Forecasting the Multidimensional Burdens of Hospital Admissions in LRTIs and COPD: Projections of Healthcare Expenditures, Resource Utilization, and Socioeconomic and Environmental Outcomes for Switzerland

Abstract

Introduction

Hospital admissions due to lower respiratory tract infections (LRTIs) and chronic obstructive pulmonary disease (COPD) represent a substantial healthcare challenge for Switzerland. Demographic shift is expected to exacerbate the burden. Thus, multidimensional projections of healthcare resource utilization, environmental, and (socio)economic impact are necessary to facilitate sustainable healthcare planning for these diseases.

Methods

BRONCH-2035 employed historical admissions data (2015–2023) and Swiss governmental population scenario forecasts to project nationwide hospital and intensive care unit (ICU) admissions due to LRTIs and COPD between 2025 and 2035. Projections were then leveraged to estimate future inpatient healthcare expenditures, environmental and socioeconomic burden, employing historic tariff data (SwissDRG) alongside data on disease-agnostic greenhouse gas emissions, Swiss governmental employment and Organization for Economic Co-operation and Development (OECD) economic activity.

Results

Compared to 2025, 10,055 (18.4%) additional hospitalizations and an additional 488 (14.3%) ICU admissions due to LRTI and COPD were projected for 2035, requiring 338 (21.8%) additional hospital and 7 (15.9%) surplus ICU beds. Future admissions are estimated to result in an additional 100.2 million Swiss francs (CHF) (18.3%) in inpatient healthcare expenditures and 1.3 million kg (21.1%) additional CO₂-equivalent greenhouse gas emissions. Over the period of 2025–2035, LRTI and COPD-related hospitalizations are projected to cause 2.94 million missed workdays, corresponding to 15,402 lost full-time workers, CHF 1.32 billion in productivity losses, 2.46 billion Purchasing Power Parity (PPP) in lost GDP and CHF 151 million in lost tax revenue.

Conclusions

BRONCH-2035 is the first study to project the healthcare, environmental and socioeconomic burdens in LRTI and COPD for Switzerland. Population growth and demographic shift alone are projected to exacerbate LRTI and COPD hospitalizations, ICU admissions, inpatient healthcare expenditures, and bed requirements by 2035, alongside socioeconomic and environmental consequences. These findings provide a robust baseline for healthcare planning, highlighting the need for consistent guideline-concordant prevention and structured outpatient care.

Supplementary Information

The online version contains supplementary material available at 10.1007/s40121-026-01377-6.

Key Summary Points

Why carry out this study?

Lower respiratory tract infections (LRTIs) and chronic obstructive pulmonary disease (COPD) are major contributors to morbidity, mortality and hospital burden in Switzerland, yet comprehensive projections of future admissions, healthcare expenditures, and intensive care unit (ICU) needs are lacking on a national and global level.

Policymakers require multidimensional forecasts, including healthcare resource utilization, socioeconomic and environmental impact, to plan capacities and prioritize preventative and chronic care strategies.

BRONCH-2035 was designed to provide a status quo baseline for these future burdens, identifying where more consistent guideline-concordant care practices and improved care structures may help limit admission burden

What was learned from this study?

Population growth and demographic change alone are projected to increase annual hospital admissions in LRTI and COPD substantially by 2035, with parallel rises in ICU admissions, hospital and ICU bed requirements and inpatient healthcare expenditures.

These additional admissions are expected to generate substantial socioeconomic burden in terms of missed workdays, productivity losses and lost tax revenue, as well as a sizeable increase in healthcare-related greenhouse gas emissions.

The findings suggest that prevention and outpatient management remain important levers to reduce future respiratory admissions, while more consistent implementation of guideline-concordant care and structured chronic care delivery may support offsetting burden.

Introduction

Lower respiratory tract infections (LRTIs) and chronic obstructive pulmonary disease (COPD) remain leading causes of morbidity and mortality worldwide, imposing a substantial burden on healthcare systems and societies [1, 2].

Demonstrating the significant global LRTI burden, worldwide non-COVID-19 incidence of LRTIs is estimated at 344 million [m] episodes per year, with adults aged 70 years and older experiencing the highest incidence rates at 18,900 episodes per 100,000 (18.9%) population [2]. LRTIs are a significant contributor to global mortality, responsible for 2.5 m deaths worldwide in 2023, disproportionately impacting children under 5 years and adults over 70 years [3]. Streptococcus pneumoniae is the leading pathogen globally in terms of mortality, responsible for approximately 25% of all LRTI deaths, followed by Staphylococcus aureus and Klebsiella pneumoniae [3]. Swiss epidemiological data on LRTIs are scarce, with recent literature focusing predominantly on specific pathogens. In line with global trends, Streptococcus pneumoniae leads as the most predominant etiological pathogen in Switzerland [4]. Among viral pathogens, respiratory syncytial virus (RSV) – the leading pathogen in pediatric LRTIs – was responsible for 35,000 hospitalizations across Switzerland between 2017 and 2023. Further, admissions generated over 16,600 hospital days per year in patients 60 years and above, with mechanical ventilation rates and in-hospital mortality reaching up to 12.6% and 7.1%, respectively [5]. The COVID-19 pandemic has highlighted the ability of LRTIs to severely challenge public health systems, underscoring the need for improved public health and hospital preparedness [3]. Hospitalization risk in LRTIs is substantially elevated in older adults and patients with chronic or immunocompromising conditions, highlighting the growing pressure on healthcare systems as populations age and multimorbidity increases [6].

In COPD, global prevalence is expected to rise by 23.3% to over 600 m patients by 2050, accompanied by a 10.8% increase in direct costs amounting to $852.4 billion [b] annually [7]. In Switzerland, COPD also poses a significant public health challenge, accounting for 26.5% of all Potentially Avoidable Hospitalizations (PAHs) and nearly half of all deaths from respiratory diseases [8–10]. Costs associated with PAHs due to COPD amounted to CHF 117.9 m in 2019 and are forecast to reach CHF 149.8 m by 2032 [9].

The Swiss Study on Air Pollution and Lung Diseases in Adults (SAPALDIA) provides key epidemiological data on COPD in the country [11, 12]. Swiss prevalence of airflow obstruction among adults was estimated at 7.0% [12], with prevalence in over 45-year-old inpatients as high as 9.1% in 2017 [13]. The interplay between COPD and LRTIs is evident in Swiss clinical settings; patients with LRTIs frequently present with acute COPD exacerbations [14].

Outcome studies evaluating burden of disease in Switzerland are scarce for both and in particular for LRTIs, with Swiss research for the latter placing a focus on approaches to minimise guideline-incompliant antibiotic use [14, 15].

