Clinical decision support system to enhance quality control of spirometry using information and communication technologies

Deep learning for spirometry quality assurance with spirometric indices and curves

Background

Spirometry quality assurance is a challenging task across levels of healthcare tiers, especially in primary care. Deep learning may serve as a support tool for enhancing spirometry quality. We aimed to develop a high accuracy and sensitive deep learning-based model aiming at assisting high-quality spirometry assurance.

Methods

Spirometry PDF files retrieved from one hospital between October 2017 and October 2020 were labeled according to ATS/ERS 2019 criteria and divided into training and internal test sets. Additional files from three hospitals were used for external testing. A deep learning-based model was constructed and assessed to determine acceptability, usability, and quality rating for FEV₁ and FVC. System warning messages and patient instructions were also generated for general practitioners (GPs).

Results

A total of 16,502 files were labeled. Of these, 4592 curves were assigned to the internal test set, the remaining constituted the training set. In the internal test set, the model generated 95.1%, 92.4%, and 94.3% accuracy for FEV₁ acceptability, usability, and rating. The accuracy for FVC acceptability, usability, and rating were 93.6%, 94.3%, and 92.2%. With the assistance of the model, the performance of GPs in terms of monthly percentages of good quality (A, B, or C grades) tests for FEV₁ and FVC was higher by ~ 21% and ~ 36%, respectively.

Conclusion

The proposed model assisted GPs in spirometry quality assurance, resulting in enhancing the performance of GPs in quality control of spirometry.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-022-02014-9.

The effects of different body positions on pulmonary function in healthy adults

Abstract Introduction: Pulmonary function testing, or spirometry, is a validated, globally recognized test that contributes to the diagnosis, staging, and longitudinal follow-up of lung diseases. The exam is most often performed in a sitting position in clinical practice; hence, there are no predicted values for its performance in other positions, such as in different decubitus. Objective: The present study aimed to evaluate the effects of position on pulmonary function test results in healthy adults. Methods: Forty-two healthy adults of both sexes, divided into male (MG) and female groups (FG), were provided respiratory questionnaires. Subsequently, the pulmonary function test was conducted to evaluate the ventilatory parameters of forced vital capacity (FVC), forced expiratory volume in 1 second (FEV1), and FEV1/FVC ratio in the sitting (S), dorsal decubitus (DD), right lateral decubitus (RLD), and left lateral decubitus (LLD) positions. A comparison of the parametric data was performed via one-way analysis of variance followed by Tukey post-hoc tests. Correlations between the S position variables along with the other positions were evaluated using the Pearson test. Results: The mean and standard error for the FVC values of the MG at positions DD (4.3 ± 0.7/L), RLD (4.1 ± 0.6/L) and LLD (4.1 ± 0.6/L) were lower when compared to S (5.05 ± 0.6 L). There was a strong positive correlation between the values of FVC, FEV1, and FEV1/FVC in the S position compared to other positions analyzed in both groups. Conclusion: Body positioning altered the parameters of the pulmonary function test in healthy adults.

Appropriateness of Overridden Alerts in Computerized Physician Order Entry: Systematic Review

BACKGROUND

The clinical decision support system (CDSS) has become an indispensable tool for reducing medication errors and adverse drug events. However, numerous studies have reported that CDSS alerts are often overridden. The increase in override rates has raised questions about the appropriateness of CDSS application along with concerns about patient safety and quality of care.

OBJECTIVE

The aim of this study was to conduct a systematic review to examine the override rate, the reasons for the alert override at the time of prescribing, and evaluate the appropriateness of overrides.

METHODS

We searched electronic databases, including Google Scholar, PubMed, Embase, Scopus, and Web of Science, without language restrictions between January 1, 2000 and March 31, 2019. Two authors independently extracted data and crosschecked the extraction to avoid errors. The quality of the included studies was examined following Cochrane guidelines.

RESULTS

We included 23 articles in our systematic review. The range of average override alerts was 46.2%-96.2%. An average of 29.4%-100% of the overrides alerts were classified as appropriate, and the rate of appropriateness varied according to the alert type (drug-allergy interaction 63.4%-100%, drug-drug interaction 0%-95%, dose 43.9%-88.8%, geriatric 14.3%-57%, renal 27%-87.5%). The interrater reliability for the assessment of override alerts appropriateness was excellent (kappa=0.79-0.97). The most common reasons given for the override were "will monitor" and "patients have tolerated before."

