The Incidence of Community-Acquired Pneumonia in the Elderly Population of the United States of America: A Systematic Review and Meta-Analysis

Current epidemiological data reports that adults aged 65 years and older comprise the most vulnerable age group with the highest proportion of CAP-attributable hospitalizations. Pneumococcal vaccine efficacy (VE) has been shown to decrease over time, contributing to increasing incidence rates of CAP. A holistic evaluation of age, sex, seasonality, and VE are is conducted in this systematic review and meta-analysis of 12 prospective and retrospective cohort studies. The findings suggest that incidence and age are positively associated and that incidence in females is more often reported to be higher in females than in males. In studies that observed seasonality of CAP, high seasons and low seasons were reported to be in winter and summer months, respectively. Lastly, studies that reviewed the effect of vaccination on incidence consistently found decreased observance of CAP in elderly adults following reception of PCV13 or PPSV23. However, one study suggested that such vaccinations may have decreased effectiveness in elderly populations and that research examining potential explanations for this require further investigation. Furthermore, distinct diagnostic and case ascertainment standards, descriptive measures, and methods of prevention and treatment of CAP used across the US are outlined in this review. Public health guidance such as encouraging the reception of pneumococcal vaccinations and mask-wearing during high seasons of CAP, and communicating the risks of not adhering to the aforementioned preventative measures can facilitate an effort to reduce the incidence of CAP and its associated adverse outcomes in the US elderly population.


Introduction
With more than 1.5 million unique American adults hospitalized for community-acquired pneumonia (CAP) annually, CAP is the leading cause of infectious disease-related death in the US (Ramirez et al., 2017). Current epidemiological data report that adults aged 65 years and older comprise the most vulnerable age group, accounting for 60% of hospitalizations attributable to pneumococcal pneumonia (Wroe et al., 2012). From 2004 to 2040, as the US population increases by 38%, it is projected that there will be a 96% increase in CAP hospitalizations (Wroe et al., 2012). CAP occurs in all age groups, however, is most fatal and prevalent in the elderly, who possess senescent immune systems and have a higher rate of comorbidity (Brown et al., 2018). In comparison to other existing reviews, this review focuses on CAP incidence specifically within the elderly population of the US. The aim of this review is to provide a critical evaluation of studies focused on the epidemiology and incidence of elderly CAP and to inform current and future public health interventions when planning prevention measures.

Population of interest
The population of interest in this review is elderly adults living in the US. Elderly adults will be defined as people aged 65 years or older, while any reference to non-elderly adults refers to people between 19 and 65 years of age. CAP hospitalizations for the elderly occurs at a notably higher rate compared to that for non-elderly adults (Ramirez et al., 2017). Increased utilization of prophylactic measures and standardization of laboratory and radiographic diagnostic criteria serve as options for the mediation of CAP and increasing accuracy of incidence estimates.

Case definition and criteria for CAP diagnosis
CAP is a lower respiratory tract infection that may be caused by the inhalation of a variety of pathogens including bacteria, viruses, and fungi in a non-clinical setting (Mayo Clinic, 2020).
Pneumonia is defined as an infection of the alveoli in either or both lungs, consequently causing the build-up of purulent material in the alveoli (American Lung Association [ALA], 2020). Such physiological changes limit the exchange of oxygen and carbon dioxide at the alveolar-capillary membrane, leading to physiological changes such as respiratory acidosis and symptoms including but not limited to shortness of breath, expectoration of phlegm, fever, and, in particular for the elderly, lowered body temperature, and changes in mental awareness (ALA, 2020; Mayo Clinic, 2020). This review focuses on geriatric cases that are clinically severe, defined as elderly patients who require hospitalization for infection treatment.

Meaningful measures of disease frequency
The studies included in this review utilize incidence as the most meaningful measure of disease frequency. Incidence will be defined here as the number of new CAP cases per 1,000 person-years.

