📄 medicine

Temporal Trends, Microbiological Characteristics, and Factors Associated with Healthcare- Associated Infections in Hospitalized Neurological Patients: An 8-Year Retrospective Study

This 8-year retrospective study of 45,166 neurological patients at Deyang People's Hospital reveals a significant decline in healthcare-associated infections from 3.5% to 1.1%, identifies lower respiratory and urinary tract infections caused by Gram-negative bacteria as primary challenges, and proposes an exploratory risk-estimation nomogram based on factors like age, comorbidities, and calendar year that requires further validation before clinical use.

Original authors: Ying Yang, Shuai Jiang, Shiru Xiao, Li Luo, Ting Zhang, Xiaoyun Jiang, Yangyun Han

Published 2026-06-28

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Ying Yang, Shuai Jiang, Shiru Xiao, Li Luo, Ting Zhang, Xiaoyun Jiang, Yangyun Han

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine a hospital's neurological ward as a busy, high-stakes airport terminal. The passengers are patients with brain and nerve conditions. Because their "flight plans" (their health) are already complicated—some can't speak clearly, some can't move, and some need special equipment to breathe or use the bathroom—they are more vulnerable to getting "lost luggage" (infections) during their stay.

This study looked at the baggage claims of 45,166 passengers over an 8-year period (2016 to 2023) at one specific hospital in China. The researchers wanted to answer three questions:

Is the rate of lost luggage going up or down?
What kind of "germs" are causing the trouble?
Which passengers are most likely to get their luggage mixed up?

Here is the story of what they found, broken down simply:

1. The Trend: A Clearing Storm

Think of the infection rate as a storm cloud. In 2016, the storm was heavy, with 3.5% of patients getting an infection. By 2023, the storm had mostly cleared, dropping to just 1.1%.

Why? The researchers believe this is because the hospital staff got better at cleaning, washing hands, and managing patients over time. It's like the airport staff learned better ways to keep the terminal clean and safe, even during the global "pandemic" period when everyone was extra careful.

2. The Culprits: Where and What?

Even though the storm got smaller, the "bad weather" wasn't spread evenly.

The Location: Almost 9 out of 10 infections happened in just two places: the lungs (breathing) and the bladder (peeing). This makes sense because many neurological patients have trouble clearing their throats or moving their bodies, making these areas easy targets for germs.
The Villains: The main bad guys were Gram-negative bacteria (a specific type of germ). The top two "captains" of this crew were Klebsiella pneumoniae (often found in the lungs) and E. coli (often found in the bladder).
The Super-Villains: A significant portion of these germs were "super-villains" known as Multidrug-Resistant Organisms (MDROs). These are germs that have learned to dodge the usual antibiotics, making them harder to defeat.

3. The Risk Factors: Who is Most Vulnerable?

The researchers built a "risk map" (a mathematical tool called a nomogram) to see which passengers were most likely to get infected. They found that the following factors made a patient much more likely to catch an infection:

Age: Older passengers were at higher risk. It's like an older car needing more maintenance; older bodies are often less able to fight off new germs.
Smoking: Smokers were much more likely to get infected. Think of smoking as rusting the air filters in a car; it makes the lungs less able to defend themselves.
Diabetes: This was the biggest risk factor. Having diabetes is like having a security system that is turned down; the body's white blood cells (the security guards) don't work as well when sugar levels are high.
History of Epilepsy: Patients with a history of seizures were at higher risk. The researchers suggest this might be because these patients often have more complex medical needs or take more medications, making them more vulnerable.
The Year: Interestingly, the year itself mattered. Being treated in 2023 was much safer than being treated in 2016, regardless of who the patient was. This confirms that the hospital's overall safety practices improved over time.

4. What About Alcohol?

The study looked at alcohol use, too. At first, it looked like drinkers were at higher risk. However, once the researchers accounted for smoking and other factors, the link to alcohol disappeared. It's like seeing that people who carry umbrellas also get wet; it's not the umbrella causing the wetness, but the rain. In this case, alcohol use seemed to be a "marker" for other risky behaviors (like smoking) rather than a direct cause of the infection itself.

5. The "Risk Map" (The Nomogram)

The team created a visual tool (a nomogram) that doctors could theoretically use to guess a patient's risk of infection based on their age, smoking status, diabetes, epilepsy history, and type of brain disease.

The Catch: While this map looked very accurate inside the study, the authors warn that it's like a weather forecast made only for one specific city. It hasn't been tested on other hospitals or in the future yet. So, while it's a helpful starting point, it's not a crystal ball that should be used to make final medical decisions just yet.

The Bottom Line

Over 8 years, the hospital got much better at preventing infections in neurological patients. However, the battle isn't over. The main enemies remain lung and bladder infections caused by tough, drug-resistant bacteria.

The study tells us that to keep patients safe, hospitals need to:

Focus extra care on the lungs and bladder.
Watch out for older patients, smokers, and those with diabetes.
Keep improving their cleaning and safety habits, because those habits clearly work.

Note: This study is a look back at past data from one hospital. It tells us what happened, but it doesn't prove that changing one specific thing (like stopping smoking) will definitely prevent an infection in every single future case.

Technical Summary: Temporal Trends, Microbiological Characteristics, and Factors Associated with Healthcare-Associated Infections in Hospitalized Neurological Patients

Problem Statement
Healthcare-associated infections (HAIs) represent a critical patient safety challenge, particularly for neurological inpatients who are vulnerable due to impaired consciousness, airway dysfunction, immobility, and the frequent need for invasive procedures. While existing literature has explored HAIs in neurological settings, significant gaps remain. Previous studies are often limited by small sample sizes, short observation periods, single-center designs, or a focus on specific conditions (e.g., traumatic brain injury) or infection types (e.g., ventilator-associated pneumonia) without integrated analysis of the full disease spectrum, pathogen distribution (including multidrug-resistant organisms), and temporal trends. Furthermore, it remains unclear whether HAI incidence among hospitalized neurological patients has genuinely declined in recent years using large-scale real-world data. This study addresses these gaps by conducting an 8-year retrospective analysis to characterize temporal trends, microbiological profiles, and risk factors, while also exploring the development of a risk-estimation tool.