CONCLUSIONS

The findings of our study show that alert override rates are high, and certain categories of overrides such as drug-drug interaction, renal, and geriatric were classified as inappropriate. Nevertheless, large proportions of drug duplication, drug-allergy, and formulary alerts were appropriate, suggesting that these groups of alerts can be primary targets to revise and update the system for reducing alert fatigue. Future efforts should also focus on optimizing alert types, providing clear information, and explaining the rationale of the alert so that essential alerts are not inappropriately overridden.

Respiratory Symptoms and Lung Function among Ethiopian Women in Relation to Household Fuel Use

Exposure to household air pollution has been linked to chronic obstructive pulmonary disease, respiratory symptoms and reduced lung function. This study aims to assess respiratory symptoms and lung function among Ethiopian women in relation to exposure to HAP. We conducted a cross-sectional study among non-smoking women responsible for household cooking. Data was collected on socio-demographic characteristics, respiratory symptoms and risk factors using a validated questionnaire. Spirometry with reversibility testing was performed according to American Thoracic Society/European Respiratory Society guidelines. We used independent t-test and multivariable logistic regression to compare the means and measure association respectively. A total of 545 women participated in the study out of which 231 (42.3%) performed spirometry with at least three acceptable manoeuvres. Everyone in the rural group and 43% of the urban group were exposed to HAP from solid fuels during cooking. The odds of developing at least one respiratory symptom when compared with those using cleaner fuels are twice as high for women cooking within the living house. We also found significantly lower forced expiratory volume in the first second (FEV1) (L) among solid fuels users compared with cleaner energy users. Given the larger population settlement in the rural areas and the use of solid fuel as the only energy source, there is a higher risk of developing chronic respiratory health problems for those women in Ethiopia.

Personalized medicine for patients with COPD: where are we?

Chronic airflow limitation is the common denominator of patients with chronic obstructive pulmonary disease (COPD). However, it is not possible to predict morbidity and mortality of individual patients based on the degree of lung function impairment, nor does the degree of airflow limitation allow guidance regarding therapies. Over the last decades, understanding of the factors contributing to the heterogeneity of disease trajectories, clinical presentation, and response to existing therapies has greatly advanced. Indeed, diagnostic assessment and treatment algorithms for COPD have become more personalized. In addition to the pulmonary abnormalities and inhaler therapies, extra-pulmonary features and comorbidities have been studied and are considered essential components of comprehensive disease management, including lifestyle interventions. Despite these advances, predicting and/or modifying the course of the disease remains currently impossible, and selection of patients with a beneficial response to specific interventions is unsatisfactory. Consequently, non-response to pharmacologic and non-pharmacologic treatments is common, and many patients have refractory symptoms. Thus, there is an ongoing urgency for a more targeted and holistic management of the disease, incorporating the basic principles of P4 medicine (predictive, preventive, personalized, and participatory). This review describes the current status and unmet needs regarding personalized medicine for patients with COPD. Also, it proposes a systems medicine approach, integrating genetic, environmental, (micro)biological, and clinical factors in experimental and computational models in order to decipher the multilevel complexity of COPD. Ultimately, the acquired insights will enable the development of clinical decision support systems and advance personalized medicine for patients with COPD.

Implementation of Home Hospitalization and Early Discharge as an Integrated Care Service: A Ten Years Pragmatic Assessment

Home Hospitalization has proven efficacy, but its effectiveness and potential as an Integrated Care Service in a real world setting deserves to be explored.

OBJECTIVE

To evaluate implementation and 10 years follow-up of Home Hospitalization and Early Discharge as an Integrated Care Service in an urban healthcare district in Barcelona.

METHODS

Prospective study with pragmatic assessment. Patients: Surgical and medical acute and exacerbated chronic patients requiring admission into a highly specialized hospital, from 2006 to 2015. Intervention: Home-based individualized care plan, administered as a hospital-based outreach service, aiming at substituting hospitalization and implementing a transitional care strategy for optimal discharge. Main measurements: Emergency Department, readmissions and mortality. Patients' and professionals' perspectives, technologies and costs were evaluated.

