent antibiotic therapy (because of possible resistance caused by recent exposure), known carrier state or risk factors for resistant organisms (for example, known carrier of MRSA or presence of bronchiectasis predisposing to Pseudomonas aeruginosa), or suspicion of possible Legionella pneumophila infection (legionnaires disease).

In 2005, the American Thoracic Society and Infectious Diseases Society of America have published guidelines suggesting antibiotics specifically for HCAP. The guidelines recommend combination therapy with an agent from each of the following groups to cover for both "Pseudomonas aeruginosa" and MRSA. This is based on studies using sputum samples and intensive care patients, in whom these bacteria were commonly found.

- cefepime, ceftazidime, imipenem, meropenem or piperacillin–tazobactam; plus

- ciprofloxacin, levofloxacin, amikacin, gentamicin, or tobramycin; plus

- linezolid or vancomycin

In one observational study, empirical antibiotic treatment that was not according to international treatment guidelines was an independent predictor of worse outcome among HCAP patients.

Guidelines from Canada suggest that HCAP can be treated like community-acquired pneumonia with antibiotics targeting Streptococcus pneumoniae, based on studies using blood cultures in different settings which have not found high rates of MRSA or Pseudomonas.

Besides prompt antibiotic treatment, supportive measure for organ failure (such as cardiac decompensation) are also important. Another consideration goes to hospital referral; although more severe pneumonia requires admission to an acute care facility, this also predisposes to hazards of hospitalization such as delirium, urinary incontinence, depression, falls, restraint use, functional decline, adverse drug effects and hospital infections. Therefore, mild pneumonia might be better dealt with inside the long term care facility. In patients with a limited life expectancy (for example, those with advanced dementia), end-of-life pneumonia also requires recognition and appropriate, palliative care.

For healthcare-associated bacteremia due to intravenous catheters, the IDSA has published guidelines for catheter removal. Short term catheters (in place 14 days) should be removed if the patient is developing signs or symptoms of sepsis or endocarditis, or if blood cultures remain positive for more than 72 hours.

Antibiotics are the treatment of choice for bacterial pneumonia, with ventilation (oxygen supplement) as supportive therapy. The antibiotic choice depends on the nature of the pneumonia, the microorganisms most commonly causing pneumonia in the geographical region, and the immune status and underlying health of the individual. In the United Kingdom, amoxicillin is used as first-line therapy in the vast majority of patients acquiring pneumonia in the community, sometimes with added clarithromycin. In North America, where the "atypical" forms of community-acquired pneumonia are becoming more common, clarithromycin, azithromycin, or fluoroquinolones as single therapy have displaced the amoxicillin as first-line therapy.

Local patterns of antibiotic-resistance always need to be considered when initiating pharmacotherapy. In hospitalized individuals or those with immune deficiencies, local guidelines determine the selection of antibiotics.

People who have difficulty breathing due to pneumonia may require extra oxygen. An extremely sick individual may require artificial ventilation and intensive care as life-saving measures while his or her immune system fights off the infectious cause with the help of antibiotics and other drugs.

Chest radiographs (X-ray photographs) often show a pulmonary infection before physical signs of atypical pneumonia are observable at all.

This is occult pneumonia. In general, occult pneumonia is rather often present in patients with pneumonia and can also be caused by "Streptococcus pneumoniae", as the decrease of occult pneumonia after vaccination of children with a pneumococcal vaccine suggests.

Infiltration commonly begins in the perihilar region (where the bronchus begins) and spreads in a wedge- or fan-shaped fashion toward the periphery of the lung field. The process most often involves the lower lobe, but may affect any lobe or combination of lobes.

Mycoplasma is found more often in younger than in older people.

Older people are more often infected by Legionella.

A clinical diagnosis may be made by taking a history and doing a brief examination. Treatment is usually started without or before confirmation by laboratory analysis.

PCR-based screening methodologies are in the process of development. Although they speed up detection immensely, they are costly and the reliability of the tests is questionable due to false positives. Nested arbitrary PCR (ARB-PCR) was used during a 2007 CRE outbreak at the University of Virginia Medical Center to identify the specific "bla" KPC plasmid involved in the transmission of the infection, and researchers suggest that ARB-PCR may also be used to identify other methods of CRE spread.

Neonatal sepsis screening:

1. DLC (differential leukocyte count) showing increased numbers of polymorphs.

2. DLC: band cells > 20%.

3. increased haptoglobins.

