Rehospitalization for pneumonia after first pneumonia admission: Incidence and predictors in a population-based cohort study.

BACKGROUND AND OBJECTIVES: Hospital readmissions are a frequent complication of pneumonia. Most data regarding readmissions are obtained from the United States, whereas few data are available from the European healthcare utilization (HCU) systems. In a large cohort of Italian patients with a previous hospitalization for pneumonia, our aim was to evaluate the incidence and predictors of early readmissions due to pneumonia.
METHODS: This is a observational retrospective, population based, cohort study. Data were retrieved from the HCU databases of the Italian Lombardy region. 203,768 patients were hospitalized for pneumonia between 2003 and 2012. The outcome was the first rehospitalization for pneumonia. The patients were followed up after the index hospital admission to estimate the hazard ratio, and relative 95% confidence interval, of the outcome associated with the risk factors that we had identified.
RESULTS: 7,275 patients (3.6%) had an early pneumonia readmission. Male gender, age ≥70 years, length of stay of the first admission and a higher burden of comorbidities were significantly associated with the outcome. Chronic use of antidepressants, antiarrhythmics, glucocorticoids and drugs for obstructive airway diseases were also more frequently prescribed in patients requiring rehospitalization. Previous use of inhaled broncodilators, including both beta2-agonists and anticholinergics, but not inhaled steroids, were associated with an increased risk of hospital readmission.
CONCLUSIONS: Frail elderly patients with multiple comorbidities and complex drug regimens were at higher risk of early rehospitalization and, thus, may require closer follow-up and prevention strategies.

Introduction

Pneumonia is one of the more frequent and potentially serious infectious diseases, being a leading cause of hospitalization worldwide. In the last decades pneumonia hospital admissions have increased by 25–50% in the US and European countries [1,2], particularly in the elderly, who have a 10-fold increased incidence [1,3]. Pneumonia has also been shown to increase short- and long-term mortality [1], mostly affecting frail populations, including elderly and patients with multiple comorbidities [4], substantially heightening the disease burden among adults overall and, thus, becoming one of the main topics in public health.

Age, gender, type of pneumonia, chronic comorbidities and severity of the disease have all been associated with increased short- and long-term mortality in patients with pneumonia [1], especially because pneumonia is a trigger for cardiovascular events and for respiratory exacerbations in predisposed patients [1].

Early hospital readmission (within 30 days) complication after a first hospitalization for pneumonia, with an incidence ranging from 11.8 to 20.8% [5]. Recurring pneumonia, cardiovascular disease and exacerbations of chronic pulmonary diseases are the most common causes of early readmission [6]. Instability factors upon discharge, such as persistent fever and respiratory failure, hospitalization in the previous 90 days and an elevated number of decompensated comorbidities, have all been associated with rehospitalization [7,8]. Nevertheless, also the complexity of medication regimens is a possible predictive factor for readmission in patients affected by pneumonia [9].

Most data concerning readmissions and risk factors come from the United States, because hospitals with higher-than-expected risk adjusted 30-day readmission rates have to deal with major financial penalties [10] imposed by Medicare and Medicaid services since the Hospital Readmission Reduction Program has been implemented in 2012.

With these premises, a large real-world retrospective cohort study was conducted using administrative databases to estimate, in patients who had been previously hospitalized for pneumonia, the incidence of early (within 30 days) pneumonia readmissions and to investigate the predisposing factors for rehospitalizations.

Materials and methods

Setting

The present study is based on computerized Healthcare Utilization (HCU) databases of Lombardy, an Italian northern region accounting for almost 10 million people (about 16% of Italy’s whole population. In Italy, the National Health Service (NHS) covers the entire resident population, and since 1997 its management in Lombardy has been associated with an automated system of HCU databases that collects and stores a variety of information, including (i) demographic and clinical data on residents who receive NHS assistance, (ii) diagnosis at discharge from public or private hospitals, (iii) reimbursable drug prescriptions dispensed outpatiently or directly administered in hospital, (iv) exemptions from healthcare co-payment for chronic diseases, and (v) outpatient visits, including specialist visits and diagnostic exams reimbursable by the NHS [11].

A detailed description of the HCU databases of Lombardy for studying the framework of respiratory diseases is available in other previous studies [12-15]. The ICD-9-CM, Anatomical-Therapeutic-Chemical (ATC) and outpatient procedure codes used in the current study are shown in .

Cohort selection

Our study was designed according to the procedure shown in . All 266,766 NHS-eligible residents in Lombardy who had experienced at least one hospital admission with pneumonia as primary or secondary diagnosis (ICD-9-CM code from 480.x to 488.x, 487.x excluded, 4870.x included) during the years 2003 to 2012, were identified. The date of the first hospital discharge was considered as the index date. Four categories of patients were excluded: (i) 50,187 patients aged ≤18 years at the index date; (ii) 2,230 patients who were beneficiaries of the NHS from less than <6 months before the index date; (iii) 438 patients with at least one hospitalization for pneumonia in the 6 months before the index date (in order to include only those patients for whom the index admission was not a readmission after previous pneumonia hospitalization); and (iv) 10,143 patients treated cumulatively with antibiotic, antiviral or antifungal medications for at least 2 months before the index date. The remaining 203,768 patients were included in the study cohort that accumulated person-years of follow-up from the index date until the earliest date among the occurrence of the outcome (see below) or censoring (occurring at death, emigration, or the 90th day after index date).

Flow-chart of inclusion and exclusion criteria for the study cohort selection.

Lombardy region, Italy, 2003–2012. NHS, national health system.

Outcome

The outcome was the first rehospitalization for pneumonia (ICD-9-CM code from 480.x to 488.x, 487.x excluded, 4870.x included) that occurred within 30 days of discharge after the first pneumonia admission; the so-called early readmissions for pneumonia.

Covariates

The baseline characteristics of cohort members (i.e., those recorded at the index date or during the previous 6 months) included gender, age, length of stay during the index hospitalization, prescriptions for antibiotic, antifungal or antiviral drugs at discharge, previous admissions for lung cancer or chronic lung disease and previous prescriptions for other drugs (antidepressants, antiarrhythmics, antithrombotics, antihypertensives, inhaled steroid drugs and bronchodilators, nonsteroidal anti-inflammatory drugs (NSAIDs), digoxin, statins and benzodiazepines). Some of these latter medications can be considered as proxies for underlying chronic medical conditions.

