Short- and long-term effectiveness of a supervised training program in spirometry use for primary care professionals

Artificial Intelligence Applied to Forced Spirometry in Primary Care

Introduction

This study aims to create an artificial intelligence (AI) based machine learning (ML) model capable of predicting a spirometric obstructive pattern using variables with the highest predictive power derived from an active case-finding program for COPD in primary care.

Material and methods

A total of 1190 smokers, aged 30-80 years old with no prior history of respiratory disease, underwent spirometry with bronchodilation. The sample was analyzed using AI tools. Based on an exploratory data analysis (EDA), independent variables (according to mutual information analysis) were trained using a gradient boosting algorithm (GBT) and validated through cross-validation.

Results

With an area under the curve close to unity, the model predicted a spirometric obstructive pattern using variables with the highest predictive power: FEV1_theoretical_pre values. Sensitivity: 93%. Positive predictive value: 94%. Specificity: 97%. Negative predictive value: 96%. Accuracy: 95%. Precision: 94%.

Conclusion

An ML model can predict the presence of an obstructive pattern in spirometry in a primary care smoking population with no prior diagnosis of respiratory disease using the FEV1_theoretical_pre values with an accuracy and precision exceeding 90%. Further studies including clinical data and strategies for integrating AI into clinical workflow are needed.

The Impact of Metabolic Syndrome Risk Factors on Lung Function Impairment: Cross-Sectional Study

BACKGROUND

Metabolic syndrome (MetS) is a constellation of risk factors increasingly present in the world's population. People with this syndrome are at an increased risk of cardiovascular disease and type 2 diabetes mellitus. Moreover, evidence has shown that it affects different organs. MetS and its risk factors are independently associated with impaired lung function, which can be quantified through spirometric variables.

OBJECTIVE

This study aims to determine whether a high number of MetS criteria is associated with increased lung function decline.

METHODS

We conducted a descriptive cross-sectional study with a random sample of 1980 workers. Workers with acute respiratory pathology (eg, influenza), chronic respiratory pathology (eg, chronic bronchitis), or exposure to substances harmful to the lungs (eg, organic and inorganic dust) were not included. MetS was established based on harmonized criteria, and lung function was assessed according to spirometric variables. On the basis of these, classification into restrictive lung disease (RLD), obstructive lung disease, and mixed lung disease (MLD) was performed. In addition, the association between MetS and lung function was established based on analysis of covariance, linear trend analysis, and multiple linear regression.

RESULTS

MetS was associated with worse lung function according to all the spirometric parameters analyzed (percentage of predicted forced expiratory volume in 1 second: mean 83, SD 13.8 vs mean 89.2, SD 12.8; P<.001 and percentage of predicted forced vital capacity: mean 85.9, SD 11.6 vs mean 92, SD 11.3; P<.001). Moreover, those diagnosed with MetS had a higher prevalence of lung dysfunction (41% vs 21.9%; P<.001), RLD (23.4% vs 11.2%; P<.001), and MLD (7.3% vs 2.2%; P<.001). Furthermore, an increasing number of MetS criteria was associated with a greater impairment of pulmonary mechanics (P<.001). Similarly, with an increasing number of MetS criteria, there was a significant linear trend (P<.001) in the growth of the prevalence ratio of RLD (0 criteria: 1, 1: 1.46, 2: 1.52, 3: 2.53, 4: 2.97, and 5: 5.34) and MLD (0 criteria: 1, 1: 2.68, 2: 6.18, 3: 9.69, and 4: 11.37). Regression analysis showed that the alteration of all MetS risk factors, adjusted for various explanatory variables, was significantly associated with a worsening of spirometric parameters, except for forced expiratory volume in 1 second/forced vital capacity.

CONCLUSIONS

The findings have shown that an increase in cardiometabolic risk factors is associated with a more significant worsening of spirometric variables and a higher prevalence of RLD and MLD. As spirometry could be a crucial tool for monitoring patients at risk of developing chronic pathologies, we conclude that this inexpensive and easily accessible test could help detect changes in lung function in patients with cardiometabolic disorders. This highlights the need to consider the importance of cardiometabolic health in lung function when formulating public health policies.

