Dataset Information

Power outage mediates the associations between major storms and hospital admission of chronic obstructive pulmonary disease

ABSTRACT:

Background

Chronic obstructive pulmonary disease (COPD) is the third-leading cause of death worldwide with continuous rise. Limited studies indicate that COPD was associated with major storms and related power outages (PO). However, significant gaps remain in understanding what PO’s role is on the pathway of major storms-COPD. This study aimed to examine how PO mediates the major storms-COPD associations.

Methods

In this time-series study, we extracted all hospital admissions with COPD as the principal diagnosis in New York, 2001–2013. Using distributed lag nonlinear models, the hospitalization rate during major storms and PO was compared to non-major storms and non-PO periods to determine the risk ratios (RRs) for COPD at each of 0–6 lag days respectively after controlling for time-varying confounders and concentration of fine particulate matter (PM2.5). We then used Granger mediation analysis for time series to assess the mediation effect of PO on the major storms-COPD associations.

Results

The RRs of COPD hospitalization following major storms, which mainly included flooding, thunder, hurricane, snow, ice, and wind, were 1.23 to 1.49 across lag 0–6 days. The risk was strongest at lag3 and lasted significantly for 4 days. Compared with non-outage periods, the PO period was associated with 1.23 to 1.61 higher risk of COPD admissions across lag 0–6 days. The risk lasted significantly for 2 days and was strongest at lag2. Snow, hurricane and wind were the top three contributors of PO among the major storms. PO mediated as much as 49.6 to 65.0% of the major storms-COPD associations.

Conclusions

Both major storms and PO were associated with increased hospital admission of COPD. PO mediated almost half of the major storms-COPD hospitalization associations. Preparation of surrogate electric system before major storms is essential to reduce major storms-COPD hospitalization.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-12006-x.

SUBMITTER: Qu Y

PROVIDER: S-EPMC8556928 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundCOPD is the third leading cause of death in the United States, with 16 million Americans currently experiencing difficulty with breathing. Power outages could be life-threatening for those relying on electricity. However, significant gaps remain in understanding the potential impact of power outages on COPD exacerbations.Research questionThe goal of this study was to determine how power outages affect COPD exacerbations.Study design and methodsUsing distributed lag nonlinear models controlling for time-varying confounders, the hospitalization rate during a power outage was compared vs non-outage periods to determine the rate ratio (RR) for COPD and its subtypes at each of 0 to 6 lag days in New York State from 2001 to 2013. Stratified analyses were conducted according to sociodemographic characteristics, season, and clinical severity; changes were investigated in numerous critical medical indicators, including length of stay, hospital cost, the number of comorbidities, and therapeutic procedures between the two periods.ResultsThe RR of COPD hospitalization following power outages ranged from 1.03 to 1.39 across lag days. The risk was strongest at lag0 and lag1 days and lasted significantly for 7 days. Associations were stronger for the subgroup with acute bronchitis (RR, 1.08-1.69) than for cases of acute exacerbation (RR, 1.03-1.40). Compared with non-outage periods, the outage period was observed to be $4.67 thousand greater in hospital cost and 1.38 greater in the number of comorbidities per case. The average cost (or number of comorbidities) was elevated in all groups stratified according to cost (or number of comorbidities). In contrast, changes in the average length of stay (-0.43 day) and the average number of therapeutic procedures (-0.09) were subtle.InterpretationPower outages were associated with a significantly elevated rate of COPD hospitalization, as well as greater costs and number of comorbidities. The average cost and number of comorbidities were elevated in all clinical severity groups.

Project description:Long term oxygen therapy (LTOT) has a strong evidence base in COPD patients with respiratory failure, but prescribing practices are recognized to need reform to ensure appropriate use and minimize costs. In the UK, since February 2006, all Home Oxygen prescription is issued by hospitals, making respiratory specialists totally in charge of home oxygen prescription. It has been widely noted that inappropriate home oxygen, often for intermittent use ("short burst"), is frequently prescribed in patients with COPD and related conditions with the intention to prevent hospital admissions outside of evidence based LTOT guidelines. We participated in a national Lung Improvement Project aimed at making LTOT use more evidence based. We utilised this unique opportunity of studying the effect of removal of oxygen from COPD patients (who did not meet LTOT criteria) on hospital admission rates.Primary and secondary care data sources were used to identify patients with COPD in a single primary care trust who were admitted to hospital at least once due to COPD between April 2007 and November 2010. Admission rates were compared between LTOT users and non-users, adjusted for age and COPD severity. LTOT users were further studied for predictors of admission in those appropriately or inappropriately given oxygen according to NICE guidance, and for admissions before and after oxygen receipt, adjusting further for co-morbidity. Mortality and economic analyses were also conducted.Readmission was more likely in LTOT users (3.18 v 1.67 per patient, p<0.001) after adjustment for FEV1 and age by multiple regression. When stratifying by appropriateness of LTOT prescription, adjusting also for Charlson index and other covariates, FEV1 predicted admission in appropriate users but there were no predictors in inappropriate users. In longitudinal analyses admission rates did not differ either side of oxygen prescription in appropriate or inappropriate LTOT users. Specialist assessment resulted in cost savings due to reduced use of oxygen.Admission to hospital is more likely in LTOT users, independent of COPD severity. Oxygen use outside NICE guidance does not appear to prevent admissions.