Other clinical and therapeutic characteristics were also measured during the index hospital stay: venous thromboembolism, thrombocytosis, Clostridium difficile infection, pressure ulcer, lung procedure, respiratory failure, mechanical ventilation use, ≥1 unit of blood transfusion.

Finally, we used the Multisource Comorbidity Score (MCS), a new comorbidity index obtained from inpatient diagnostic information and outpatient drug prescriptions, validated using data from Lombardy and other regions of Italy [11]. Patients were categorized as having low (0–9) or high (≥10) MCS score.

Data analysis

Demographic data, clinical characteristics and therapeutic regimens were compared among patients who had a pneumonia readmission and those who did not. A cohort design was implemented and a Cox proportional hazard model was fitted to estimate hazard ratios (HR), and relative 95% confidence intervals (CI), for the association between the selected potential predisposing factors and the risk of early readmission for pneumonia. The estimates were adjusted by those covariates listed in the “Covariates” section. The effect of the predisposing factors on the risk of a pneumonia rehospitalization was evaluated both for the entire cohort and, separately, for age classes (<70 and ≥70 years).

Finally, the robustness of estimates regarding potential bias introduced by unmeasured confounders (e.g., smoking status) was investigated by using the rule-out approach described by Schneeweiss [16]. This approach involves detecting the amount of the overall confounding required to fully account for the exposure–outcome association, thus moving the observed point estimate to null.

The Statistical Analysis System software (version 9.4; SAS Institute, Cary, North Carolina, USA) was used to perform the analyses. For all hypotheses tested, two-tailed p-values <0.05 were considered to be significant.

Ethical approval and consent to participate

According to the rules from the Italian Medicines Agency (available at: http://www.agenziafarmaco.gov.it/sites/default/files/det_20marzo2008.pdf), retrospective studies using administrative databases do not require Ethics Committee protocol approval. Furthermore, according to General Authorization for the Processing of Personal Data for Scientific Research Purposes issued by the Italian Privacy Authority on August 10, 2018 (available at: https://www.garanteprivacy.it/web/guest/home/docweb/-/docweb-display/docweb/9124510) this study was exempt from informed consent. In order to protect privacy, and to guarantee individual records anonymity, after the record-linkage between HCU databases and the data extraction procedure, the individual identification codes were automatically converted into anonymous by the regional IT technicians, so that researchers had access to full anonymized data.

Results

Study population

The 203,768 cohort members accumulated 44,226 person-years of follow-up, on average 0.22 years per patient. During this period 33,746 patients (16.6%) experienced a 30-day all-cause hospital readmission, whereas 7,275 patients (3.6%) required an early pneumonia readmission (with an incidence rate of 16.4 cases every 100 person-years).

Demographic data, clinical characteristics and therapeutic regimens of patients undergoing a first hospitalization for pneumonia between the years 2003 and 2012 in the whole cohort, and according to whether or not patients experienced the outcome of interest, are summarised in Table 1. Patients who required rehospitalization were more frequently male, with a longer length of hospital stay during the first pneumonia admission and a higher burden of comorbidities, measured by the MCS ().

Table 1

Baseline demographic and clinical characteristics and therapeutic regimens of the 203,768 patients considered in the study, in the whole cohort and according to whether they experienced the outcome of interest.

Lombardy, Italy, 2003–2012.

	Whole cohort (n = 203,768)	Pneumonia rehospitalization (n = 7,275)	No pneumonia rehospitalization (n = 196,493)	P value
Demographics
Male (%)	112,698 (55.3)	4,509 (62.0)	108,189 (55.1)	< .0001
Age, years: mean (SD)	71.2 (16.3)	71.6 (16.2)	71.2 (16.3)	0.3486
Length of stay (LOS): mean (SD)	15.4 (16.2)	16.5 (16.9)	15.4 (16.1)	< .0001
Multisource Comorbidity Score α (%)
Low	139,732 (68.6)	4,579 (62.9)	135,153 (68.8)	< .0001
High	64,036 (31.4)	2,696 (37.1)	61,340 (31.2)
Previous diseases (%)
CV disease	115,118 (56.5)	4,305 (59.2)	110,813 (56.4)	< .0001
Cerebrovascular disease	27,034 (13.3)	1,015 (14.0)	26,019 (13.2)	0.0795
Liver disease	9,154 (4.5)	343 (4.7)	8,811 (4.5)	0.3510
Renal disease	14,765 (7.3)	680 (9.4)	14,085 (7.2)	< .0001
Cancer	34,263 (16.8)	1,543 (21.2)	32,720 (16.7)	< .0001
Lung cancer	4,170 (12.2)	175 (2.4)	3,995 (2.0)	0.0276
Chronic lung disease	31,369 (15.4)	1,205 (16.6)	30,164 (15.4)	0.0049
Previous medications (%)
Drugs for obstructive airway diseases	46,239 (22.7)	1,781 (24.5)	44,458 (22.6)	0.0002
Inhaled steroids	35,584 (17.5)	1,388 (19.1)	34,196 (17.4)	0.0002
Inhaled broncodilators β	29,012 (14.2)	1,186 (16.3)	27,826 (14.16)	< .0001
Persistent aspirin therapy	1,972 (1.0)	91 (1.3)	1,881 (1.0)	0.0120
Persistent NSAID therapy γ	4,889 (2.4)	178 (2.5)	4,711 (2.4)	0.7877
Persistent glucocorticoid therapy	4,037 (2.0)	184 (2.5)	3,853 (2.0)	0.0006
Antidepressants	26,374 (12.9)	1,021 (14.0)	25,353 (12.9)	0.0047
Antiarrhythmics	15,028 (7.4)	647 (8.9)	14,381 (7. 3)	< .0001
Statins	28,751 (14.1)	999 (13.7)	27,752 (14.1)	0.3460
Antihypertensives	120,543 (59.2)	4,348 (59.8)	116,195 (59.1)	0.2816
During hospital stay (%)
Venous thromboembolism	882 (0.4)	32 (0.4)	850 (0.4)	0.9260
Clostridium difficile infection	2,031 (1.0)	93 (1.3)	1,938 (1.0)	0.0138
Pressure ulcer	2,371 (1.2)	115 (1.6)	2,256 (1.2)	0.0007
Chest / lung surgery §	17,190 (8.4)	772 (10.6)	16,418 (8.4)	< .0001
Respiratory failure	33,940 (16.7)	1,604 (22.1)	32,336 (16.5)	< .0001
Mechanical ventilation	11,879 (5.8)	698 (9.6)	11,181 (5.7)	< .0001
≥ 1 units blood transfusion	13,716 (6.7)	707 (9.7)	13,009 (6.6)	< .0001

Abbreviation: SD, standard deviation; CV, cardiovascular; NSAID, nonsteroidal anti-inflammatory drug

The clinical status was assessed by the Multisource Comorbidity Score (MCS) according to the hospital admission and the drugs prescribed in the six-months period before the index date. Two categories of clinical status (MCS) were considered: good (Low MCS: score < 10) and poor (High MCS: score ≥ 10).