Deep Learning-Based Analytic Models Based on Flow-Volume Curves for Identifying Ventilatory Patterns

INTRODUCTION

Spirometry, a pulmonary function test, is being increasingly applied across healthcare tiers, particularly in primary care settings. According to the guidelines set by the American Thoracic Society (ATS) and the European Respiratory Society (ERS), identifying normal, obstructive, restrictive, and mixed ventilatory patterns requires spirometry and lung volume assessments. The aim of the present study was to explore the accuracy of deep learning-based analytic models based on flow-volume curves in identifying the ventilatory patterns. Further, the performance of the best model was compared with that of physicians working in lung function laboratories.

METHODS

The gold standard for identifying ventilatory patterns was the rules of ATS/ERS guidelines. One physician chosen from each hospital evaluated the ventilatory patterns according to the international guidelines. Ten deep learning models (ResNet18, ResNet34, ResNet18_vd, ResNet34_vd, ResNet50_vd, ResNet50_vc, SE_ResNet18_vd, VGG11, VGG13, and VGG16) were developed to identify patterns from the flow-volume curves. The patterns obtained by the best-performing model were cross-checked with those obtained by the physicians.

RESULTS

A total of 18,909 subjects were used to develop the models. The ratio of the training, validation, and test sets of the models was 7:2:1. On the test set, the best-performing model VGG13 exhibited an accuracy of 95.6%. Ninety physicians independently interpreted 100 other cases. The average accuracy achieved by the physicians was 76.9 ± 18.4% (interquartile range: 70.5-88.5%) with a moderate agreement (κ = 0.46), physicians from primary care settings achieved a lower accuracy (56.2%), while the VGG13 model accurately identified the ventilatory pattern in 92.0% of the 100 cases (P < 0.0001).

CONCLUSIONS

The VGG13 model identified ventilatory patterns with a high accuracy using the flow-volume curves without requiring any other parameter. The model can assist physicians, particularly those in primary care settings, in minimizing errors and variations in ventilatory patterns.

Management of Children with Acute Asthma Attack: A RAND/UCLA Appropriateness Approach

Bronchial asthma is the most frequent chronic disease in children and affects up to 20% of the pediatric population, depending on the geographical area. Asthma symptoms vary over time and in intensity, and acute asthma attack can resolve spontaneously or in response to therapy. The aim of this project was to define the care pathway for pediatric patients who come to the primary care pediatrician or Emergency Room with acute asthmatic access. The project was developed in the awareness that for the management of these patients, broad coordination of interventions in the pre-hospital phase and the promotion of timely and appropriate assistance modalities with the involvement of all health professionals involved are important. Through the application of the RAND method, which obliges to discuss the statements derived from the guidelines, there was a clear increase in the concordance in the behavior on the management of acute asthma between primary care pediatricians and hospital pediatricians. The RAND method was found to be useful for the selection of good practices forming the basis of an evidence-based approach, and the results obtained form the basis for further interventions that allow optimizing the care of the child with acute asthma attack at the family and pediatric level. An important point of union between the primary care pediatrician and the specialist hospital pediatrician was the need to share spirometric data, also including the use of new technologies such as teleconsultation. Monitoring the progress of asthma through spirometry could allow the pediatrician in the area to intervene early by modifying the maintenance therapy and help the patient to achieve good control of the disease.

Conocimiento de la enfermedad pulmonar obstructiva crónica (EPOC) en la población gallega: estudio «CoñecEPOC»

Objectives

In 2011, only 18% of the population in Galician knew the COPD. Since then, activities have been carried out to publicize this disease. The objective of this study was to evaluate the current situation regarding the knowledge of COPD in the Galician population.

Methodology

Cross-sectional study, through telephone surveys. Variables included in the questionnaire, related to knowledge of the disease, were analyzed.

Results

872 respondents, 53% women, mean age 54 years. 63% with secondary/university studies. 40% has knowledge of COPD. In contrast, more than 90% of respondents know other high-frequency diseases (diabetes, stroke, asthma). The factors most associated with knowledge of COPD were female gender, having secondary/university studies, and having previously performed spirometry.

Conclusions

The knowledge of COPD in the Galician population is 40% now, higher than in 2011, but it is far from that of other prevalent diseases.