RESULTS

4,165 admissions (71 ± 15 yrs; Charlson Index 4 ± 3). In-hospital stay was 1 (0-3) days and the length of home-based stay was 6 (5-7) days. The 30-day readmission rate was 11% and mortality was 2%. Patients, careers and health professionals expressed high levels of satisfaction (98%). At the start, the service was reimbursed at a flat rate of 918€ per patient discharged, significantly lower than conventional hospitalization (2,879€) but still allowing the hospital to keep a balanced budget. At present, there is no difference in the payment schemes for both types of services.

CONCLUSIONS

The service freed an average of 6 in-hospital days per patient. The program showed health value generation, as well as potential for synergies with community-based Integrated Care Services.

Quality and learning aspects of the first 9000 spirometries of the LifeGene study

Spirometry requires the patient to cooperate and do the manoeuvre 'right' for reliable results. Algorithms to assess test quality as well as educational recommendations for personnel are defined in guidelines. We compared the quality of forced spirometry tests performed by spirometry technicians with little or no previous experience of spirometry using spirometry systems with different modes of feedback. In both cases, the spirometry technician received general feedback on the screen based on ATS/ERS guidelines, such as 'exhale faster' and 'exhale longer'. The major difference was whether quality grading system of the complete session was available simultaneously on screen, or in the printed report afterwards. Two parts of the same population-based study (LifeGene), the pilot (LG1) and the first part (LG2) of the subsequent study, were compared retrospectively. In LG1 (on-screen grading) approved examination quality was achieved for 88% of the 10 first subjects for each spirometry technician compared to 70% in LG2 (printed grading afterwards). The corresponding values after 40 subjects was 94 % in LG1, compared to 73% in LG2, and after the first ten subjects there was no apparent quality improvement in either LG1 or LG2. The quality for LG1 is among the highest reported in the literature even though the spirometry technician were relatively inexperienced. We conclude that on-screen grading in addition to general technical quality feedback is powerful in enhancing the spirometry test session quality.

On-screen, real-time feedback for technicians and patients during spirometry raises the efficiency and quality of tests. Spirometry measures air flow passing through the lungs to determine lung function. However, the test relies on patients co-operating and performing at their best to achieve high quality results and avoid mis-diagnosis. As part of the LifeGene study at Karolinska University in Stockholm, Sweden, Mikaela Qvarfordt and co-workers trialled an automated method of providing feedback during testing to inexperienced spirometry technicians and their patients. 9000 patients were split into two groups, LG1 and LG2. LG1 received on-screen feedback plus a quality grade during testing, while LG2 received grading feedback after tests had finished. For every technician, approved quality tests were achieved for 94 per cent of their first 40 patients in LG1, compared with 73 per cent in LG2.

Automated Spirometry Quality Assurance: Supervised Learning From Multiple Experts

Forced spirometry testing is gradually becoming available across different healthcare tiers including primary care. It has been demonstrated in earlier work that commercially available spirometers are not fully able to assure the quality of individual spirometry manoeuvres. Thus, a need to expand the availability of high-quality spirometry assessment beyond specialist pulmonary centres has arisen. In this paper, we propose a method to select and optimise a classifier using supervised learning techniques by learning from previously classified forced spirometry tests from a group of experts. Such a method is able to take into account the shape of the curve as an expert would during visual inspection. We evaluated the final classifier on a dataset put aside for evaluation yielding an area under the receiver operating characteristic curve of 0.88 and specificities of 0.91 and 0.86 for sensitivities of 0.60 and 0.82. Furthermore, other specificities and sensitivities along the receiver operating characteristic curve were close to the level of the experts when compared against each-other, and better than an earlier rules-based method assessed on the same dataset. We foresee key benefits in raising diagnostic quality, saving time, reducing cost, and also improving remote care and monitoring services for patients with chronic respiratory diseases in the future if a clinical decision support system with the encapsulated classifier is to be integrated into the work-flow of forced spirometry testing.