Inhaled broncodilators: both beta2-agonists and anticholinergics.

§ Chest / lung surgery: all operations on the respiratory system, including excision of lung and bronchus, but excluding chest drainage and bronchoscopy.

Chronic renal and lung diseases, as well as previous cardiovascular events and history of cancer, were more frequent in patients rehospitalized for pneumonia compared to those who were not. Chronic use of medications—antidepressants, antiarrhythmics, inhaled drugs for obstructive airway diseases, including steroids and bronchodilators, and persistent systemic glucocorticoid therapy—were more frequently prescribed in patients requiring rehospitalization ().

Factors complicating the first hospitalization, such as respiratory failure and mechanical ventilation use, pressure ulcers and need of blood transfusions, were more frequent among patients requiring a readmission.

Risk factors for pneumonia rehospitalization

Possible risk factors for early pneumonia rehospitalizations are summarised in . Male gender, age ≥70 years, longer length of hospital stay during the first hospitalization and a MCS ≥10 were all associated with an increased risk of pneumonia rehospitalization. Previous medications—inhaled bronchodilators, including both beta2-agonists and anticholinergics—were associated with increased risk of hospital readmission, whereas this association was not observed for inhaled steroids, either alone or in association with broncodilators. Chronic aspirin and systemic glucocorticoid therapy were also risk factors for pneumonia readmissions, but NSAID therapy was not. As for medical history, renal and chronic lung disease were associated with an increased risk of pneumonia rehospitalization.

Risk factors for pneumonia rehospitalization, stratified by age

When stratifying the analysis by age classes (≥70 years vs. <70 years), similar risk factors were observed, with the exception of complications during hospital stay (). C. difficile infection was associated with increased risk of readmission for pneumonia only in elderly patients. A need for blood transfusion appeared to be a stronger risk factor in younger patients.

Age-stratified Hazard ratios (HR), and relative 95% confidence intervals (CI), of early rehospitalizations for pneumonia associated with selected risk factors, estimated with a multivariable Cox proportional hazard model.

Lombardy region, Italy, 2003–2012. Estimates, stratified for age classes (<70 and ≥70 years), were obtained through a multivariable Cox proportional hazard model. For each covariate of which HR of the outcome was reported in the figure, the reference category was that of unexposed patients to this covariate (e.g., for persistent aspirin therapy the reference category was that of patients who was not exposed to a persistent aspirin therapy). NSAIDs, nonsteroidal anti-inflammatory drugs.

Sensitivity analysis

The robustness of estimates regarding bias introduced by an unmeasured confounder, like smoking status, was evaluated using the rule-out approach (). We performed this sensitivity analysis taking into account two of those predisposing factors that were found to have a significant association with the outcome, as well as being associated with the unmeasured confounder (i.e., respiratory failure and mechanical ventilation use during the index hospitalization). The rule-out approach was applied to evaluate if the observed harmful effect of those two risk factors was overinflated by the unmeasured confounder. For example, patients with a respiratory failure during the index hospitalization had a 3-fold smokers’ prevalence than those who had not a respiratory failure (exposure-confounder odds ratio, OREC, = 3). In order to nullify the observed harmful effect of the respiratory failure, smoking status should increase the risk of experiencing an hospital readmission for pneumonia by 3-fold (confounder-outcome relative risk, RRCD, = 3). On the other hand, admitting that smoking increases the risk of the outcome by 2-fold, prevalence of smokers among patients who experienced a respiratory failure should be 6-fold higher than those who did not, in order to nullify the observed effect.

Influence of smoking status as an unmeasured confounder on the association between respiratory failure (or mechanical ventilation use) during the index hospital stay and the risk of experiencing a pneumonia readmission.

Lombardy region, Italy, 2003–2012. The graph indicates the combinations of confounder-outcome and exposure-confounder associations that would be required to move the observed effect of the considered risk factors towards the null. We set the possible generic unmeasured confounder, e.g., smoking status: (i) to have a 21% prevalence of exposure among patients hospitalized for pneumonia [17], (ii) to increase the risk of the outcome onset up to 10-fold more in patients exposed to the confounder than in those not exposed, and (iii) to be up to 10-fold more common among patients exposed to the predisposing factor than in those not exposed.

Similar results and interpretation, if not even more marked and less likely, can also be observed considering the mechanical ventilation use.

Discussion

In the cohort of 203,768 adult patients hospitalized for pneumonia in the northern Italian region of Lombardy between 2003 and 2012, the majority were elderly, with a high burden of comorbidities (particularly cardiovascular events) and on treatment with multiple medications.

Out of these 203,768 patients, 33,746 (16.6%) experienced an all-cause rehospitalization within 30 days after the index hospital discharge, meaning that one in six patients hospitalized for pneumonia was readmitted for any cause within 30 days of discharge, according to the results reported in several observational studies. In literature, all-cause 30-day readmission rates after a first pneumonia hospitalization ranged from 8.6 to 20.8% [8,18], whereas many studies that have focused on the Medicare population showed all-cause 30-day readmission rate of 17%-25% [19,20].

Furthermore, in the present study pneumonia-related hospital readmissions accounted for about 22% of total 30-day readmissions, e.g., 7,275 patients (3.6%) required an early readmission (within 30 days of hospital discharge) for a new case of pneumonia.

In a multivariate analysis, the main risk factors for early rehospitalization were male gender, older age, higher burden of comorbidities and longer length of hospital stay during the first pneumonia hospitalization.