Switzerland’s healthcare expenditures are rising dramatically, with those related to COPD exacerbating by 32.7% between 2012 and 2017 [16]. Epidemiologically, this rise was attributed to population growth and aging. However, critically for policymakers, an aggravation in average per-patient spending was responsible for 84.7% (83 of 98 m CHF) of the total cost increase – reflecting a significant increase in the quantity and intensity of healthcare services consumed by patients with COPD. In communicable diseases, including LRTIs (not including HIV/AIDS), the impact of average per-patient spending was even larger, responsible for 98.0% of the respective total cost increase [calculated based on percentage changes in Table 5 in [16]]. Furthermore, inpatient healthcare expenditures may underestimate total costs as unaccounted post-discharge expenditures, may nearly double the costs of a single admission [17].

Healthcare resource utilization (HCRU) is higher in patients suffering from chronic diseases compared to otherwise healthy individuals with acute conditions [18]. In turn, they are driven by price and volume of health service utilization, which is influenced by demographics, disease burden and care preferences [19–21]. In the European comparison, Switzerland leads in excess healthcare utilization and inpatient healthcare cost burden – defined as excess healthcare utilization costs relative to total healthcare expenditure [19]. Responsible for 27.6% of healthcare expenditures, inpatient care is a major component of total health spending in Switzerland [calculated based on absolute totals depicted in Fig. 3 in [16]]. These include costs incurred by intensive care units (ICUs), which are disproportionately resource-intensive [22–24].

Understanding of the holistic burdens of LRTIs and COPD across HCRU, socioeconomic, and environmental outcomes is limited, while projections of future healthcare needs, productivity and environmental impact remain insufficient. Thus, given the substantial impact of population growth and aging on inpatient healthcare utilization in Switzerland – and the significant burden of COPD and LRTIs on hospital admissions – demographics-based projections of national hospital and ICU admissions may prove valuable to support public health planning. Though one recent study forecasted PAHs for various chronic diseases, including COPD, it did not address LRTIs or COPD admissions beyond PAHs [9]. Forecasts of admissions-related healthcare expenditures, hospital bed capacities, and environmental impact may further support policymakers and healthcare administrators. To address these needs, we developed BRONCH-2035 (Burden, Resource, and CO₂e Outlook for COPD and Hospitalized LRTIs in Switzerland to 2035), a nationwide forecasting study.

Hospital admissions and, in particular, ICU stays are recognized contributors to healthcare-related greenhouse gas (GHG) emissions, driven by high energy demand, resource-intensive care processes, and the use of single-use medical products [25]. As Switzerland has committed to achieving net-zero emissions by 2050 [26], projections of future clinical demand and its environmental impact of inpatient care are increasingly relevant for national and cantonal policymakers. Projections quantifying environmental outcomes for rising COPD and LRTI admissions can therefore support strategic planning, enabling decision-makers to anticipate pressures on both healthcare resources and sustainability targets. Embedding environmental impact alongside epidemiological and socioeconomic forecasts ensures that BRONCH-2035 provides a multidimensional evidence base to guide long-term health system preparedness.

Therefore, the objective of BRONCH-2035 is to project hospital and ICU admission rates for LRTIs, COPD, and acute exacerbations of COPD triggered by LRTIs through 2035, based on governmental demographic projections for Switzerland and assuming continuation of current patterns of diagnosis, prevention, and treatment. To deliver a holistic forecast of future burden and needs, secondary aims include estimating associated inpatient healthcare expenditures, analyzing outcomes related to healthcare resource utilization, including general ward and ICU bed utilization, quantifying their environmental impact via GHG emissions and projecting the socioeconomical burden of LRTI and COPD.

Methods

Study Design

This study employed a two-part design combining retrospective analysis of observational data and prospective modeling. Routinely collected, anonymized, nationwide administrative hospital inpatient data were obtained from the Swiss Federal Statistical Office (FSO) [27]. Each admission is ICD-10-coded for the main and secondary diagnoses, including comorbidities and complications. Additional variables include age groups, sex, year and month of hospitalization, canton of hospital, effective cost weight, Swiss Diagnosis Related Group (SwissDRG), hospital length of stay (LOS), ICU admission, ICU length of stay (ICU LOS), 30-day readmission and in-hospital mortality. The study leveraged anonymous data, conformed with local law and ethical review and research policies, adhering to the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines [28].

Definition of Chorts

Based on the primary ICD-10 code, we defined three cohorts: LRTI, COPD, and COPD with LRTI. ICD-10 codes for LRTIs and COPD with LRTI were defined by integrating code lists from validation studies of administrative data and epidemiological definitions, supplemented by pathogen-specific (e.g., RSV, COVID-19) codes [6, 29]. ICD-10 codes per cohort and exclusion criteria are described in Table S1.

Statistical Analyses

Statistical analyses were performed using R version 4.2.3 with the packages stats (4.2.3), tableone (version 0.13.2), and epitools (version 0.5–10.1). No data imputation or linkage was conducted. For descriptive analyses, non-normally distributed continuous variables are presented as median and interquartile range (IQR), and categorical variables as counts and percentages in each category. Multivariable logistic regression was used to test for associations with binary categorical outcomes. For the right-skewed outcome measures LOS and LOS ICU, multivariable linear regression was performed on log-transformed values, and the resulting estimated coefficients were back-transformed to allow interpretation as percentage increase/decrease. Models were computed for each cohort individually and adjusted for disease burden using Elixhauser comorbidity categories derived from ICD-10-GM coding, the binary variables smoking status, sex, hospitalization during (2020–2022) or beyond the COVID-19 pandemic years, and the categorical variable age group, defined as Infants & toddlers (0–4), Children & Adolescents (5–17), Adults (18–39), Middle-aged adults (40–64), Late adults (65–79), Seniors (80 +). Furthermore, the models for the LRTI cohort included secondary COPD as covariate, and the models for the COPD cohort included secondary LRTI.

Swiss Population Data

Retrospective Swiss population data stratified by age and sex for the years 1980–2024 were obtained from the FSO [30]. Projections of the Swiss population size per age, sex, and canton for the years 2025–2055 were also obtained from the FSO [31]. These projections include three scenarios – a reference scenario (“AR-00–2025"), a high scenario (“BR-00–2025") and a low scenario (“CR-00-2025"), assuming different birth, mortality, and migration rates.