Project description:BackgroundTelemonitoring of symptoms and physiological signs has been suggested as a means of early detection of chronic obstructive pulmonary disease (COPD) exacerbations, with a view to instituting timely treatment. However, algorithms to identify exacerbations result in frequent false-positive results and increased workload. Machine learning, when applied to predictive modelling, can determine patterns of risk factors useful for improving prediction quality.ObjectiveOur objectives were to (1) establish whether machine learning techniques applied to telemonitoring datasets improve prediction of hospital admissions and decisions to start corticosteroids, and (2) determine whether the addition of weather data further improves such predictions.MethodsWe used daily symptoms, physiological measures, and medication data, with baseline demography, COPD severity, quality of life, and hospital admissions from a pilot and large randomized controlled trial of telemonitoring in COPD. We linked weather data from the United Kingdom meteorological service. We used feature selection and extraction techniques for time series to construct up to 153 predictive patterns (features) from symptom, medication, and physiological measurements. We used the resulting variables to construct predictive models fitted to training sets of patients and compared them with common symptom-counting algorithms.ResultsWe had a mean 363 days of telemonitoring data from 135 patients. The two most practical traditional score-counting algorithms, restricted to cases with complete data, resulted in area under the receiver operating characteristic curve (AUC) estimates of 0.60 (95% CI 0.51-0.69) and 0.58 (95% CI 0.50-0.67) for predicting admissions based on a single day's readings. However, in a real-world scenario allowing for missing data, with greater numbers of patient daily data and hospitalizations (N=57,150, N+=55, respectively), the performance of all the traditional algorithms fell, including those based on 2 days' data. One of the most frequently used algorithms performed no better than chance. All considered machine learning models demonstrated significant improvements; the best machine learning algorithm based on 57,150 episodes resulted in an aggregated AUC of 0.74 (95% CI 0.67-0.80). Adding weather data measurements did not improve the predictive performance of the best model (AUC 0.74, 95% CI 0.69-0.79). To achieve an 80% true-positive rate (sensitivity), the traditional algorithms were associated with an 80% false-positive rate: our algorithm halved this rate to approximately 40% (specificity approximately 60%). The machine learning algorithm was moderately superior to the best symptom-counting algorithm (AUC 0.77, 95% CI 0.74-0.79 vs AUC 0.66, 95% CI 0.63-0.68) at predicting the need for corticosteroids.ConclusionsEarly detection and management of COPD remains an important goal given its huge personal and economic costs. Machine learning approaches, which can be tailored to an individual's baseline profile and can learn from experience of the individual patient, are superior to existing predictive algorithms and show promise in achieving this goal.Trial registrationInternational Standard Randomized Controlled Trial Number ISRCTN96634935; http://www.isrctn.com/ISRCTN96634935 (Archived by WebCite at http://www.webcitation.org/722YkuhAz).

Project description:IntroductionPatients hospitalized for chronic obstructive pulmonary disease (COPD) exacerbations are at risk of further readmissions, increased treatment costs, and excess mortality. This study evaluated inpatient admissions and readmissions in patients receiving initial maintenance therapy with umeclidinium/vilanterol (UMEC/VI) versus tiotropium (TIO).MethodsThis retrospective, matched cohort study identified patients with COPD who initiated maintenance therapy with UMEC/VI or TIO from Optum's de-identified Clinformatics Data Mart database between January 1, 2013, and December 31, 2018 (index date defined as earliest dispensing). Eligibility criteria included: ≥ 1 medical claim for COPD pre-index or on the index date; ≥ 12 months of continuous eligibility pre-index; age ≥ 40 years at index; no pre- or post-index asthma diagnosis; and no pre-index claims for medications containing inhaled corticosteroids, long-acting β2-agonists, or long-acting muscarinic antagonists. Outcomes included time to first on-treatment COPD-related inpatient admission, rate of on-treatment COPD-related admissions, and rate of all-cause and COPD-related readmissions within 30 and 90 days. Propensity score matching was used to adjust for potential confounders.ResultsMatched UMEC/VI and TIO cohorts each included 7997 patients and were balanced on baseline characteristics (mean age 70.9 years; female 47.1-47.6%). Over 12 months, patients initiating UMEC/VI had significantly reduced risk (hazard ratio [95% CI]: 0.87 [0.79, 0.96]; p = 0.006) and rates (rate ratio [95% CI]: 0.80 [0.72, 0.92]; p = 0.008) of COPD-related inpatient admissions compared with TIO. While all-cause readmission rates were similar between treatment cohorts, readmission rates among patients with an initial admission length of stay of 1-3 days were numerically lower for UMEC/VI versus TIO (30-day readmissions: 10.5% vs. 12.4%; 90-day readmissions: 15.5% vs. 19.8%). Similar patterns were observed for COPD-related readmissions.ConclusionsThese findings highlight the real-world benefits of dual therapy with UMEC/VI versus TIO in reducing inpatient admissions and readmissions in patients with COPD, which may translate to lower healthcare costs.