4. micro ESR (Erythrocyte Sedimentation Rate) titer > 15mm.

5. gastric aspirate showing > 5 polymorphs per high power field.

6. newborn CSF (Cerebrospinal fluid) screen: showing increased cells and proteins.

7. suggestive history of chorioamnionitis, PROM (Premature rupture of membranes), etc...

Culturing for microorganisms from a sample of CSF, blood or urine, is the gold standard test for definitive diagnosis of neonatal sepsis. This can give false negatives due to the low sensitivity of culture methods and because of concomitant antibiotic therapy. Lumbar punctures should be done when possible as 10-15% presenting with sepsis also have meningitis, which warrants an antibiotic with a high CSF penetration.

CRP is not very accurate in picking up cases.

Cultures of stool samples are examined to identify the organism causing dysentery. Usually, several samples must be obtained due to the number of amoebae, which changes daily. Blood tests can be used to measure abnormalities in the levels of essential minerals and salts.

Recovery from an anaerobic infection depends on adequate and rapid management. The main principles of managing anaerobic infections are neutralizing the toxins produced by anaerobic bacteria, preventing the local proliferation of these organisms by altering the environment and preventing their dissemination and spread to healthy tissues.

Toxin can be neutralized by specific antitoxins, mainly in infections caused by Clostridia (tetanus and botulism). Controlling the environment can be attained by draining the pus, surgical debriding of necrotic tissue, improving blood circulation, alleviating any obstruction and by improving tissue oxygenation. Therapy with hyperbaric oxygen (HBO) may also be useful. The main goal of antimicrobials is in restricting the local and systemic spread of the microorganisms.

The available parenteral antimicrobials for most infections are metronidazole, clindamycin, chloramphenicol, cefoxitin, a penicillin (i.e. ticarcillin, ampicillin, piperacillin) and a beta-lactamase inhibitor (i.e. clavulanic acid, sulbactam, tazobactam), and a carbapenem (imipenem, meropenem, doripenem, ertapenem). An antimicrobial effective against Gram-negative enteric bacilli (i.e. aminoglycoside) or an anti-pseudomonal cephalosporin (i.e. cefepime ) are generally added to metronidazole, and occasionally cefoxitin when treating intra-abdominal infections to provide coverage for these organisms. Clindamycin should not be used as a single agent as empiric therapy for abdominal infections. Penicillin can be added to metronidazole in treating of intracranial, pulmonary and dental infections to provide coverage against microaerophilic streptococci, and Actinomyces.

Oral agents adequate for polymicrobial oral infections include the combinations of amoxicillin plus clavulanate, clindamycin and metronidazole plus a macrolide. Penicillin can be added to metronidazole in the treating dental and intracranial infections to cover "Actinomyces" spp., microaerophilic streptococci, and "Arachnia" spp. A macrolide can be added to metronidazole in treating upper respiratory infections to cover "S. aureus" and aerobic streptococci. Penicillin can be added to clindamycin to supplement its coverage against "Peptostreptococcus" spp. and other Gram-positive anaerobic organisms.

Doxycycline is added to most regimens in the treatment of pelvic infections to cover chlamydia and mycoplasma. Penicillin is effective for bacteremia caused by non-beta lactamase producing bacteria. However, other agents should be used for the therapy of bacteremia caused by beta-lactamase producing bacteria.

Because the length of therapy for anaerobic infections is generally longer than for infections due to aerobic and facultative anaerobic bacteria, oral therapy is often substituted for parenteral treatment. The agents available for oral therapy are limited and include amoxacillin plus clavulanate, clindamycin, chloramphenicol and metronidazole.

In 2010 the American Surgical Society and American Society of Infectious Diseases have updated their guidelines for the treatment of abdominal infections.

The recommendations suggest the following:

For mild-to-moderate community-acquired infections in adults, the agents recommended for empiric regimens are: ticarcillin- clavulanate, cefoxitin, ertapenem, moxifloxacin, or tigecycline as single-agent therapy or combinations of metronidazole with cefazolin, cefuroxime, ceftriaxone, cefotaxime, levofloxacin, or ciprofloxacin. Agents no longer recommended are: cefotetan and clindamycin ( Bacteroides fragilis group resistance) and ampicillin-sulbactam (E. coli resistance) and ainoglycosides (toxicity).