Standardization of Spirometry 2019 Update. An Official American Thoracic Society and European Respiratory Society Technical Statement

Background: Spirometry is the most common pulmonary function test. It is widely used in the assessment of lung function to provide objective information used in the diagnosis of lung diseases and monitoring lung health. In 2005, the American Thoracic Society and the European Respiratory Society jointly adopted technical standards for conducting spirometry. Improvements in instrumentation and computational capabilities, together with new research studies and enhanced quality assurance approaches, have led to the need to update the 2005 technical standards for spirometry to take full advantage of current technical capabilities.Methods: This spirometry technical standards document was developed by an international joint task force, appointed by the American Thoracic Society and the European Respiratory Society, with expertise in conducting and analyzing pulmonary function tests, laboratory quality assurance, and developing international standards. A comprehensive review of published evidence was performed. A patient survey was developed to capture patients' experiences.Results: Revisions to the 2005 technical standards for spirometry were made, including the addition of factors that were not previously considered. Evidence to support the revisions was cited when applicable. The experience and expertise of task force members were used to develop recommended best practices.Conclusions: Standards and consensus recommendations are presented for manufacturers, clinicians, operators, and researchers with the aims of increasing the accuracy, precision, and quality of spirometric measurements and improving the patient experience. A comprehensive guide to aid in the implementation of these standards was developed as an online supplement.

[Quality of spirometry in Primary Care in Cantabria 10 years later. EspiroCan Study]

INTRODUCTION

To study the quality of spirometry performed in Primary Care in Cantabria (Spain) and the level of compliance with the regulations of this technique of the Spanish Society of Chest Diseases and Thoracic Surgery (SEPAR).

MATERIAL AND METHODS

Cross-sectional descriptive study in the Community of Cantabria using an original questionnaire based on the latest SEPAR regulations. The questionnaire was distributed in 2018 by internal mail to the 42 Primary Care Centres of the Cantabrian Health Service, with the coordinators of the PCC and the nurses responsible having been personally contacted by the researchers or by telephone. This is an original survey, based on the regulations of the SEPAR, which evaluates the quality of the material used (model, years of operation, calibration, cleanliness), as well as the training of personnel who perform spirometry.

RESULTS

The health personnel responsible for performing spirometry were surveyed in the 42 Primary Care Centres of the Cantabrian Health Service throughout the Cantabrian area. A mean of 564 spirometries are performed per month. A significant number (13%) of spirometers are never calibrated, and only 10.5% of the spirometers are calibrated by the personnel who perform the test in the centre itself. More than half (54, 53.7%) of these professionals have never received specific training to perform spirometry, and only 3.8% of them have experience in performing the test as recommended by SEPAR. As for the cleaning of the devices, 30% of the technicians do not clean the spirometer or the adapter.

CONCLUSIONS

10 years after our initial study lack of training is still being observed in the professionals, and probably translates into invalid spirometry.

Effects of the Spirometry Learning Module on the knowledge, confidence, and experience of spirometry operators

Our study measures effects of the Spirometry Learning Module (SLM) on health-care professionals' knowledge of spirometry test quality and perceived confidence, experience, and understanding of spirometry measurements and interpretation. Professionals from both primary and hospital-based settings enrolled in the SLM, a training model focusing on spirometry test performance and interpretation, including an online interactive learning component and a face-to-face workshop. Participants were asked to submit patient spirometry assessment worksheets for feedback on quality and interpretation. Data were collected at baseline, SLM completion (20 weeks), and 12 months after SLM completion. Knowledge of spirometry test quality was evaluated with questions relating to five case-based assessments of common spirometric patterns. Perceived confidence, experience, and knowledge in test performance were measured using a 7-point Likert scale. The Friedman test combined with post hoc analyses were used to analyse differences between baseline, 20-week, and 12-month post completion. Qualitative interviews were performed to assess reasons for non-completion. Of the 90 participants enrolled in the SLM and consented to research, 48 completed the 20-week measurement and 11 completed the 12-month measurement. Statistically significant improvements were detected in all outcomes in participants who completed the SLM to 20-week and 12-month follow-up assessments (all p values < 0.01). Barriers to completion were limited access to patients requiring spirometry, high clinic workload, and having a different spirometer at the workplace compared to the one used during SLM demonstrations. Our data suggest that participants' confidence, experience, and knowledge regarding spirometry may improve through SLM completion.