Hospital and ICU Admissions

Retrospective (2015–2024) and prospective (2025–2034) Swiss population counts were converted to sex-stratified 5-year age groups (0–4 to 95 +), matching the admissions data. Due to systematic biases in hospitalization patterns during the COVID-19 pandemic, the years 2020–2022 were excluded. Projections were made for all-cause hospitalizations (nationwide hospitalizations from any disease) as well as each cohort (LRTI, COPD, COPD, with LRTI) individually. First, the yearly number of hospitalizations per sex-stratified 5-year age group was calculated for the years 2015–2019 and 2023. These were then divided by the recorded population count of the corresponding group in that year, resulting in yearly hospitalization rates per group. Next, the median hospitalization rates across the years 2015–2019 and 2023 were calculated for each group. Future hospitalizations were then projected by multiplying the median hospitalization rates of each group by its projected population size, and annual totals were obtained by summing the group-specific projections. This calculation was repeated for each year from 2025 to 2035. Using the same approach, we also projected the total annual numbers of ICU admissions and the annual LOS and ICU LOS. To account for uncertainties in the FSO population projections, all forecasts were repeated using the low and high population scenarios, which are presented as value ranges "(low scenario-high scenario)" in the results section. As an internal validation, projections were repeated using the median annual values from 2015–2019 (5 instead of 6 historic years) to predict values for 2023. A zero-intercept linear regression model was then used to quantify the proportional bias between these projected values and the observed values for 2023, with the slope indicating the extent of systematic over- or under-estimation.

Inpatient Healthcare Expenditures

All hospitalizations included in this study were reimbursed under the SwissDRG (acute hospital admissions) or ST-Reha systems (inpatient rehabilitation). Under SwissDRG, each case is assigned a relative cost weight. For hospitalizations beyond the DRG-specific lower/upper LOS thresholds, this relative weight is adjusted into an effective cost weight via per diem deductions or supplements, which is then multiplied by the negotiated base rate to obtain the reimbursement amount [32, 33]. Under ST-Reha, each case is assigned to a rehabilitative cost group (RCG) with phase-based daily cost weights. The reimbursable (effective) cost weight is calculated by summing daily weights across phases (with weights potentially changing between phases, commonly stepping down over time) and multiplying by the applicable base rate [34].

To estimate inpatient healthcare expenditures, canton-specific base rates for 2023 were derived by averaging all available hospital base rates separately for SwissDRG and ST-Reha tariffs, applying the appropriate canton-level base rate for each hospitalization [35]. SwissDRG costs by case were calculated by multiplying the effective cost weight by the corresponding DRG base rate of the canton [36]. For ST-Reha cases, only the broader rehabilitation cost group category (arcg) was available in the administrative dataset, whereas reimbursement is formally based on more granular phase-specific RCG subclasses. Therefore, a single average rehabilitation cost weight per rehabilitation category was calculated using all corresponding RCG-specific daily cost weights from the official ST-Reha catalog [37]. This value was multiplied by the canton-specific ST-Reha base rate [35] for each hospitalization. Using the estimated case-level costs, the mean cost per hospitalization for each cohort was calculated for 2023, along with the total annual number of hospitalizations in each subgroup. Finally, annual inpatient healthcare expenditures were generated for each cohort, year, and population scenario, based on the projected number of hospitalizations. All projected expenditures are expressed in constant 2023 reimbursement terms, assuming stable canton-level base rates and tariff structures over time.

To facilitate inpatient healthcare proportion estimates, we additionally projected costs for all non-psychiatric admissions, reimbursed exclusively under the SwissDRG or ST-Reha and costed using the same canton-level base rates and cost-weight assumptions.

Hospital and ICU Bed Utilization

The yearly number of required hospital beds per cohort was calculated as

$${\mathit{Beds}}_{c,y}=\frac{\sum {\mathit{LOS}}_{c,y}}{365\ast 0.8}$$

where Beds_c,y is the number of occupied beds by cohort c in year y, LOS is length of stay in days, 365 the number of days in a year, and the factor 0.8 corrects for an empirical bed occupancy rate of 80%, as previously published [9].

Similarly, ICU bed utilization was calculated as

$${\mathit{ICUBeds}}_{c,y}=\frac{\sum {\mathit{LOSICU}}_{c,y}}{365\ast 24\ast 0.8}$$

where ICU Beds_c,y is the number of occupied ICU beds by cohort c in year y, LOS ICU is length of stay on the ICU in hours, 365 * 24 is the total number of hours per year, and the factor 0.8 corrects for 80% occupancy rate [9].

Socioeconomic Outcomes

Projections of socioeconomic outcomes included patient and caregiver absenteeism due to hospitalizations, patient absenteeism due to premature death, productivity, gross domestic product (GDP) PPP, full-time equivalent (FTE) workers, tax revenue, applying the methodologies developed in IMPACT-CKD to quantify burden across all outcomes [38]. These socioeconomic methodologies are disease-agnostic, converting hospitalizations and in-hospital deaths into economic outcomes using standard human capital approaches and population-level labor market parameters, independent of the underlying admitting diagnosis. For each cohort, outcomes were estimated annually by age group and subsequently aggregated across all ages to generate yearly estimates and cumulatively across all years to capture total socioeconomic impact over time.

Patient absenteeism due to hospitalizations was estimated only for individuals aged 15 + years and adjusted for governmental age-specific employment rates for Switzerland (ranging from 54.6% for ages 15–19 to 90.6% for ages 45–49; full detail in Supplementary Table S2), to reflect that not all patients are employed [39–41]. Missed workdays were limited to working days: for average hospital stays longer than 5 days, patients were assumed to miss 5 working days plus any additional working days beyond a 7-day period. For patients aged 0–14 years, work loss was attributed to caregivers, using the governmental standardized employment rate for Switzerland (67.3%) as a proxy for caregiver labor force participation [41]. Absenteeism due to death was calculated as the work time patients would have contributed had they survived, based on an average employment start of 16.5 years – an assumption of the age at which 50% of the Swiss age cohort are employed [39] – and retirement ages [42]. Remaining working years were adjusted using the Swiss standardized employment rate (67.3%) — which represents a lifetime-averaged employment probability — rather than age-specific rates, to avoid speculative assumptions about individual future employment trajectories over multi-decade time horizons [39].

Lost productivity was derived by multiplying missed workdays due to absenteeism with Swiss-specific values for average daily salary [43]. Similarly, lost GDP was derived by multiplying missed workdays due to absenteeism by GDP per workday (Purchasing Power Parity (PPP)) – in turn derived by multiplying GDP per hour worked (PPP) with the Swiss average for hours worked per workday [44, 45]. Lost FTEs were calculated by dividing total missed workdays due to absenteeism by the average number of days worked per year by a full-time worker, adjusted from the Swiss average for hours worked per workday [44, 46]. Lost tax revenue was calculated by applying the average tax rate to income lost due to absenteeism – assumed to be 50% of lost productivity, as per IMPACT-CKD [38, 47]. Age-stratified economic activity rates in Switzerland are outlined in Table S2 and further input parameters are summarised in Table S3.

Greenhouse Gas Emissions

Estimates of GHG emissions, expressed as carbon dioxide equivalents (CO₂e), associated with a 1-day disease-agnostic stay in a Swiss hospital or 1-day stay in ICU were obtained from IMPACT-CKD [38, 48]. Annual CO₂e-emissions attributable to hospitalizations were then calculated for each cohort by

$${C{O}_{2}e}_{c,y}=\sum {\mathit{LOS}}_{c,y}\ast 12.7kgC{O}_{2}e/d$$

where CO₂e_c,y denotes the amount of CO₂e emitted through hospitalizations for cohort c in year y; LOS_c,y represents the length of stay in days; and 12.7 kg CO₂e/d corresponds to the emission factor per hospitalization day [38, 48, 49].