The most interesting finding of the present study was the overall readmission rate occurring in the first 30 days (3.6%), leading to the conclusion that patients in the first month are at risk for infectious complications and therefore may require a close follow-up. Most recent pneumonia guidelines do not suggest a follow-up of patients discharged from the hospital, with the exception of the British Thoracic Society guidelines, which suggest arranging a clinical review by the general practitioner or in a hospital clinic at ~6 weeks for all patients [21]. According to our results, a timely follow-up within 4 weeks of hospital discharge especially among patients who can most benefit from this intervention (i.e., elderly patients with a high burden of comorbidities including chronic lung diseases) is suggested. The incidence of 30-day pneumonia readmissions differs between European countries and increases with age in all countries. In most cases, pneumonia affects already frail populations, including the elderly and those with multiple chronic conditions. Thus, rehospitalizations, including those for pneumonia, load an additional burden on these vulnerable population. Therefore, various interventions aimed at decreasing the risk of hospital readmissions by targeting transitional and territorial care and post-discharge care coordination are needed.

In our study, inhaled broncodilators were associated with an increased risk of hospital readmission, whereas inhaled steroids, either alone or in association with bronchodilators, were not associated to the outcome; however, only 15% of patients in our cohort had chronic lung diseases. In contrast, the association between the use of inhaled steroids and the risk of developing pneumonia has been described in patients with chronic obstructive pulmonary disease, particularly those receiving fluticasone, but it is still a matter of debate [22-24].

Although our results seem to support the safe use of inhaled steroids in association to the risk of rehospitalization for pneumonia in the general population, they may not apply to the subgroup of patients with specific clinical conditions, such as chronic lung disease.

Chronic aspirin therapy were also found to be risk factors for pneumonia readmissions, whereas this association was not observed for chronic NSAID therapy. Aspirin has been considered a possible add-on therapy in patients with pneumonia and significant coronary risk factors because of its preventive role in cardiovascular events [25]. These are common complications in patients with community-acquired pneumonia, particularly those with hypoxemia and high systemic inflammatory response [25]. However, NSAID therapy, including the use of aspirin, prior hospital referral for pneumonia has been associated with an increased number of pleuropulmonary complications, such as pleural empyema and lung cavitations [26,27]. The association found in our population between chronic aspirin use and pneumonia readmissions may be due to those delayed pleuropulmonary complications.

Chronic use of systemic steroids, as well as other immunosuppressive therapies not included in this study because of the small number of cases, is also a recognized risk factor for pneumonia [28].

Despite the high population of elderly patients in our study, we found no difference in risk factors for rehospitalization between younger and older patients, with the exception of C. difficile infection and a need for blood transfusions during the index hospital stay. In the elderly, an acute disease such as pneumonia can cause a loss of physiologic reserve [29] that may appear as various complications, such as C. difficile infection, in those who require prolonged antibiotic therapies [30].

Our study has several elements of strength: (i) the investigation was based on a very large unselected population, which was made possible because of the cost-free health care system in Italy, which covers virtually all citizens; (ii) the drug prescription database provided highly accurate data, because pharmacists are required to report prescriptions in detail to obtain reimbursement, and incorrect reports about the dispensed drugs have legal consequences [31]; (iii) our study is one of the firsts to describe the incidence and risk factors for rehospitalization due to pneumonia in a population of patients with a first hospitalization for pneumonia; (iv) we excluded patients who had been treated with antibiotic, antiviral or antifungal medications for at least 2 months before the first pneumonia hospitalization. By doing so, we ruled out patients with chronic infections that might have led to an overestimation of the outcome. (v) We excluded patients who experienced a hospitalization for pneumonia in the 6 months before the index date; with this cautionary criterium we ensured the inclusion in the cohort of patients for whom the index admission was not a rehospitalization after previous pneumonia hospital admission, avoiding a potential misclassification of the outcome. Finally, (vi) a sensitivity analysis for the presence of unmeasured confounding confirmed the robustness of findings provided by the main analysis.

Our investigation also has several limitations beyond those inherent the observational studies. A main limitation is that, because of privacy regulations, hospital records were not available for scrutiny, which means that the diagnostic validity of pneumonia could not be checked. Another limitation of our study is that, although the patients’ clinical status can be inferred (and the data adjusted for) from knowledge of hospitalizations, treatments for pneumonia and pulmonary diseases and assumption of antibiotic, antifungal, antiviral therapy and other drugs, information does not include blood pressure, fever, respiratory rate, new-onset confusion, severity of pneumonia, and other clinical variables. Thus, as with any observational investigation, our findings may be affected by unmeasured confounding factors. However, the sensitivity analysis we carried out suggest that it is unlikely that the results of the study could be confounded by factors not measurable in the HCU databases (e.g., smoking status).

Furthermore, the data analyzed in this study relate to several years ago and are not updated, for this reason our study may not have been able to include some changes that have recently occurred in the care of hospitalized patient for pneumonia. However, our all-case 30-day readmission rates after a first pneumonia hospitalization agree with those reported in recent observational studies, ranging from 8.6 to 25% [6,8,19,20], thus we can likely suppose that all-case 30-day readmission rates after a first pneumonia hospitalization remained stable.

These limitations notwithstanding, our investigation offers quantitative evidence that patients in the first 30 days after discharge are at higher risk of hospital readmission for pneumonia. Frail patients (especially the elderly, who have a higher burden of comorbidities and chronic use of medications) are at higher risk of rehospitalization and, thus, may require closer follow-up and prevention strategies.

Future perspectives should include a better characterisation of chronic medication use in correlation with the underlying disease to evaluate prescriptive appropriateness. A better knowledge of risk factors for pneumonia rehospitalization should guide large-scale prevention efforts.

ICD-9-CM diagnostic, Anatomical-Therapeutic-Chemical (ATC) medication and outpatient procedure codes considered in the current study.

Lombardy region, Italy, 2003–2012.

(DOC)

Click here for additional data file.

17 Feb 2020

PONE-D-19-32710

Rehospitalization for pneumonia after first pneumonia admission: Incidence and predictors in a population-based cohort study

PLOS ONE

Dear Dr. Faverio,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Thank you for your submission to PLOS ONE. You will note that both reviewers identified a number of strengths as well as a number of limitations to your paper. In most cases, the noted limitations relate to some inherent issue in the administrative data or coding; and while it is recognized that you will not be able to address these in your analysis, some further nuanced discussion of these issues is warranted. Further, you will see that both reviewers picked up the 90-days readmissions - where one reviewer thought further discussion was needed and the other thought that the 90-day readmission analysis should be dropped altogether. I will leave it to you and your co-authors to decide; however, if exploring 90-day readmissions had been part of your original objectives then I would encourage to consider keeping it but better explain why no additional analyses are presented.

