Background: Respiratory tract infections (RTIs) are a major public health concern. Global studies published in Lancet Infect. Dis. highlight the persistent morbidity and mortality from RTIs, with upper- and lower-RTIs collectively accounting for more than 100 million disability-adjusted-life-years per year. During menopause, hormonal changes alongside other factors increase the risk for illnesses, such as RTIs, COPD, cardiovascular disease, and diabetes. However, it remains unknown how hormone-replacement therapy during menopause might impact the frequency or severity of RTIs. While hormone replacement therapy (HRT) is often prescribed for menopausal symptom relief, its potential impact on RTI risk and severity has not been examined. Objective: This observational cohort study aims to compare and predict the risk of RTI among postmenopausal women, with a particular focus on the influence of HRT. The principal aim is to compare the rates and severity of respiratory tract infections in postmenopausal women taking or not taking HRT. The secondary aims are to characterize risk factors for RTI in postmenopausal women and identify signals in wearable data that predict the onset of an RTI before symptoms become apparent. Methods: 400 women aged 40-60 will be studied, stratified into two groups: postmenopausal women taking HRT, and postmenopausal women not taking HRT. Participants will each be followed for six months, with RTI episodes recorded through self-reporting and confirmed by laboratory tests. Wearable devices will continuously monitor physiological parameters (e.g., heart rate, sleep patterns), and questionnaires will assess lifestyle factors, medical history, and environmental exposure. Statistical modeling and machine learning approaches will be used to analyze infection predictors and develop a model that predicts the risk of onset of an RTI. Impact: Half of the world's population inevitably undergoes menopause, and this important life transition has wide-ranging impacts on women's health and quality of life for decades. Studies show that women spend more of their lives in poor health than men, with far-reaching impacts on a woman's participation in society, career performance, and ability to care for other family members. A better understanding of risk factors for respiratory infections in menopausal women and whether hormone-replacement therapy influences RTIs will contribute much-needed knowledge to enable better health management strategies for women. Furthermore, an "early-warning" system based on wearable signals will provide a valuable tool for quick intervention and to reduce the spread of infectious illnesses. Such an "early-warning" system will subsequently be tested for applicability across a broader representation of society as a preventive health measure and tool for pandemic preparedness. Conclusion: Findings will enhance understanding of RTI risk and management in menopausal women and contribute to the development of personalized prevention strategies. Future applications include a wearable-based medical device for real-time RTI risk assessment, potentially reducing antibiotic overuse and improving healthcare efficiency. By enabling early detection and risk stratification, this study paves the way for a proactive and personalized approach to respiratory health in postmenopausal women, ultimately shifting the focus to prevention.
Study Type
OBSERVATIONAL
Enrollment
400
University Hospital of Bern, Department of Gynecological Endocrinology & Reproductive Medicine
Bern, Switzerland
RECRUITINGNumber of Respiratory Tract Infection (RTI) Episodes per Participant Over 6 Months
Count of RTI episodes per participant during the 6-month observation period. An RTI episode is defined as the period beginning with participant-reported symptom onset and ending when: * all symptoms return to baseline, or * a maximum of 10 days of residual symptoms has elapsed. Symptom presence is documented via daily responses to the Wisconsin Upper Respiratory Symptoms Survey-21 (WURSS-21).
Time frame: 6 months from enrollment (per participant)
Cumulative Symptom Severity Score of All RTI Episodes Over 6 Months
Cumulative RTI severity, calculated as the sum of daily WURSS-21 scores across all RTI episodes occurring during the 6-month observation period. The WURSS-21 (Wisconsin Upper Respiratory Symptom Survey-21) is a validated instrument containing 21 items scored from 0 to 7 each: * 0 = "not at all" (no symptom / no functional impact) * 7 = "severe" Higher scores indicate worse symptoms and greater functional impairment. Per-episode scoring begins on the first day of participant-reported symptom onset and continues until resolution or up to 10 days of persistent residual symptoms.
Time frame: 6 months from enrollment (per participant)
Daily Individual Probability of Respiratory Tract Infection (RTI) Onset Based on Integrated Participant-Level and Environmental Data
For each participant and each study day, a continuous RTI-risk probability (unit: probability, 0-1) will be generated by statistical/machine-learning models. Predictors include: * wearable-derived physiological parameters (resting heart rate, respiration rate, sleep phase durations, step count, stress index) * questionnaire-derived behavioral, psychosocial, medical, and social variables * laboratory-measured biological variables (e.g., hormone concentrations, in original units) environmental/epidemiological variables (temperature \[°C\], humidity \[%\], PM10 \[µg/m³\], circulating respiratory viruses detected in national surveillance). Unit of Measure: Probability (0-1) per participant per day.
Time frame: 6 months after enrollment (per participant)
Correlation Between Participant-Level Characteristics and Number of RTI Episodes
The RTI count is defined as the number of RTI episodes per participant over 6 months, as ascertained through: * daily WURSS-21 reporting, * Jackson-scale symptom onset rules, * nasal swab confirmation when available. For each participant characteristic (predictor), the association will be quantified via regression coefficients from generalized linear models with Poisson distribution. Predictors include: * biological measures (e.g., hormone concentrations, in pg/mL or IU/L), * medical history variables (binary/continuous), * psychosocial questionnaire scales (e.g., PHQ-9 score, 0-27), * lifestyle/behavior measures (sleep duration from wearable, minutes/day; step - count/day; stress index score 1-100), * social context variables (household size: number of cohabitants). Unit of Measure: Regression coefficient (unitless) per predictor describing effect on RTI count.
Time frame: Time Frame: 6 months after enrollment (per participant)
Correlation Between Participant-Level Characteristics and Cumulative RTI Severity
Cumulative RTI severity is defined as the sum of daily WURSS-21 scores (range 0-140 per day) across all RTI days during the 6-month observation period. Associations with participant characteristics will be quantified as regression coefficients from generalized linear models assuming a Tweedie compound-Poisson distribution. Predictor variables and measurement tools (same categories as 4): * hormone concentrations (IU/L or pg/mL), * medical history variables, * psychosocial scales (PHQ-9), * lifestyle/behavior variables (sleep, steps, stress index), * social context variables (household size). Unit of Measure: Regression coefficient (unitless) per predictor describing effect on cumulative WURSS-21 severity score.
Time frame: 6 months after enrollment (per participant)
Aggregate Pre-Symptomatic Deviation of Physiological and Questionnaire-Derived Signals During the 14 Days Prior to RTI Symptom Onset
For each confirmed RTI episode, wearable-derived parameters and weekly questionnaire variables will be evaluated in the 14 days preceding the first reported RTI symptoms (WURSS-21 start date). The following raw indicators will be extracted but converted into standardized deviations relative to each participant's personal 30-day baseline (unit: z-score, unitless): * resting heart rate (beats/min), * respiration rate (breaths/min), * sleep structure (minutes in REM/light/deep), * stress index (1-100 score), * step count (steps/day), * participant-reported stress level (Likert scale), * sleep quality (Likert scale), * lifestyle behaviors (binary/ordinal). These will be aggregated into one composite pre-symptomatic deviation score, calculated as the mean z-standardized deviation of all included signals across the 14-day pre-onset window. Unit of Measure: Composite standardized deviation score (unitless z-score).
Time frame: 6 months after enrollment (per participant)
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.