Similarly, annual CO₂e emissions attributable to ICU stays were calculated for each cohort as:

$$C{O}_{2}eIC{U}_{c,y}=\sum LOSIC{U}_{c,y}\ast 24\ast 37.9kgC{O}_{2}e/d$$

where CO₂e ICU_c,y denotes the amount of CO₂e emitted through ICU stays of cohort c in year y; LOS ICU represents the length of ICU stay in hours, 24 is the total number of hours per day, and 37.9 kg CO₂e/d corresponds to the emission factor per ICU day [38, 48].

Canton-Level Burden Projections

Additional canton-level projection analyses were performed for exploratory regional analyses. Age- and sex-specific hospitalization rates were estimated using historic canton-level hospitalization data and applied to canton-specific population projections to estimate future regional variation in admissions, ICU admissions, and inpatient healthcare expenditures. Canton-specific inpatient healthcare expenditures were estimated using canton-specific SwissDRG and ST-Reha base rates derived from 2023 hospitalization data. Canton-level projections are presented in the Supplementary Materials.

Sensitivity Analyses

Sensitivity analyses were additionally performed to evaluate the impact of hypothetical annual changes in age- and sex-specific hospitalization rates between 2025 and 2035. A baseline scenario assuming stable hospitalization rates over time was compared with scenarios applying either a 1% annual decrease or a 1% annual increase in hospitalization rates across all age and sex strata. Projected admissions, ICU admissions, bed demand, costs, and greenhouse gas emissions were recalculated under each scenario using the reference population projection.

Results

Descriptive Statistics

A total of 12.9 m hospitalizations were documented between 2015 and 2023. 466,023 (3.6%) fell in scope (Figure S1). Among these, 365,937 (79%) were due to LRTIs, 56,329 (12%) due to COPD and 43,757 (9%) were coded COPD with LRTI. Table 1 provides an overview of cohort-stratified patient characteristics. Observed and projected hospitalization prevalences per year and cohort are shown in Table S4, and observed age-standardized rate per 100,000 people in Table S5, allowing comparison across cohorts independent of their differing age distributions. LRTIs consistently demonstrated the highest age-standardized admission rates. Between 2015 and 2023, yearly hospital admissions due to LRTI and/or COPD rose from 45,049 to 54,631 (21.3%), with LRTIs increasing from 34,134 to 42,289 (23.9%), COPD from 5634 to 6908 (22.6%) and COPD with LRTI from 5281 to 5434 (2.9%; all in Table S6). Patients aged ≥ 80 had the longest hospitalizations (Figure S2A) and the shortest ICU stays (Figure S2C). The median LOS of patients with COPD or COPD with LRTI was 2 days longer than the median LOS of patients with LRTI (Figure S2B) but patients with LRTI had the longest LOS ICU (Figure S2D). Table S6 presents additional information on comorbidities and data stratified by year and season. The mean yearly number of admissions across the COVID-19 pandemic years compared to years beyond the pandemic was 28% higher for LRTIs (47,706 vs. 37,137), 7% lower for COPD (5977 vs. 6400), and 28% lower for COPD with LRTI (3868 vs. 5359). Hospitalizations involving LRTIs were seasonal, with more than a third of LRTI (39%), and COPD with LRTI (36%) admissions recorded in the winter months (Figure S3A–C).

Table 1.

Baseline characteristics of the patients overall and stratified by cohort

	LRTI and/or COPD2	LRTI	COPD	COPD with LRTI
	n = 466,023	n = 365,937	n = 56,329	n = 43,757
Age category (%)
Infants & toddlers (0–4)	70,439 (15.1)	70,285 (19.2)	138 (0.2)1	16 (0.0)1
Children & Adolescents (5–17)	10,162 (2.2)	10,094 (2.8)	52 (0.1)1	16 (0.0)1
Adults (18–39)	15,041 (3.2)	14,723 (4.0)	192 (0.3)1	126 (0.3)1
Middle-aged adults (40–64)	86,009 (18.5)	62,391 (17.0)	14,668 (26.0)	8950 (20.5)
Late adults (65–79)	148,976 (32.0)	96,368 (26.3)	29,747 (52.8)	22,861 (52.2)
Seniors (80 +)	135,396 (29.1)	112,076 (30.6)	11,532 (20.5)	11,788 (26.9)
Sex = Male (%)	254,565 (54.6)	202,498 (55.3)	28,079 (49.8)	23,988 (54.8)
Smoker (%)3	18,850 (4.0)	5249 (1.4)	8604 (15.3)	4997 (11.4)
Disease burden (median  [IQR])4	2 [0, 3]	2 [0, 3]	2 [1, 3]	2 [1, 4]
Length of stay [days] (median [IQR])	6 [3, 11]	6 [3, 10]	8 [4, 15]	8 [5, 13]
In-hospital death (%)
Death	19,529 (4.2)	15,792 (4.3)	1813 (3.2)	1924 (4.4)
Discharged alive	442,979 (95.1)	347,196 (94.9)	54,186 (96.2)	41,597 (95.1)
Unknown	3515 (0.8)	2949 (0.8)	330 (0.6)	236 (0.5)
ICU admission (%)	31,483 (6.8)	22,021 (6.0)	4297 (7.6)	5165 (11.8)
Length of ICU stay [hours] (median [IQR])5	65 [28, 144]	74 [32, 176]	42 [22, 76]	52 [26, 98]
30 days readmission (%)	44,039 (9.4)	28,756 (7.9)	9358 (16.6)	5925 (13.5)
Respiratory failure (%)	102,189 (21.9)	84,489 (23.1)	8049 (14.3)	9651 (22.1)

ICU Intensive care unit, LRTI lower respiratory tract infection, COPD Chronic obstructive pulmonary disease

Patients with “discharge type unknown” and “in-hospital death” were excluded from regression analyses. ¹Cohorts were defined by primary ICD-10 hospitalization codes. A small number of pediatric COPD cases (n = 222 in 2015–2023) likely reflect miscoding or rare conditions, a known limitation of administrative data. These cases represent only 0,22% of COPD hospitalizations and were excluded, as regression analyses and projections for COPD and COPD with LRTI were restricted to patients aged ≥ 40 years, in line with the Agency for Healthcare Research and Quality (AHRQ) methodology for Prevention Quality Indicator 05 calculations [50]. ²LRTI and/or COPD: Cumulative values for LRTI, COPD and “COPD with LRTI”. ³Smoking status was defined as the presence/absence of any F17* ICD-10 code as secondary diagnosis. ⁴Based on secondary diagnoses, patients were assigned a binary variable for the presence/absence of each of the Elixhauser comorbidities [51], except “Chronic pulmonary disease” (30/31 groups). For each patient, the sum across these binary Elixhauser comorbidity variables was calculated to serve as a numeric indicator for overall disease burden. ⁵Median and IQR of length of ICU stay across patients admitted to the ICU

Independent Associations with Clinical Outcomes

Multivariable regression analyses, as presented in Supplementary Tables S7-9 and Figure S4, revealed several independent associations. Most notably, smoking patients were independently associated with longer LOS among all cohorts, and in COPD with/without LRTI they had increased odds of 30-day readmissions. The pandemic was associated with ICU admissions and longer ICU LOS in LRTIs only. Patients aged 80 or older had decreased odds of ICU admissions and were associated with shorter ICU LOS among all cohorts.