We would appreciate receiving your revised manuscript by Apr 02 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Andrea Gruneir

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements:

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We noticed you have some minor occurrence(s) of overlapping text with the following previous publication(s), which needs to be addressed:

https://doi.org/10.1016/j.ejim.2017.09.023

https://doi.org/10.1002/pds.4206

https://doi.org/10.1016/j.ejim.2017.09.023

https://doi.org/10.1186/s12883-017-0796-3

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the Methods section. Further consideration is dependent on these concerns being addressed.

3. In the ethics statement in the manuscript and in the online submission form, please provide additional information about the patient records/samples used in your retrospective study. Specifically, please ensure that you have discussed whether all data/samples were fully anonymized before you accessed them and/or whether the IRB or ethics committee waived the requirement for informed consent. If patients provided informed written consent to have data/samples from their medical records used in research, please include this information.

4. Thank you for stating the following in the Competing Interests section:

"I have read the journal's policy and the authors of this manuscript have the following competing interests: G.C. received research support from the European Community (EC), the Italian Medicines Agency (AIFA), and the Italian Ministry of Education, Universities and Research (MIUR). He took part in a variety of projects that were funded by pharmaceutical companies (Novartis, GSK, Roche, AMGEN and BMS). He also received honoraria as a member of the Advisory Board of Roche. No other potential conflicts of interest were declared."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

5. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability.

Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized.

Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access.

We will update your Data Availability statement to reflect the information you provide in your cover letter.

6. Your ethics statement must appear in the Methods section of your manuscript. If your ethics statement is written in any section besides the Methods, please move it to the Methods section and delete it from any other section. Please also ensure that your ethics statement is included in your manuscript, as the ethics section of your online submission will not be published alongside your manuscript.

7. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Partly

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This study shows the factors associated with readmissions in a large cohort of patients with pneumonia. Also, this study adds evidence about complexity medication regimens as a possible predictive factor for readmission.

The follow-up of patients included early and late readmission (30 and 90 days). Unfortunately, the analysis of patients who had a late readmission were not presented. According to your study aimed, information should be provided regarding patient characteristics and associated factors with late readmissions.

The study data were retrieved from the health care utilization database of NHS. Community-acquired pneumonia and hospital-acquired pneumonia were included? Please, add the pneumonia criteria and ICD-9CM codes includes of index admission. Interestingly, perhaps you have include information about type of pneumonia and causative microorganism.

You need explain if a collinearity analysis was perform. Why did you not show the use of statins (Table 1)?

Tables, figures a references need a review (references duplicates nº 5 and 14)

The introduction and discussion needs a little information regarding frail early patients.

Reviewer #2: The authors present a retrospective review of a large cohort of patients admitted with pneumonia to multiple hospitals in Northern Italy between 2003 and 2012 in which they assess the rate of pneumonia-specific readmission and its contributing cofactors. As opposed to most other studies on pneumonia readmission rates, the authors focus on related, pneumonia-specific readmissions, rather than all-cause readmissions. This is a strength of the study. Limitations include the significant delay between data collection and manuscript.

Major critiques

The authors should explain why they feel it is relevant to consider readmissions that occur beyond 30 days in their analysis. Intuitively it seems that these so called “late readmissions” are unrelated to factors involved in the original case. Furthermore the rate of late readmissions is very low. Unless the authors can add further justification for including this in the paper I would suggested removal and focusing on the readmissions within 30 days which is the standard most other research in this area examines.

The data being analyzed in this study is currently 8 – 17 years old. Certainly temporal changes in the care of hospitalized patients have transpired, and the fact that the conclusions are based on old data should be clarified as a limitation in the discussion sections. Ideally the authors would frame their results against comparative readmission rates from more recent data and put their results into context. Unfortunately few studies exist with pneumonia specific readmission rates, but if the authors have access to all-cause 30 day readmission rates, they could make the case for whether temporal trends suggest increasing, decreasing or stable rates compared to the study period.

In Table 1, there appears to be significant differences in various comorbidities between readmitted and not readmitted patients; this is pointed out in line 15 on p 9. However when adjusted hazard ratios are calculated (Table 2) only renal disease appears to be a predictor. In the Results section (p10, line 4-7) the authors note that use of meds for chronic lung disease were associated with increased readmission risk, but chronic lung disease as a comorbidity was not significant. Aren't some of these meds in fact "proxies" for the underlying chronic medical conditions? The authors should discuss these dichotomies and the implications for the limitations of their data in the discussion section.

The paper would benefit from more careful proofreading and correction of grammatical errors, and some effort to improve the clarity of writing.

Data Table 1 lacks column headings, these should be added.

Minor critiques

Late readmissions are mentioned in the introduction but given the lack of clear clinical relevance and the exceedingly small number identified by the author, I think the paper would be more effective if this detail was omitted throughout the manuscript.

The statement about readmission penalties in the US is inaccurate in that it refers to the “private health insurance system.” In fact, 30-day readmission penalties in the US are imposed by Medicare which is a federally funded health insurance program for adults age 65 and older. It is not private insurance.

Why does p 7 line 16 state that info was collected about prior admissions for pneumonia when this group of patients was excluded according to the text and Figure 1?

I had difficulty linking a clinical relationship to pneumonia readmission rates and some of the factors on the list of clinical characteristics detailed on p7 line 20. Can the authors explain why they chose to include thrombocytosis and pressure ulcer as clinical covariates?

P9 line 5-8: More reason to exclude mention of the concept of late readmissions in this analysis. It adds little other than distracting from the main findings.

Figure 1 – graphic is blurred

Page 12 line 22, 23 – further explanation as to why inhaled bronchodilators were associate with increased risk of admission where as ICS-Bronchodilator combination was not

Figure 2 – no point estimate labeled for younger or older patients, therefor unable to visually compare point estimates between both groups

P13 line 1-6, the authors follow discussion of a finding that states that inhaled steroids were not a predictor of readmission with a reference to a study that found that inhaled steroids are associated with higher risk of developing pneumonia. Was this a study about pneumonia readmission rates or primary episodes of pneumonia? If the latter, then the link between the 2 thoughts does not logically connect. Consider rewriting or removing.