For high risk community-acquired infections in adults, the agents recommended for empiric regimens are: meropenem, imipenem-cilastatin, doripenem, piperacillin-tazobactam, ciprofloxacin or levofloxacin in combination with metronidazole, or ceftazidime or cefepime in combination with metronidazole. Quinolones should not be used unless hospital surveys indicate >90% susceptibility of "E. coli" to quinolones.

Aztreonam plus metronidazole is an alternative, but addition of an agent effective against gram-positive cocci is recommended. The routine use of an aminoglycoside or another second agent effective against gram-negative facultative and aerobic bacilli is not recommended in the absence of evidence that the infection is caused by resistant organisms that require such therapy.

Empiric use of agents effective against enterococci is recommended and agents effective against methicillin-resistant "S. aureus" (MRSA) or yeast is not recommended in the absence of evidence of infection due to such organisms.

Empiric antibiotic therapy for health care-associated intra-abdominal should be driven by local microbiologic results. Empiric coverage of likely pathogens may require multidrug regimens that include agents with expanded spectra of activity against gram-negative aerobic and facultative bacilli. These include meropenem, imipenem-cilastatin, doripenem, piperacillin-tazobactam, or ceftazidime or cefepime in combination with metronidazole. Aminoglycosides or colistin may be required.

Antimicrobial regimens for children include an aminoglycoside-based regimen, a carbapenem (imipenem, meropenem, or ertapenem), a beta-lactam/beta-lactamase-inhibitor combination (piperacillin-tazobactam or ticarcillin-clavulanate), or an advanced-generation cephalosporin (cefotaxime, ceftriaxone, ceftazidime, or cefepime) with metronidazole.

Clinical judgment, personal experience, safety and patient compliance should direct the physician in the choice of the appropriate antimicrobial agents. The length of therapy generally ranges between 2 and 4 weeks, but should be individualized depending on the response. In some instances treatment may be required for as long as 6–8 weeks, but can often be shortened with proper surgical drainage.

A lumbar puncture (LP) is necessary to diagnose meningitis. Cerebrospinal fluid (CSF) culture is the most important study for the diagnosis of neonatal bacterial meningitis because clinical signs are non-specific and unreliable. Blood cultures may be negative in 15-55% of cases, deeming it unreliable as well. However, a CSF/blood glucose ratio below two-thirds has a strong relationship to bacterial meningitis. A LP should be done in all neonates with suspected meningitis, with suspected or proven sepsis (whole body inflammation) and should be considered in all neonates in whom sepsis is a possibility. The role of the LP in neonates who are healthy appearing but have maternal risk factors for sepsis is more controversial; the yield of the LP in these patients may be low.

Early-onset is deemed when infection is within one week of birth. Late-onset is deemed after the first week.

Babies born from mothers with symptoms of Herpes Simplex Virus (HSV) should be tested for viral infection. Liver tests, complete blood count (CBC), cerebrospinal fluid analyses, and chest X-ray should all be completed to diagnose meningitis. Samples should be taken from skin, conjunctiva (eye), mouth and throat, rectum, urine, and the CSF for viral culture and PCR analysis with respect to the sample from CSF.

The disc diffusion method can be used by hospital laboratories to screen for CRE. In this technique, antibiotic discs are placed onto plates of Mueller Hinton agar that have already been inoculated with the sample strain. The plates are then incubated overnight at 37 °C. Following incubation, the zones of inhibition surrounding the various antibiotic discs are measured and compared with Clinical and Laboratory Standard Institute guidelines. Identification of KPCs, MBLs and OXAs can be achieved by demonstrating synergistic inhibition with phenyl boronic acid, EDTA or neither, respectively.

In a Thailand-based study of CRE in hospital settings, carbapenem resistance was defined as any strain that shows resistance to at least one of three carbapenem antibiotics tested.