Projections of Hospital Admissions

Between 2025 and 2035, the annual number of all-cause hospitalizations was projected to rise from 1,574,051 to 1,766,233 (12.2%; Table S10). During the same period, annual admissions were projected to increase from 41,409 to 49,062 (18.5%) for LRTIs; from 7225 to 8433 (16.7%) for COPD, and from 6153 to 7347 (19.4%) for COPD with LRTI (Fig. 1; Table 2). Figure S5 shows a validation in which hospitalization admissions in 2023 were projected with the median annual values from 2015–2019 and compared to the observed values in 2023. Year-by-year projections for all outcomes in the period 2025–2035 are shown in Table S10.

Fig. 1.

Observed and projected yearly hospital admissions (one dot per year; 2015–2035). Annual total hospital admissions due to A) LRTI, B) COPD and C) COPD with LRTI. Red and green dots show observed values during (2020–2022) and beyond (2015–2019, 2023) the COVID-19 pandemic. Blue dots show base case projections calculated with BFS reference population scenario and yellow dots show projected values based on the BFS low/high population scenarios. LRTI lower respiratory tract infection, COPD chronic obstructive pulmonary disease

Table 2.

Projections of healthcare resource utilization and inpatient healthcare expenditures

Outcome	Cohort	Projections for 20251	Projections for 20351	Absolute change 2	Percent change 2
				2025–2035	2025–2035
Hospital admissions	LRTI	41,409 (41,148–41,680)	49,062 (46,104–52,086)	7653	18.5
	COPD	7225 (7216–7233)	8433 (8247–8606)	1208	16.7
	COPD with LRTI	6153 (6143–6162)	7347 (7149–7537)	1194	19.4
	LRTI and/or COPD3	54,787 (54,507–55,075)	64,842 (61,500–68,229)	10,055	18.4
ICU admissions	LRTI	2089 (2078–2100)	2380 (2269–2488)	291	13.9
	COPD	582 (581–582)	666 (654–676)	84	14.4
	COPD with LRTI	736 (735–737)	849 (833–863)	113	15.4
	LRTI and/or COPD3	3407 (3394–3419)	3895 (3756–4027)	488	14.3
Bed utilization	LRTI	1064 (1059–1069)	1313 (1241–1387)	249	23.4
	COPD	267 (266–267)	312 (305–318)	45	16.9
	COPD with LRTI	220 (220–221)	264 (257–271)	44	20
	LRTI and/or COPD3	1551 (1545–1557)	1889 (1803–1976)	338	21.8
ICU bed utilization	LRTI	29 (29–29)	33 (31–34)	4	13.8
	COPD	6 (6–6)	7 (7–7)	1	16.7
	COPD with LRTI	9 (9–9)	11 (11–11)	2	22.2
	LRTI and/or COPD3	44 (44–44)	51 (49–52)	7	15.9
Inpatient healthcare expenditures [CHF]	LRTI	388,004,824 (385,562,198–390,544,596)	459,718,371 (432,001,219–488,050,029)	71,713,547	18.5
	COPD	83,401,246 (83,301,468–83,499,356)	97,347,739 (95,207,559–99,342,204)	13,946,493	16.7
	COPD with LRTI	74,976,304 (74,861,574–75,096,472)	89,535,372 (87,117,637–91,843,582)	14,559,068	19.4
	LRTI and/or COPD3	546,382,374 (543,725,240–549,140,424)	646,601,482 (614,326,415–679,235,815)	100,219,108	18.3
GHG emissions [kg CO2e]	LRTI	4,268,415 (4,248,534–4,289,849)	5,232,313 (4,949,430–5,522,038)	963,898	22.6
	COPD	1,056,631 (1,055,413–1,057,816)	1,231,320 (1,205,148–1,255,568)	174,689	16.5
	COPD with LRTI	920,319 (918,969–921,706)	1,096,399 (1,068,082–1,123,393)	176,080	19.1
	LRTI and/or COPD3	6,245,365 (6,222,916–6,269,371)	7,560,032 (7,222,660–7,900,999)	1,314,667	21.1

Projections of hospital admissions, ICU admissions, hospital and ICU bed utilization, inpatient healthcare expenditures, and GHG emissions

ICU intensive care unit, LRTI lower respiratory tract infection, COPD chronic obstructive pulmonary disease, GHG greenhouse gas

¹Base case projections derived from reference population scenario, with low to high population scenario in brackets. ²Absolute and percent change based on projections with reference scenario for 2025 and 2035. ³Cumulative projections across LRTI, COPD, COPD with LRTI

Projections of ICU Admissions

From 2025 until 2035, the yearly number of ICU admissions were projected to increase from 2089 to 2380 (13.9%) for LRTIs; from 582 to 666 (14.4%) for COPD, and from 736 to 849 (15.4%) for COPD with LRTI (Figure S6; Table 2). Figure S7 shows a validation in which ICU admissions in 2023 were projected with the median annual values from 2015–2019 and compared to the observed values in 2023.

Projections of Inpatient Healthcare Expenditures

From 2025 until 2035, annual inpatient healthcare expenditures due to LRTI and/or COPD were projected to increase from CHF 546.4 to 646.6 m (CHF 100.2 m; 18.3%; Fig. 2C; Table 2). Annual inpatient healthcare expenditures due to LRTIs were projected to rise from CHF 388.0 to 459.7 m (18.5%), while those due to COPD were projected to increase from CHF 83.4 to 97.3 m (16.7%). In line with the admissions projections, the largest increase was estimated for annual inpatient healthcare expenditures due to COPD with LRTI, from CHF 75.0 to 89.5 m (19.4%; Figure S8G-I; Table 2). In 2035, inpatient healthcare expenditures due to LRTI and/or COPD (CHF 646.6 m) were projected to represent 3.6% of all non-psychiatric inpatient healthcare expenditures (CHF 18.2 b; Table S10).

Fig. 2.