Since you have a separate paragraph addressing the relationship of systemic steroids and pneumonia readmissions (p13 line 17-18), consider removing mention of steroids in line 7 and just focus on ASA in this paragraph.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

20 May 2020

Reviewer #1:

This study shows the factors associated with readmissions in a large cohort of patients with pneumonia. Also, this study adds evidence about complexity medication regimens as a possible predictive factor for readmission.

The follow-up of patients included early and late readmission (30 and 90 days). Unfortunately, the analysis of patients who had a late readmission were not presented. According to your study aimed, information should be provided regarding patient characteristics and associated factors with late readmissions.

Reply: We thank the reviewer for this suggestion. As the reviewer said, in our study the follow-up of patients included as outcome the first hospitalization for pneumonia; these rehospitalizations included both early (within 30 days of discharge after the first pneumonia admission) and late (between 31 and 90 days after discharge) readmissions for pneumonia. Therefore, the aim of our study was to evaluate the incidence and predictors of early and late readmissions due to pneumonia.

However, from a first explorative analysis, as reported at page 9, lines 5-8 of the old version of our manuscript, it resulted that, from 31 to 90 days after the index date, only 47 patients (0.02%) experienced a late readmission for pneumonia. Thus, considering that the rate of late readmissions is very low, statistical analyses were carried out only for early readmissions. Statistical analyses were not carried out also for late readmissions for pneumonia because a lack of statistical power would have been observed, mainly due to the low number of events considered, which would not have allowed to highlight appreciable statistically significant differences in the predictor variables.

Then, considering the potential lack of statistical power that could have occurred analyzing late pneumonia readmissions, and also taking into consideration the various suggestions raised by the reviewer #2 who proposed not to include late readmissions in the main analyses, we decided to remove from the main analyses the readmissions that occurred beyond 30 days after the index date. We revised the paper focusing on the early readmissions (i.e., those within 30 days after the index discharge for pneumonia), being these early readmissions the standard that most other research in this area examined.

However, we mentioned late readmissions a few times in the manuscript to broaden the discussion and introduce potential future developments of this research.

The study data were retrieved from the health care utilization database of NHS. Community-acquired pneumonia and hospital-acquired pneumonia were included? Please, add the pneumonia criteria and ICD-9CM codes included of index admission. Interestingly, perhaps you have included information about type of pneumonia and causative microorganism.

Reply: As the Reviewer suggested, our study is based on HealthCare Utilization (HCU) databases on Lombardy region, and all the 266,766 residents in Lombardy region assisted by the NHS wo had experienced at least one hospital admission with primary or secondary diagnosis of pneumonia, during the years 2003-2012, were firstly identified to be included in the study. Hospitalizations with pneumonia as secondary diagnosis were also considered because both the primary and secondary diagnoses contributed decisively in the hospital admission decision. As diagnosis of pneumonia for the index hospitalization, we considered those that had been recorded with an ICD-9-CM code from 480.x to 488.x, excluding 487.x but including 4870.x.

We considered also hospital admissions with primary or secondary diagnosis with the ICD-9-CM code 486.x, so we could reasonably state that both Hospital-Acquired and Community-Acquired Pneumonia (HAP and CAP, respectively) has been included in the index hospitalization, since the ICD-9-CM code 486.x is used to record also community acquired pneumonia, healthcare associated pneumonia and hospital acquired pneumonia.

Furthermore, since the major outcome was the first rehospitalization for pneumonia within 30 days after the discharge from the first pneumonia admission, both HAP and CAP had been included also in this category.

As requested by the reviewer, we added in the manuscript the pneumonia ICD-9-CM codes that were considered for the identification of the index hospitalization [Materials and Methods section, Setting subsection, page 6, lines 19-20; Outcome subsection, page 7, lines 9-10].

Further information about type of pneumonia and causative microorganism were not available, since one of the major limitations of our study is that the information recorded in HCU databases did not include clinical data as causative microorganism, etc.

You need explain if a collinearity analysis was performed. Why did you not show the use of statins (Table 1)?

Reply: We have considered collinearity in the context of a statistical model that is used to estimate the relationship between one response variable and a set of predictor variables; multicollinearity occurs when there are high correlations among predictor variables, leading to unreliable and unstable estimates of regression coefficients. We performed a collinearity analysis considering most widely-used diagnostic for multicollinearity, the Variance Inflation Factor (VIF). VIF has been calculated for each covariate by performing a linear regression model of that predictor on all the other covariates. The VIF has been calculated using the following formula: 1/(1-R2); with R2 estimated from the above-mentioned linear regression model. This parameter, i.e., VIF, estimates how much the variability of a coefficient is inflated because of linear dependence with other predictor variables. The VIF has a lower bound of 1 but no upper bound; in order to classify VIF as “High”, we fixed the threshold of 3. We included in the main analysis only those predictor variables that did not have high VIFs.

According to the reviewer’s suggestion, information regarding the previous use of lipid-lowering and antihypertensive medications has been added in Table 1 [pages 23-24]

Tables, figures, and references need a review (references duplicate nº 5 and 14)

Reply: According to the suggestion of the reviewer, we had carefully read the manuscript many times, particularly focusing on Tables, Figures and References. References duplicated has been corrected, and several modifications has been apported to tables, figures, figures legends and references’ format. Furthermore, Table 1 has been consistently modified [pages 23-24]. Also, Table 2 [pages 25] and Figure 2 were consistently revised.

The introduction and discussion need a little information regarding frail elderly patients.

Reply: We thank the reviewer for this advice. According to his suggestion, the topic of frail patients has been introduced and discussed thoughtfully in the manuscript, and it has been now reported in the Introduction and Discussion section [Introduction, Page 5, Lines 1-8; Discussion, Page 14, Lines 4-11; Page 16, lines 23-24; Page 17, lines 1-2]

Reviewer #2:

The authors present a retrospective review of a large cohort of patients admitted with pneumonia to multiple hospitals in Northern Italy between 2003 and 2012 in which they assess the rate of pneumonia-specific readmission and its contributing cofactors. As opposed to most other studies on pneumonia readmission rates, the authors focus on related, pneumonia-specific readmissions, rather than all-cause readmissions. This is a strength of the study. Limitations include the significant delay between data collection and manuscript.

Major critiques

The authors should explain why they feel it is relevant to consider readmissions that occur beyond 30 days in their analysis. Intuitively it seems that these so called “late readmissions” are unrelated to factors involved in the original case. Furthermore, the rate of late readmissions is very low. Unless the authors can add further justification for including this in the paper, I would suggest removal and focusing on the readmissions within 30 days which is the standard most other research in this area examines.