Note that, in neonates, sepsis is difficult to diagnose clinically. They may be relatively asymptomatic until hemodynamic and respiratory collapse is imminent, so, if there is even a remote suspicion of sepsis, they are frequently treated with antibiotics empirically until cultures are sufficiently proven to be negative. In addition to fluid resuscitation and supportive care, a common antibiotic regimen in infants with suspected sepsis is a beta-lactam antibiotic (usually ampicillin) in combination with an aminoglycoside (usually gentamicin) or a third-generation cephalosporin (usually cefotaxime—ceftriaxone is generally avoided in neonates due to the theoretical risk of kernicterus.) The organisms which are targeted are species that predominate in the female genitourinary tract and to which neonates are especially vulnerable to, specifically Group B Streptococcus, "Escherichia coli", and "Listeria monocytogenes" (This is the main rationale for using ampicillin versus other beta-lactams.) Of course, neonates are also vulnerable to other common pathogens that can cause meningitis and bacteremia such as "Streptococcus pneumoniae" and "Neisseria meningitidis". Although uncommon, if anaerobic species are suspected (such as in cases where necrotizing enterocolitis or intestinal perforation is a concern, clindamycin is often added.

Granulocyte-macrophage colony stimulating factor (GM-CSF) is sometimes used in neonatal sepsis. However, a 2009 study found that GM-CSF corrects neutropenia if present but it has no effect on reducing sepsis or improving survival.

Trials of probiotics for prevention of neonatal sepsis have generally been too small and statistically underpowered to detect any benefit, but a randomized controlled trial that enrolled 4,556 neonates in India reported that probiotics significantly reduced the risk of developing sepsis. The probiotic used in the trial was "Lactobacillus plantarum".

A very large meta-analysis investigated the effect of probiotics on preventing late-onset sepsis (LOS) in neonates. Probiotics were found to reduce the risk of LOS, but only in babies who were fed human milk exclusively. It is difficult to distinguish if the prevention was a result of the probiotic supplementation or if it was a result of the properties of human milk. It is also still unclear if probiotic administration reduces LOS risk in extremely low birth weight infants due to the limited number of studies that investigated it. Out of the 37 studies included in this systematic review, none indicated any safety problems related to the probiotics. It would be beneficial to clarify the relationship between probiotic supplementation and human milk for future studies in order to prevent late onset sepsis in neonates.

The decision to treat bacteriuria depends on the presence of accompany symptoms and comorbidities.

Testing for bacteriuria is often performed in those with symptoms of a urinary tract infection. Testing is often done in other scenarios as in failure to thrive of a newborn or confusion in the elderly. Screening for bacteriuria is recommended in pregnancy as there is evidence that asymptomatic bacteriuria can lead to low birth weight and preterm delivery.

- Bacteriuria can be detected by urine dipstick test. The urinary nitrite test will be able to detect any nitrate-reducing bacteria in the urine. The leukocyte esterase test detects the presence of leukocytes (white blood cells) in the urine which can be associated with a urinary tract infection.The urine dipstick test is readily available and provides fast results.

- Microscopy can also be used to detect bacteriuria. It is more specific, especially when used with gram staining, but requires more time and equipment.

- The gold standard for detecting bacteriuria is a bacterial culture which identifies the actual organism. This test is more specific but can take several days to obtain a result. As a result, clinicians will often treat a bacteriuria based on the results of the urine dipstick test while waiting for the culture results. The culture will often provide antibiotic sensitivity.

Bacteriuria can be confirmed if a single bacterial species is isolated in a concentration greater than 100,000 colony forming units per millilitre of urine in clean-catch midstream urine specimens (one for men, two consecutive specimens with the same bacterium for women). For urine collected via bladder catheterization in men and women, a single urine specimen with greater than 100,000 colony forming units of a single species per millilitre is considered diagnostic. The threshold is also 100 colony forming units of a single species per millilitre for women displaying UTI symptoms.

A determination of whether or not the person has dehydration is an important part of the assessment, with dehydration typically divided into mild (3–5%), moderate (6–9%), and severe (≥10%) cases. In children, the most accurate signs of moderate or severe dehydration are a prolonged capillary refill, poor skin turgor, and abnormal breathing. Other useful findings (when used in combination) include sunken eyes, decreased activity, a lack of tears, and a dry mouth. A normal urinary output and oral fluid intake is reassuring. Laboratory testing is of little clinical benefit in determining the degree of dehydration. Thus the use of urine testing or ultrasounds is generally not needed.

Analysis of the urine may show signs of urinary tract infection. Specifically, the presence of nitrite and white blood cells on a urine test strip in patients with typical symptoms are sufficient for the diagnosis of pyelonephritis, and are an indication for empirical treatment. Blood tests such as a complete blood count may show neutrophilia. Microbiological culture of the urine, with or without blood cultures and antibiotic sensitivity testing are useful for establishing a formal diagnosis, and are considered mandatory.