METHODS
A systematic review and meta-analysis using PRISMA methodology was implemented to investigate the incidence of CAP amongst the US elderly population, using three electronic databases: Web of Science, PubMed, and Scopus.

Literature search
Systematic search methods were performed using Global Health, PubMed, and Scopus using MeSH terms as appropriate. Pilot examination of studies was carried out in order to identify key MeSH terms used in relevant literature. Search filters were not used when selecting studies to avoid the exclusion of potentially admissible studies. The terms utilized in literature searches are detailed in Table 1.

Scopus
(elderly OR senior OR geriatric) AND ("community-acquired pneumonia" OR pneumonitis) AND (epidemiolog* OR aetiolog OR incidence OR cross-section*) AND ("United States of America" OR American) were deemed eligible for inclusion. Figure 1 illustrates the methodology used for the identification, screening, and eligibility-determination of included publications.

Exclusion Criteria
Study exclusively reports hospital-acquired or ventilator-acquired pneumonia data Study does not define community-acquired pneumonia CAP incidence data not provided Study was deemed a systematic review or meta-analysis Study considers geographic locations that are not considered as part of the US Study did not provide CAP incidence data broken down by age band Data 35 years or older was used Studies that exclusively report CAP infections that did not require hospitalization Language of study is not in English Study does not include elderly-aged patients or focuses on foreign-born Americans Study includes non-human data

Overview of included studies
12 studies were deemed eligible and included in this review.

Diagnostic standards and case ascertainment
Out The studies included in this review used symptomatic abnormalities, laboratory evidence, and radiographic approaches to diagnose cases of CAP. The methods used for case ascertainment vary between reports due to differing diagnostic criteria. Six studies reported cough, sputum production, fever, chills, hypothermia, chest pain, dyspnea, and tachypnea as potential  Table 4 provides an overview on the determinants for each of the included studies.

CAP Prophylaxis and Antibiotic Treatment
Current US strategies to prevent pneumonia among elderly adults include immunization with pneumococcal polysaccharide, conjugate, and Despite the fact that vaccination rates for both influenza and pneumococcal polysaccharide vaccines were high in the elderly study population (>75% and >70%, respectively), rates of CAP were high, especially among those aged ≥ 85 years; the study reported a correlation between influenza virus circulation and CAP incidence

Isturiz 2019 (17)
Only patients who met study eligibility criteria, had a final diagnosis of CAP confirmed by chest X-rays, and did not receive pneumococcal vaccination within 30-days of study enrollment were included in the primary analysis population; a decline in PCV13 serotypes was reported in the study, which may reflect the decreased effectiveness of vaccination among elderly adults 20

McLaughlin 2015 (18)
At the time of the study, only PPSV23 was approved for routine use in adults; the study reported a significant burden of CAP in the veteran population (median age 65 years) despite VHA patients having better PPSV23 coverage than the general, non-VA population

Quality of studies
Depending on the design and methods of a particular study, different limitations to reported results may exist. In an effort to account for such limitations, Table 6 summarizes key limitations, biases (if present), and strengths of the studies included in this review. To conduct an evaluation of PCV13 vaccine effectiveness against hospitalized vaccine-type CAP in the US in adults aged ≥65 years The study population is restricted to elderly adults residing in Louisville, Kentucky, limiting generalizability; cohort limited to those who provided their pneumococcal vaccination history Aspects of the study rely on observational data and thus is not immune to selection bias The test-negative method is good for evaluating vaccine effectiveness against infectious respiratory diseases; laboratory and radiographic data were used to define CAP

Kollef 2005 (16)
To characterize the microbiology and outcomes among patients with culturepositive CAP, HCAP, HAP, and VAP A cohort of 4,543 patients with culturepositive pneumonia were included in the study -underestimation of incidence as blood cultures show poor sensitivity in pneumonia: results often present false negatives (Gamache, 2020) Diagnostic methods for CAP are limited to laboratory findings without inclusion of radiographic confirmation

Incidence
Amongst the 12 included studies, eight reported incidence estimates for CAP in the US elderly population ( elderly per year. Incidence estimates depended largely on how cohort age bands were specified by their respective authors. In addition to the variation caused by differing age band specifications between studies, as summarized in Table 7, estimates varied notably with regards to sex, seasonality, and vaccination history.