Methodology

Study Design and Population: A retrospective observational study was conducted at the Center for Neurological Diseases of Deyang People's Hospital (Southwest China) involving 45,166 eligible admissions to the Departments of Neurology and Neurosurgery between January 2016 and December 2023.
Data Source: Data were retrieved from the Hospital Infection Monitoring System, linked to the hospital information system (HIS), laboratory information system (LIS), nursing, and anesthesia records.
Definitions: HAIs were identified based on the 2001 Chinese Ministry of Health "Diagnostic Criteria for Nosocomial Infections (Trial)," with infections occurring >48 hours post-admission classified as HAIs. Multidrug-resistant organisms (MDROs) were defined as isolates non-susceptible to at least one agent in three or more antimicrobial categories.
Statistical Analysis:
- Descriptive: Annual HAI incidence, infection sites, and microbiological findings were summarized.
- Multivariable Analysis: Logistic regression was used to identify factors associated with HAI occurrence. Variables included age, sex, smoking, alcohol use, diabetes, epilepsy history, neurological disease category, and calendar year. Variables representing infection manifestations (e.g., fever, length of stay) were excluded from the primary predictive model to avoid reverse causation.
- Risk Modeling: An exploratory nomogram for in-hospital HAI risk estimation was developed using five clinically interpretable variables (age, smoking, diabetes, epilepsy history, neurological disease category). The model was internally validated using a 7:3 training/validation split, with discrimination assessed via the Area Under the Receiver Operating Characteristic Curve (AUC).
- Sensitivity: Firth penalized logistic regression was performed to evaluate stability against sparse-data bias.

Key Results

Temporal Trends: The overall HAI incidence was 2.3% (1,046 cases). A significant downward trend was observed, with incidence decreasing from 3.5% in 2016 to 1.1% in 2023 ( $P < 0.001$ ).
Infection Sites: The burden was concentrated in the lower respiratory tract (72.9%) and urinary tract (15.9%).
Microbiological Characteristics: Gram-negative bacteria (GNB) predominated (72.3% of isolates), with Klebsiella pneumoniae (27.4%) and Escherichia coli (25.6%) as leading pathogens. K. pneumoniae was most common in respiratory specimens, while E. coli dominated urine specimens. MDROs included ESBL-producing E. coli (36.8% of resistant isolates), ESBL-producing K. pneumoniae (23.0%), and Methicillin-resistant Staphylococcus aureus (17.7%).
Associated Factors: Multivariable analysis identified the following as independently associated with HAI occurrence:
- Older Age: Patients $\ge$ 80 years had significantly higher odds of HAI compared to those <45 years (aOR 2.92).
- Comorbidities: Diabetes (aOR 45.65) and a history of epilepsy (aOR 4.44) were strong risk factors.
- Lifestyle: Smoking was associated with higher odds (aOR 4.16), whereas the association for alcohol drinking was attenuated and non-significant after adjustment.
- Neurological Disease Category: Using neoplasms as a reference, traumatic injuries were associated with lower odds (aOR 0.39), while vascular disorders showed no significant association.
- Calendar Year: All subsequent years showed lower odds compared to 2016, with 2023 showing the lowest risk (aOR 0.15), reflecting secular improvements in care.
Nomogram Performance: The exploratory nomogram demonstrated high apparent discrimination, with an AUC of 0.909 in the training set and 0.925 in the internal validation set.

Key Contributions

Longitudinal Surveillance: Provided an 8-year, large-scale (45,166 admissions) real-world assessment of HAI trends in a tertiary neurological center in China, confirming a significant decline in incidence.
Integrated Microbiological Profile: Detailed the site-specific distribution of pathogens and MDROs, highlighting the dominance of Gram-negative organisms and the specific prevalence of ESBL producers and carbapenem-resistant strains in neurological patients.
Risk Factor Identification: Identified specific patient-level vulnerabilities (age, diabetes, epilepsy, smoking) and disease categories associated with HAI risk, distinguishing them from infection-related manifestations.
Exploratory Prediction Tool: Developed and internally validated a nomogram for estimating in-hospital HAI risk based on readily available clinical variables, offering a potential framework for targeted surveillance.

Significance and Claims
The authors claim that their findings support the prioritization of respiratory and urinary infection prevention strategies in neurological wards, particularly for clinically vulnerable subgroups (older patients, those with diabetes or epilepsy). The study emphasizes that HAI risk is shaped not only by host factors but also by the evolving infection-control environment, as evidenced by the strong temporal trend.

However, the paper maintains a modest stance regarding its conclusions:

Causality: The authors explicitly state that the findings represent associations rather than causal effects due to the retrospective design.
Generalizability: The single-center nature of the study limits external generalizability.
Nomogram Utility: The nomogram is presented as an "exploratory" tool. The authors caution that its high internal discrimination may be overestimated due to the lack of external or temporal validation and the potential for predictors to reflect in-hospital care trajectories rather than pre-existing risk. They assert that the tool requires further validation before clinical application.
Limitations: The study acknowledges limitations including missing data for BMI, the inability to time-stamp invasive procedures relative to infection onset (leading to their exclusion from the model), and the lack of direct measures for neurological severity or glycaemic control.

In conclusion, the study provides locally actionable evidence for infection prevention in neurological settings but calls for future multicenter studies with direct severity measures and external validation to strengthen causal interpretation and generalizability.