From Systems Understanding to Personalized Medicine: Lessons and Recommendations Based on a Multidisciplinary and Translational Analysis of COPD

Systems medicine, using and adapting methods and approaches as developed within systems biology, promises to be essential in ongoing efforts of realizing and implementing personalized medicine in clinical practice and research. Here we review and critically assess these opportunities and challenges using our work on COPD as a case study. We find that there are significant unresolved biomedical challenges in how to unravel complex multifactorial components in disease initiation and progression producing different clinical phenotypes. Yet, while such a systems understanding of COPD is necessary, there are other auxiliary challenges that need to be addressed in concert with a systems analysis of COPD. These include information and communication technology (ICT)-related issues such as data harmonization, systematic handling of knowledge, computational modeling, and importantly their translation and support of clinical practice. For example, clinical decision-support systems need a seamless integration with new models and knowledge as systems analysis of COPD continues to develop. Our experience with clinical implementation of systems medicine targeting COPD highlights the need for a change of management including design of appropriate business models and adoption of ICT providing and supporting organizational interoperability among professional teams across healthcare tiers, working around the patient. In conclusion, in our hands the scope and efforts of systems medicine need to concurrently consider these aspects of clinical implementation, which inherently drives the selection of the most relevant and urgent issues and methods that need further development in a systems analysis of disease.

Integrated care services: lessons learned from the deployment of the NEXES project

Objectives

To identify barriers to deployment of four articulated Integrated Care Services supported by Information Technologies in three European sites. The four services covered the entire spectrum of severity of illness. The project targeted chronic patients with obstructive pulmonary disease, cardiac failure and/or type II diabetes mellitus.

Setting

One health care sector in Spain (Barcelona) (n = 11.382); six municipalities in Norway (Trondheim) (n = 450); and one hospital in Greece (Athens) (n = 388).

Method

The four services were: (i) Home-based long-term maintenance of rehabilitation effects (n = 337); (ii) Enhanced Care for frail patients, n = 1340); (iii) Home Hospitalization and Early Discharge (n = 2404); and Support for remote diagnosis (forced spirometry testing) in primary care (Support) (n = 8139). Both randomized controlled trials and pragmatic study designs were combined. Two technological approaches were compared. The Model for Assessment of Telemedicine applications was adopted.

Results

The project demonstrated: (i) Sustainability of training effects over time in chronic patients with obstructive pulmonary disease (p < 0.01); (ii) Enhanced care and fewer hospitalizations in chronic respiratory patients (p < 0.05); (iii) Reduced in-hospital days for all types of patients (p < 0.001) in Home Hospitalization/Early Discharge; and (iv) Increased quality of testing (p < 0.01) for patients with respiratory symptoms in Support, with marked differences among sites.

Conclusions

The four integrated care services showed high potential to enhance health outcomes with cost-containment. Change management, technological approach and legal issues were major factors modulating the success of the deployment. The project generated a business plan to foster service sustainability and health innovation. Deployment strategies require site-specific adaptations.

Clinical Decision Support Systems (CDSS) for preventive management of COPD patients

BACKGROUND

The use of information and communication technologies to manage chronic diseases allows the application of integrated care pathways, and the optimization and standardization of care processes. Decision support tools can assist in the adherence to best-practice medicine in critical decision points during the execution of a care pathway.

OBJECTIVES

The objectives are to design, develop, and assess a clinical decision support system (CDSS) offering a suite of services for the early detection and assessment of chronic obstructive pulmonary disease (COPD), which can be easily integrated into a healthcare providers' work-flow.

METHODS

The software architecture model for the CDSS, interoperable clinical-knowledge representation, and inference engine were designed and implemented to form a base CDSS framework. The CDSS functionalities were iteratively developed through requirement-adjustment/development/validation cycles using enterprise-grade software-engineering methodologies and technologies. Within each cycle, clinical-knowledge acquisition was performed by a health-informatics engineer and a clinical-expert team.

RESULTS

A suite of decision-support web services for (i) COPD early detection and diagnosis, (ii) spirometry quality-control support, (iii) patient stratification, was deployed in a secured environment on-line. The CDSS diagnostic performance was assessed using a validation set of 323 cases with 90% specificity, and 96% sensitivity. Web services were integrated in existing health information system platforms.

CONCLUSIONS

Specialized decision support can be offered as a complementary service to existing policies of integrated care for chronic-disease management. The CDSS was able to issue recommendations that have a high degree of accuracy to support COPD case-finding. Integration into healthcare providers' work-flow can be achieved seamlessly through the use of a modular design and service-oriented architecture that connect to existing health information systems.