Age
Age constitutes an established risk factor for CAP infections in the elderly. Figure 3 compares the incidences reported for elderly cohorts within varying age bands. The median and mean incidences for elderly patients between the ages of 65 to 74 were 8. comparisons between studies with regards to age are limited.

Sex
Differences between incidence estimates for males and females were reported in five of the 12 included studies detailed in

Seasonality
Winter months constitute the high season for CAP infections, whereas fewer events are presented during the summer months (Jackson et al., 2006

Synthesis of results and discussion of findings
In this review, the incidence of CAP among elderly adults in the US was examined. The 12 included studies suggested that incidence and age are positively associated and that incidence in females was reported to be higher more often in females than in males with one study reporting the opposite findings. In studies that observed seasonality of CAP, high seasons and low seasons were reported to be in winter and summer months, respectively. Lastly, studies that reviewed the effect of vaccination on incidence consistently found decreased observance of CAP in elderly adults following reception of PCV13 or PPSV23. However, one study suggested that such vaccinations may have decreased effectiveness in elderly populations and that research examining potential explanations for this require further investigation.

Limitations of this review and meta-analysis
Although sample sizes were sufficiently large, some cohorts consisted of patients limited to one or multiple cities within a specified geographic region of the US, which limits generalizability of the results. Additional limitations to generalizability include limiting cohort sampling to acute care hospitals and excluding patients from federal and long-term care hospitals, limiting CAP incidence estimates to an uncomprehensive range of causative pathogens rather than considering the array of causative agents, and a lack of consideration of race and ethnicity in particular studies when determining incidence.
A lack of uniformity of diagnostic criteria methodologies creates challenges for comparing studies. Some studies limit CAP diagnosis to laboratory findings while others require both laboratory and radiographic evidence to diagnose cases. Radiographic confirmation of laboratory results is the standard procedure used by clinicians to diagnose CAP and a lack of such evidence can lead to overestimation or underestimation if laboratory results provide false positive or false negative data, respectively. Furthermore, because incidence data from over two decades ago was used by three of the included studies, changes in ICD-9-CM coding and less advanced diagnostic methods during their respective study periods likely led to less precise incidence estimates.
In addition to the aforementioned limitations, inconsistency of age band designations between the included studies presented additional challenges in comparing age-specific incidence estimates.

Implications of findings
The limitations identified in this review indicate a need for uniform utilization of age band designations and diagnostic standards when providing epidemiological estimates for a particular disease. Additionally, if researchers seek to provide national estimates of incidence, generalizability, in terms of geographic breadth of study, the inclusion of different types of relevant hospitals treating the disease of interest, and consideration of relevant demographic data and medical history should be included to maximize external validity. Beginning with increased application of vaccinations and standardization of reliable laboratory and radiographic diagnostic criteria, the US must tend to the healthcare needs of its growing elderly population.

Conclusions
This review and meta-analysis provides evidence of increasing incidence of CAP with progressing age in the US elderly population. Additionally, it presents an argument for differences in gender-specific incidence of CAP as more studies found higher rates in females than in males. After adjusting for BMI and immunocompromised status, the included studies demonstrate the effectiveness of pneumococcal vaccinations, highlighting a need to ensure that elderly patients establish and maintain their immunity. Due to findings of decreasing vaccination serotypes in elderly patients over time, booster shot administration is important to decrease the likelihood of adverse outcomes. This review delineates that the elderly population is particularly vulnerable to CAP during the winter months, and so mask wearing policies and patient recommendations should be considered to initiate protective behavior changes. Lastly, wide-ranging diagnostic and case ascertainment standards of CAP across the US signal a need for nation-wide standardization so that methods of reporting CAP incidence are consistent.