Reply: We thank the reviewer for this suggestion. As the reviewer said, the aim of our study was to evaluate the incidence and predictors of early and late readmissions due to pneumonia.

However, from a first explorative analysis, as reported at page 9, lines 5-8 of the old version of our manuscript, it resulted that, from 31 to 90 days after the index date, only 47 patients (0.02%) experienced a late readmission for pneumonia. Thus, considering that the rate of late readmissions is very low and the potential lack of statistical power that could have occurred analyzing separately late pneumonia readmissions, according to the reviewer suggestion we decided to remove from the main analyses the readmissions that occurred beyond 30 days after the index date. We revised the paper focusing on the early readmissions (i.e., those within 30 days after the index discharge for pneumonia), being these early readmissions the standard that most other research in this area examined.

However, we mentioned late readmissions a few times in the manuscript to broaden the discussion and introduce potential future developments of this research.

The data being analyzed in this study is currently 8 – 17 years old. Certainly, temporal changes in the care of hospitalized patients have transpired, and the fact that the conclusions are based on old data should be clarified as a limitation in the discussion sections. Ideally the authors would frame their results against comparative readmission rates from more recent data and put their results into context. Unfortunately, few studies exist with pneumonia specific readmission rates, but if the authors have access to all-cause 30-day readmission rates, they could make the case for whether temporal trends suggest increasing, decreasing or stable rates compared to the study period.

Reply: we thank the reviewer for this advice. We discussed this topic in the Discussion section regarding the limitations of the study [page 16, lines 16-21].

Furthermore, as suggested by the reviewer, all-cause 30-day readmission rate was calculated and widely discussed along the manuscript [page 10, lines 4-5; page 13, lines 5-10; page 16, lines 16-21].

As suggested by the reviewer the data analyzed in this study relate to several years ago and are not updated, for this reason our study may not have been able to include some changes that have recently occurred in the care of hospitalized patient for pneumonia. However, our all-case 30-day readmission rates after a first pneumonia hospitalization agree with those reported in recent observational studies, ranging from 8.6 to 25% [6,8,19-20], thus we can likely suppose that all-case 30-day readmission rates after a first pneumonia hospitalization remained stable

In Table 1, there appears to be significant differences in various comorbidities between readmitted and not readmitted patients; this is pointed out in line 15 on p 9. However, when adjusted hazard ratios are calculated (Table 2) only renal disease appears to be a predictor. In the Results section (p10, line 4-7) the authors note that use of meds for chronic lung disease were associated with increased readmission risk, but chronic lung disease as a comorbidity was not significant. Aren't some of these meds in fact "proxies" for the underlying chronic medical conditions? The authors should discuss these dichotomies and the implications for the limitations of their data in the discussion section.

Reply: We thank the reviewer for this suggestion. Reading the manuscript and checking the results, we found that a result was incorrectly reported: in Table 2 the Hazard ratios (HR), and relative 95% confidence intervals (CI), of early rehospitalizations for pneumonia for patients having Chronic lung disease was not 0.95 (0.88-1.02), but the correct estimates (now reported in Table 2) are 1.12 (1.03-1.25). The appropriate and related changes were therefore made in the text [Page 7, lines 24-25; Page 11, lines 5-6; Table 2; Figure 2].

The paper would benefit from more careful proofreading and correction of grammatical errors, and some effort to improve the clarity of writing.

Reply: We thank the reviewer for this advice. According to this suggestion, we had the proofs of the manuscript corrected by a professional translator Ms. Mary McKenney.

Data Table 1 lacks column headings, these should be added.

Reply: According to the reviewer’s suggestion, Table 1 has been consistently modified, and also column headings have been added.

Minor critiques

Late readmissions are mentioned in the introduction but given the lack of clear clinical relevance and the exceedingly small number identified by the author, I think the paper would be more effective if this detail was omitted throughout the manuscript.

Reply: We thank the reviewer for this suggestion. According to this, as better explained in the response to his first suggestion in the Major critiques section, we revised the paper focusing on the early readmissions (i.e., those within 30 days after the index discharge for pneumonia), being these the standard that most other respiratory medicine research examines. However, we mentioned late readmissions a few times in the manuscript to broaden the discussion and introduce potential future developments of this research.

The statement about readmission penalties in the US is inaccurate in that it refers to the “private health insurance system.” In fact, 30-day readmission penalties in the US are imposed by Medicare which is a federally funded health insurance program for adults age 65 and older. It is not private insurance.

Reply: We agree with the suggestion of the reviewer and the statement about readmission penalties in the US previously referring to the “private health insurance system” has been rewritten [page 5, lines 22-23; page 6, lines 1-2].

Why does p 7 line 16 state that info was collected about prior admissions for pneumonia when this group of patients was excluded according to the text and Figure 1?

Reply: We thank the reviewer for his suggestion, this step of the cohort selection procedure is not very clear and misleading. According to the reviewer suggestion the sentence “previous admissions for pneumonia (up to 6 months before the index date)” was deleted.

We still want to highlight that patients who experienced a hospitalization for pneumonia in the 6 months before the index date has been excluded; with this cautionary criterium we ensured the inclusion in the cohort of patients for whom the index admission was not a rehospitalization after prior recent pneumonia hospital admission, avoiding a potential misclassification of the outcome.

I had difficulty linking a clinical relationship to pneumonia readmission rates and some of the factors on the list of clinical characteristics detailed on p7 line 20. Can the authors explain why they chose to include thrombocytosis and pressure ulcer as clinical covariates?

Reply: We thank the reviewer for this comment. We made a selection of covariates starting from the scientific literature (Weinreich M et al. Predicting the Risk of Readmission in Pneumonia: A Systematic Review of Model Performance. Ann Am Thorac Soc. 2016 Sep;13(9):1607-14; Makam AN et al. Predicting 30-Day Pneumonia Readmissions Using Electronic Health Record Data Journal of Hospital Medicine 2017 Vol 12. No 4). Furthermore, thrombocytosis is often used as a proxy of activated inflammatory state and pressure ulcers are more common in frail patients, a population analyzed in our study.

P9 line 5-8: More reason to exclude mention of the concept of late readmissions in this analysis. It adds little other than distracting from the main findings.

Reply: According to the reviewer this suggestion, and the previous ones, this sentence has been deleted.

Figure 1 – graphic is blurred.