Projected yearly hospital resource utilization (one dot per year; 2025–2035). Projected annual total hospital beds, ICU beds, inpatient healthcare expenditures (costs), and GHG emissions attributed to LRTI and/or COPD hospitalizations and ICU stays. Cumulative projections including LRTI, COPD, COPD with LRTI. Blue dots show base case projections calculated with BFS reference population scenario and yellow dots show projected values based on the BFS low/high population scenario. ICU intensive care unit, LRTI lower respiratory tract infection, COPD Chronic obstructive pulmonary disease

Projections of Hospital and ICU Bed Utilization

Between 2025 and 2035, annual hospital bed utilization for LRTI and/or COPD was estimated to rise by 338 beds (21.8%; Fig. 2; Table 2), with an increase of 249 beds (23.4%) for LRTIs, 45 beds (16.9%) for COPD, and 44 additional beds (20.0%) for COPD with LRTI (Figure S8A–C; Table 2).

ICU bed requirements for LRTI and/or patients with COPD were projected to rise by seven beds (15.9%) from 2025 to 2035 (Fig. 2; Table 2), with an increase by four ICU beds (13.8%) for LRTIs, one ICU bed (16.7%) for COPD, and two ICU beds (22.2%) for COPD with LRTI (Figure S8; Table 2).

Projections of Greenhouse Gas Emissions

Between 2025 and 2035, annual GHG emissions attributable to hospital and ICU admissions for LRTI and/or COPD were projected to rise by 1,314,667 kg CO₂e (21.1%; Fig. 2; Table 2). This increase comprises 963,898 kg CO₂e due to LRTIs, 174,689 kg CO₂e due to COPD, and 176,080 kg CO₂e due to COPD with LRTI (Figure S8J-L; Table 2).

Projections of Socioeconomic Outcomes

Over the 11-year projection period (2025–2035), hospitalizations due to LRTI and/or COPD were projected to result in 2,938,290 missed workdays for patients and caregivers due to hospitalizations and in-hospital death, corresponding to 15,402 lost full-time equivalent workers. These missed workdays were estimated to result in 2.46 b PPP in lost GDP, CHF 1.32 b in lost productivity and CHF 151,112,806 in lost tax revenue (Table 3).

Table 3.

Projection of socioeconomic burden. Projected cumulative socioeconomic burden for the 11-year projection period between 2025 and 2035

Cohort	Absenteeism due to hospitalizations and in-hospital death	Lost productivity	Lost GDP	Lost FTEs	Lost tax revenue
	[days]	[CHF]	[PPP]		[CHF]
LRTI	2,199,483	987,919,955	1,841,777,440	11,529	113,116,835
COPD	396,282	177,994,045	331,833,986	2077	20,380,318
COPD with LRTI	342,526	153,848,494	286,819,478	1795	17,615,653
LRTI and/or COPD1	2,938,290	1,319,762,495	2,460,430,903	15,402	151,112,806

LRTI lower respiratory tract infection, COPD chronic obstructive pulmonary disease, GDP gross domestic product, FTE full-time equivalent

¹Cumulative values across LRTI, COPD, COPD with LRTI

Sensitivity Analyses

Sensitivity analyses demonstrated that modest annual changes in age- and sex-specific hospitalization rates substantially influenced projected burden trajectories (Table S11). Under a hypothetical 1% annual reduction in hospitalization rates, projected combined LRTI and/or COPD admissions and ICU admissions in 2035 decreased by 9.6% relative to the base case (58,642 vs. 64,842 admissions; 3522 vs. 3895 ICU admissions), corresponding to reductions in projected hospital bed occupancy (1708 vs. 1889 beds; 46 vs. 51 ICU beds), inpatient healthcare expenditures (CHF 584.8 million vs. CHF 646.6 million), and greenhouse gas emissions (6.84 vs. 7.56 million kg CO₂e). Conversely, a hypothetical 1% annual increase in hospitalization rates resulted in a 10.5% increase in projected admissions and associated healthcare burden indicators.

Projections of Canton-Level Burden

Canton-level projections demonstrated substantial regional variation in projected burden increases between 2025 and 2035 (Table S12). Across the analyzed cohorts, the largest absolute increases in projected hospital admissions were observed in Bern, Zurich, and Aargau. Canton-level projections for ICU admissions and inpatient healthcare expenditures are also reported in Table S12.

Discussion

Key Findings

Nearly 13 m hospitalizations were recorded in Switzerland between 2015 and 2023. Almost half a million (3.6%) were due to LRTI and/or COPD. BRONCH-2035 projects population growth and aging in Switzerland to cause a 10,055 (18.4%) increase to 64,842 LRTI and/or COPD admissions in 2035. Between 2025 and 2035, hospital admissions due to LRTI, COPD and COPD with LRTI will rise to 49,062 (18.5%), 8433 (16.7%), and 7347 (19.4%), respectively. These additional admissions are projected to result in an additional CHF 100.2 m in annual inpatient healthcare expenditures and require 338 additional hospital and 7 additional ICU beds. Projected non-ICU beds would predominantly be required within internal medicine or pulmonology departments, though the specific allocation depends on cantonal and institutional organizational structures, which vary across Switzerland. The projected increase in hospital admissions across all cohorts will lead to 1.31 m kg additional annual CO₂e-emissions, reflecting the significant environmental burden associated with in-hospital LRTI and COPD care. The socioeconomic impact between 2025 and 2035 is substantial: hospital admissions and in-hospital death related to LRTI and COPD are projected to result in nearly 3 m days in absenteeism, CHF 1.3 b in lost productivity, 2.5 b PPP in lost GDP, 15,402 lost FTEs and CHF 151 m in lost tax revenue. These projections represent a status quo scenario, based solely on demographic change and current patterns of diagnosis, prevention and treatment, without modeling future changes or improvements in respiratory care.

Comparison with Previous Studies

To our knowledge, our study is the first globally to estimate future hospitalizations in LRTIs, as well as ICU admissions for LRTI and COPD, and therefore fills an important literature gap. Particularly relevant for infectious disease services and winter-surge planning, BRONCH-2035 is also the first to project Switzerland’s multidimensional burdens related to admissions in LRTI and COPD. A recent Swiss study forecasted PAHs in COPD to rise by 13.6% [9], while BRONCH-2035 projects a similar increase of 12.9% for COPD admissions with/without LRTI (13,378 to 15,107) in this period. A Europe-wide study [52] projected healthcare and economic burdens for COPD, leveraging two different prevalence sources. Forecasts, based on estimated COPD prevalences from a meta-analysis [1] anticipate relative stability, projecting a rise in annual direct costs of 4.0% (170.01 b to 173.28 b) and a decrease in annual exacerbation counts of approximately 1.5% (49.07 m to 48.34 m) between 2025 and 2035. The sensitivity analysis, leveraging Global Burden of Disease (GBD) COPD prevalence estimates [53], forecasts similar changes. Our projected national growth in hospitalization burden, with a 16.7% increase in admissions due to COPD and 19.4% in COPD with LRTI between 2025 and 2035, is higher than these broader European forecasts. While methodological differences likely contribute to this divergence, including the use of nationwide real-world hospitalization data in BRONCH-2035 versus broader modeled European estimates, structural differences are also likely important. Switzerland has one of the most rapidly aging populations in Western Europe alongside particularly high per-capita healthcare utilization and inpatient healthcare expenditure [16, 19]. As respiratory hospitalizations disproportionately affect older adults, these demographic and healthcare system characteristics may amplify future admissions burden in Switzerland compared with broader European averages. BRONCH-2035 therefore provides a complementary Swiss-specific perspective using real-world hospitalization and cost data together with official demographic projections.