Reply: According to the reviewer’s suggestion, Figure 1 has been modified and saved in a new format, in order to be clearer and sharper.

For a better view of the Figure 1, from the PDF file submitted to the Editorial Manager of the Journal, we please the reviewer not to view the Figure 1 directly from the PDF of the submission, but by going to the link (top right, “Click here to access / download ; Figure; Figure 1. Flow-chart.tiff”) and view the image after having downloaded it from the corresponding webpage; by doing this, the resolution of the image of Figure 1 will be higher.

Page 12 line 22, 23 – further explanation as to why inhaled bronchodilators were associate with increased risk of admission whereas ICS-Bronchodilator combination was not.

Reply: We thank the reviewer for this comment. This result may be due to the effect of the association with inhaled steroids. In fact, inhaled steroids were not associated with an increased risk of hospital readmission.

However, this is a population study based on HCU databases and not a pharmacological study, therefore the effect of individual molecules or associations cannot be assessed.

Figure 2 – No point estimate labeled for younger or older patients, therefore unable to visually compare point estimates between both groups

Reply: According to the reviewer suggestion, Figure 2 has been modified labelling the point estimates for younger and older patients.

P13 line 1-6, the authors follow discussion of a finding that states that inhaled steroids were not a predictor of readmission with a reference to a study that found that inhaled steroids are associated with higher risk of developing pneumonia. Was this a study about pneumonia readmission rates or primary episodes of pneumonia? If the latter, then the link between the 2 thoughts does not logically connect. Consider rewriting or removing.

Reply: We thank the reviewer for this observation and removed from the text the citation.

Since you have a separate paragraph addressing the relationship of systemic steroids and pneumonia readmissions (p13 line 17-18), consider removing mention of steroids in line 7 and just focus on ASA in this paragraph.

Reply: We agree with the reviewer on this suggestion and changed the text accordingly.

Submitted filename: Rebuttal Letter.doc

Click here for additional data file.

9 Jun 2020

PONE-D-19-32710R1

Rehospitalization for pneumonia after first pneumonia admission: Incidence and predictors in a population-based cohort study

PLOS ONE

Dear Dr. Faverio,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process

Thank you for taking the time to revise and resubmit your manuscript to PLOS ONE. As you will see, both of the Reviewers were satisfied with the revisions and only one has pointed out that the Abstract needs to be revised to reflect the updated Results. In addition to revisions to the Abstract, I would also like to see the following adjustments:

- In your Discussion, please remove any reference to the late pneumonia readmissions. This is never really addressed earlier in the manuscript (at least not now with the changes) so I think it should now be removed from here as well. This modification will also require some changes to the paragraph starting line 20 on page 14.

Please ensure that your decision is justified on PLOS ONE’s publication criteria and not, for example, on novelty or perceived impact.

==============================

Please submit your revised manuscript by Jul 24 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Andrea Gruneir

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper has improved substantially focusing the analysis on patients readmitted within 30 days.

The authors have adequately addressed the reviewer comments.

Minor critiques:

The authors should review abstract, objective does not correspond with the main text (predictors of late readmissions were not evaluated).

Reviewer #2: My comments were addressed sufficiently and the paper now appears to be appropriate for publication in my opinion.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

10 Jun 2020

Academic Editor: In your Discussion, please remove any reference to the late pneumonia readmissions. This is never really addressed earlier in the manuscript (at least not now with the changes) so I think it should now be removed from here as well. This modification will also require some changes to the paragraph starting line 20 on page 14.

We thank the Academic Editor for this suggestion. We changed the text accordingly.

Reviewer #1: This paper has improved substantially focusing the analysis on patients readmitted within 30 days.

The authors have adequately addressed the reviewer comments.

Minor critiques:

The authors should review abstract, objective does not correspond with the main text (predictors of late readmissions were not evaluated).

We thank the reviewer for this suggetion. We changed the text accordingly.

Reviewer #2: My comments were addressed sufficiently and the paper now appears to be appropriate for publication in my opinion.

Thank you

Submitted filename: Response to reviewers 10.06.doc

Click here for additional data file.

17 Jun 2020

Rehospitalization for pneumonia after first pneumonia admission: Incidence and predictors in a population-based cohort study

PONE-D-19-32710R2

Dear Dr. Faverio,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Andrea Gruneir

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

19 Jun 2020

PONE-D-19-32710R2

Rehospitalization for pneumonia after first pneumonia admission: Incidence and predictors in a population-based cohort study

Dear Dr. Faverio:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Andrea Gruneir

Academic Editor

PLOS ONE

Table 2

Hazard Ratios (HR), and relative 95% Confidence Intervals (CI), of early rehospitalizations for pneumonia in the whole cohort (203,768 patients), estimated by a multivariable Cox proportional hazard model.

Covariates	HR (95% CI)
Baseline
Sex
Female	1.00
Male	1.31 (1.25–1.37)
Age (years)
<70	1.00
≥70	1.11 (1.05–1.17)
Length of stay (days)
<10	1.00
≥10	1.10 (1.05–1.17)
Multisource Comorbidity Score α
Low	1.00
High	1.23 (1.16–1.30)
During hospital stay
Respiratory failure	1.35 (1.26–1.43)
Mechanical ventilation	1.50 (1.37–1.64)
≥1 unit blood transfusion	1.44 (1.33–1.56)
Pressure ulcer	1.34 (1.11–1.61)
Previous medications
Drugs for obstructive airway diseases	0.90 (0.79–1.03)
Inhaled steroids	1.08 (0.96–1.21)
Inhaled broncodilators	1.14 (1.03–1.26)
Persistent aspirin therapy	1.30 (1.05–1.60)
Persistent NSAID therapy	1.03 (0.88–1.19)
Persistent glucocorticoid therapy	1.21 (1.04–1.40)
Antidepressants	1.09 (1.01–1.16)
Antiarrhythmics	1.14 (1.05–1.24)
Previous diseases
Liver disease	0.99 (0.89–1.11)
Renal disease	1.19 (1.09–1.29)
Lung cancer	0.98 (0.84–1.15)
Chronic lung disease	1.12 (1.03–1.25)

The clinical status was assessed by the Multisource Comorbidity Score (MCS) according to the hospital admission and the drugs prescribed in the six-months period before the index date. Two categories of clinical status (MCS) were considered: good (Low MCS: score < 10) and poor (High MCS: score ≥ 10).