Interpretation of Regression Results

Smoking was associated with increased LOS in all cohorts, and with 30-day readmissions among patients with COPD (with/without LRTIs). These findings suggest that changes in smoking prevalence and smoking-related disease burden may influence future respiratory healthcare utilization beyond demographic effects alone. As BRONCH-2035 intentionally models a status quo baseline assuming continuation of current prevention and treatment patterns, future improvements in smoking prevention, structured outpatient care, vaccination uptake, or chronic disease management could modify the projected burden trajectories. Across all cohorts, hospitalizations of seniors aged 80 and older were associated with lower odds of ICU admission and shorter ICU LOS. This pattern may reflect differences in frailty, treatment intensity, or ICU admission practices in elderly patients. In the context of projected demographic aging in Switzerland, these findings suggest that future increases in healthcare burden may be driven not only by admission counts, but also by prolonged occupancy of non-ICU hospital beds among older patients. These results broadly align with previous findings on the negative correlation between older age with ICU admissions and ICU LOS in acute respiratory infections [54]. Studies investigating ICU admissions in LRTI subtypes – RSV and Streptococcus pneumoniae – have further confirmed the trend of shorter ICU LOS in older patients, though patient age was not correlated with risk for ICU admissions in RSV [55, 56]. ICU admission rates and seasonality in LRTI align with findings in Italy (6.0% vs. 6.6%), though Swiss in-hospital mortality and hospital LOS were comparatively reduced [57]. There was a sharp increase in hospital admissions due to LRTIs during the pandemic years, with increased ICU admissions and ICU LOS. Concurrently, there was a strong reduction in hospitalizations and ICU admissions due to COPD and COPD with LRTI. This is in line with findings from a meta-analysis, reporting a 50% reduction of admissions for COPD exacerbations during the pandemic [58]. Lower hospitalization rates might be linked to preventative measures reducing the risk of respiratory viral infections, as well as policy changes implemented to tackle resource scarcity. It is worth noting that the data from the pandemic years was excluded from all our projections to prevent biases. However, a future pandemic would likely lead to substantially higher LRTI hospitalizations and ICU admissions, and a potential reduction in COPD exacerbations.

Environmental

BRONCH-2035 revealed that by 2035 hospital admissions for LRTI and/or COPD will result in 7.56 m kg of annual CO₂e-emissions, equivalent to 160 full transatlantic passenger flights from Zurich to New York [59]. Hospital admissions for LRTIs alone accounted for approximately 70% of these emissions. Overall, surplus admissions represented an additional 1.31 million kg of CO₂e per year compared to 2025. This substantial environmental burden highlights the importance of integrating sustainability into strategic planning for inpatient care, linking admission projections to Switzerland’s commitments towards climate resilience and low-carbon health systems [26]. BRONCH-2035 provides actionable evidence supporting progress toward netzero healthcare emission goals.

Strengths and Limitations

A key strength of BRONCH-2035 is the utilization of comprehensive real-world evidence on nationwide administrative hospital admissions and costs from the FSO, alongside official FSO population projections. Furthermore, our study provides a holistic projection of the multidimensional burdens related to hospital admissions in LRTI and COPD, projecting inpatient healthcare expenditures, hospital capacity needs, socioeconomic outcomes and GHG emissions. Despite the use of comprehensive real-world data, our study has limitations. While demographic projections from the FSO incorporate anticipated changes in health-related behaviors, including smoking rates, BRONCH-2035 could not account for potential shifts in disease prevalence caused by risk factor dynamics [60, 61]. Thus, depending on how disease prevalences evolve, our projections may either over- or underestimate the true number of admissions. BRONCH-2035 did not model the potential impact of improved preventative interventions or enhanced implementation of Guideline-Directed Medical Therapy (GDMT). Such improvements could occur via policy changes or care structure reforms, potentially reducing projected burden. Instead, BRONCH-2035 maintains the status quo, with projections reflecting current patterns of diagnosis and treatment. BRONCH-2035 should therefore be interpreted as a baseline scenario rather than a fixed prediction, against which the impact of enhanced prevention and chronic care management can be evaluated. Cost projections are expressed in constant 2023 reimbursement terms and assume stable canton-level base rates and cost weights over time, as changes to future tariff structures and reimbursement systems remain uncertain and were therefore not incorporated into the projections [32]. This approach was chosen to isolate projected expenditure changes attributable to demographic developments and admission volumes rather than speculative future reimbursement reforms. Additionally, rehabilitation expenditures could only be estimated using broader rehabilitation cost group categories available in the administrative dataset, rather than exact RCG subclass reimbursement structures. Consequently, average rehabilitation cost weights across corresponding RCG subclasses were applied, which may introduce approximation error for individual rehabilitation stays.

As with all administrative healthcare datasets, the analyses are dependent on the completeness and accuracy of clinical coding. Smoking-related ICD-10 codes are likely underreported because they may not be systematically coded during hospitalization episodes unless considered clinically relevant for management or reimbursement. In the present study, smoking-related coding frequencies in the COPD (15.3%) and COPD with LRTI (11.4%) cohorts were lower than the reported smoking prevalence in the Swiss general population in 2022 (23.9%), despite smoking representing a major COPD risk factor [62]. Consequently, absence of smoking-related ICD-10 codes should not be interpreted as absence of tobacco exposure, and observed associations involving smoking-related coding should be interpreted accordingly. Similar considerations regarding coding completeness apply to other lifestyle factors, such as alcohol use.

Population-level mortality was incorporated through the official Swiss Federal Statistical Office population projection scenarios, which account for anticipated future changes in mortality, fertility and migration [31]. In-hospital mortality rates observed during the non-pandemic study years were additionally incorporated into the socioeconomic burden modeling using age- and sex-specific estimates. However, the model assumed stable disease-specific hospitalization, ICU admission and in-hospital mortality rates over time and did not explicitly account for future changes in critical care, discharge practices, frailty, or advances in prevention and treatment that may alter survival.

Generalizability

Though BRONCH-2035 relied on Swiss real-world data and used country-specific population projections, similar admission trends may be expected in other high-income countries experiencing aging populations. Projections could therefore inform outcomes in comparable settings. However, differences in healthcare system organization, coding rules, handling of hospital admissions, and regional prevalence variations should be considered.

Policy and Practice Implications

BRONCH-2035 suggests that, if current patterns of diagnosis, prevention, and treatment persist, demographic change alone will substantially increase LRTI- and COPD-related hospital admissions, as well as ICU demand, inpatient healthcare expenditures and healthcare-related GHG emissions. These findings support proactive health system planning, investment in mitigation strategies via guideline-concordant LRTI and COPD management, chronic care programs, and targeted expansion of hospital and ICU bed capacity in high-burden regions. The additional canton-level projections demonstrated regional heterogeneity in projected burden increases, further supporting the need for region-specific healthcare planning and resource allocation strategies. LRTIs are a risk factor for the onset of acute COPD exacerbations and increasing LRTI frequency and severity aggravate disease burden, mortality, and admissions [63]. Guidelines currently prioritize prevention via pathogen-specific vaccination, while innovative biologic therapies are in development to treat LRTIs [64, 65]. The Global Initiative for Chronic Obstructive Lung Disease (GOLD) guidelines recommend vaccination against influenza, SARS-CoV-2, RSV, and Streptococcus pneumoniae to mitigate LRTI incidence and COPD exacerbations [66–68]. Swiss practice mirrors these recommendations [65]. Evidence of the potential of prevention in LRTIs was provided during the COVID-19 pandemic, where significant reductions in admissions due to COPD exacerbations were documented, likely via reduction of transmission of respiratory viruses due to isolation policies and restrictions [58].

Because LRTI admissions are highly seasonal, projected additional bed requirements vary throughout the year. The projections highlight the need for flexible surge capacity during winter months, including adaptable general internal or pulmonology bed capacity, seasonal staffing plans, and pathways for outpatient care. During periods of lower LRTI activity, such capacity could support elective surgical interventions and recovery, rehabilitation, or chronic disease management services. Although pandemic years were excluded from the projection modeling to avoid distortion of baseline hospitalization patterns, the findings remain relevant for healthcare preparedness planning. Increasing baseline respiratory healthcare burden may reduce hospital and ICU resilience during future infectious disease outbreaks or seasonal respiratory surges, highlighting the importance of flexible capacity planning and prevention-focused respiratory healthcare policies.

Sensitivity analyses demonstrated that relatively modest annual changes in age- and sex-specific hospitalization rates could meaningfully alter projected burden trajectories. A hypothetical 1% annual reduction in hospitalization rates attenuated projected admissions, healthcare expenditures, bed occupancy, and emissions by approximately 10% by 2035, whereas a corresponding annual increase amplified projected burden. These findings suggest that preventive interventions, including vaccination uptake, smoking reduction, optimization of outpatient respiratory care and advances in COPD management, may substantially influence future respiratory healthcare burden beyond demographic change alone.

In COPD, hospital admissions are, by GOLD definition, associated with severe exacerbations, rendering prevention a priority to mitigate healthcare burden [66]. Previous hospitalization for COPD exacerbations is a predictor for further hospital admissions [69, 70]. In patients experiencing exacerbations, GOLD recommends reducing the risk of additional hospital admissions via treatment escalation, including GDMT up-titration to triple therapy or biologics [66]. A recent study demonstrated optimised GDMT-implementation from current levels to reduce PAHs in Switzerland by up to 14%, though enabling structures are not yet in place [9]. Further, measures such as smoking and vaping cessation programs, alongside regulatory actions to improve air quality would lower risk factor exposure [60, 61].

In both diseases, integrated care structures and community-based Chronic Care Management (CCM) programs offer a path towards operationalisation of prevention, optimal GDMT-implementation and guideline-concordant care at scale. In COPD, multi-component integrated care structures and CCM programs reduce hospital admissions, bed days and emergency room visits, while shortening LOS compared to usual care [71–73]. In Switzerland, implementation of structured care programs is lacking, with adherence to GOLD guidelines suboptimal both in the inpatient and outpatient settings [74–76]. Thus, the multidimensional burden projections of this study may help support structural reforms, ultimately enabling a more resilient healthcare system.

Conclusions

BRONCH-2035 aimed to project the multidimensional burdens of hospital admissions in LRTI and COPD for Switzerland through 2035. Population growth and demographic shift alone are projected to substantially increase annual hospitalizations (+ 18.4%), ICU admissions (+ 14.3%), inpatient healthcare expenditures (+ 18.3%), and healthcare-related greenhouse gas emissions (+ 21.1%), alongside considerable socioeconomic burden. These projections represent a status quo baseline, assuming continuation of current patterns of diagnosis, prevention, and treatment. Preventive measures, guideline-concordant care, and structural reforms in chronic care delivery may help mitigate this projected burden, while adequate healthcare capacity planning remains essential.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

Thomas Campbell-James and Michel Fries conceptualized the study. Desiree Schnidrig, Thomas Campbell-James, Josia D. Schramm, Lindsay Nicholson, Michel Fries and Patrick E. Beeler generated the data for input parameters. Desiree Schnidrig, Thomas Campbell-James and Josia D. Schramm conducted statistical analysis and modeling. Noel Ackermann, Samuel C. Robson, Jörg D. Leuppi and Patrick E. Beeler validated analyses and results. Desiree Schnidrig, Thomas Campbell-James, Lindsay Nicholson and Patrick E. Beeler wrote the first draft. Desiree Schnidrig and Thomas Campbell-James edited the first draft. Desiree Schnidrig developed the figures. All authors reviewed and provided edits to the manuscript. All authors read and approved the final version of the manuscript.

Funding

This work, including the preparation of the manuscript, was funded by AstraZeneca. The journal’s Rapid Service Fee was also funded by AstraZeneca.

Data Availability

Datasets generated by the study, beyond those published in the main manuscript or Supplementary Information, are available from the corresponding author upon reasonable request. The inpatient hospital admissions dataset (Medical Statistics) is available upon request from the Swiss Federal Statistical Office (FSO) and cannot be shared by the authors. All other data used for this study are detailed or referenced in the main manuscript and the Supplementary Information file.

Declarations

Conflict of Interest

At the time of redaction, Thomas Campbell-James and Michel Fries were employed by AstraZeneca Switzerland. Desiree Schnidrig received fees from AstraZeneca Switzerland for this study. Maverex Limited received consulting fees to support the development of environmental input calculations. At the time of redaction, Lindsay Nicholson was employed by Maverex. Jörg D. Leuppi reports having received unrestricted grants from AstraZeneca Switzerland, GSK Switzerland, OM Pharma Switzerland, and Sanofi Switzerland. Patrick E. Beeler received fees from AstraZeneca for this study and for other advisory and research work. Josia D. Schramm, Noel Ackermann, and Samuel C. Robson have nothing to disclose.

Ethical Approval

Not applicable. Only anonymized data were utilized, making individual identification impossible. The study conformed with Swiss law, ethical review, and research policies, adhering to the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) guidelines.