[Feasibility and benefits of training to screen for chronic obstructive pulmonary disease]

INTRODUCTION

Though still under-diagnosed, chronic obstructive pulmonary disease (COPD) currently affects nearly 3.5 million people in France. The present study presents the results of continuing medical education sessions on COPD screening by electronic mini-spirometry.

METHODS

From April 2013 to December 2015, the sessions involved 73 health professionals. The study analysed three questionnaires administered before, after, and long after sessions led by experts within a professional associative network.

RESULTS

The sessions proved efficient in increasing the participants' theoretical knowledge. It increased the percentage of correct answers regarding the nature of COPD (90 % vs. 81%), the functions, features, and outputs of mini-spirometers, and the treatment recommendations. The sessions led to non-negligible changes in everyday medical practice regarding the acquisition of a mini-spirometer (+13 devices), the presentation of COPD to the patients (+33 practitioners), the dialogue on tobacco use (+32 practitioners), vaccination (+33 practitioners), and compliance with the treatment recommendations (+43 practitioners).

CONCLUSION

These results encourage both holding and following up such sessions. The specialized professional environment ensures knowledge updates and offers subsequent assistance. Further improving these sessions will increase their benefits in terms of diagnosis, treatment, and health economy.

Family medicine physician teachers and residents' intentions to prescribe and interpret spirometry: a descriptive cross-sectional study

Background: Spirometry is the best test to demonstrate airway obstruction, but remains underused in primary care. Objectives: We assessed, among family medicine physician teachers and residents, their intention to prescribe spirometry in patients suspected of chronic obstructive pulmonary disease and their intention to interpret the results. This evaluation is based on the theoretical framework proposed by Godin et al. for the study of factors influencing healthcare professionals' behavior. Methods: Participants of this descriptive cross-sectional study were recruited from eight Family medicine units (FMUs) of Laval University's network. They completed a 23-item self-administered questionnaire measuring their intention to prescribe and to interpret spirometry as well as some determinants of this intention (beliefs about capabilities, beliefs about consequences, social influence and moral norm). Answers to each of the items were scored on a Likert scale (score 1 to 7) where a higher score indicated a greater agreement with the statement. Results: Of the 284 eligible physicians, 104 were included. The mean score ± standard deviation of physicians' intention to prescribe spirometry (6.6 ± 0.7) was higher than to interpret the results (5.8 ± 1.5). Mean scores for all determinants of intention measured were also higher for prescription than for interpretation of spirometry. Conclusion: The results suggest that participants have a very strong intention to prescribe spirometry. Although the intention to interpret the results is positive, it is weaker than for the prescription of the test. Further studies will be needed to assess the barriers to spirometry interpretation.

Validation of the portable Air-Smart Spirometer

BACKGROUND

The Air-Smart Spirometer is the first portable device accepted by the European Community (EC) that performs spirometric measurements by a turbine mechanism and displays the results on a smartphone or a tablet.

METHODS

In this multicenter, descriptive and cross-sectional prospective study carried out in 2 hospital centers, we compare FEV1, FVC, FEV1/FVC ratio measured with the Air Smart-Spirometer device and a conventional spirometer, and analyze the ability of this new portable device to detect obstructions. Patients were included for 2 consecutive months. We calculate sensitivity, specificity, positive and negative predictive value (PPV and NPV) and likelihood ratio (LR +, LR-) as well as the Kappa Index to evaluate the concordance between the two devices for the detection of obstruction. The agreement and relation between the values of FEV1 and FVC in absolute value and the FEV1/FVC ratio measured by both devices were analyzed by calculating the intraclass correlation coefficient (ICC) and the Pearson correlation coefficient (r) respectively.

RESULTS

200 patients (100 from each center) were included with a mean age of 57 (± 14) years, 110 were men (55%). Obstruction was detected by conventional spirometry in 73 patients (40.1%). Using a FEV1/FVC ratio smaller than 0.7 to detect obstruction with the Air Smart-Spirometer, the kappa index was 0.88, sensitivity (90.4%), specificity (97.2%), PPV (95.7%), NPV (93.7%), positive likelihood ratio (32.29), and negative likelihood ratio (0.10). The ICC and r between FEV1, FVC, and FEV1 / FVC ratio measured by the Air Smart Spirometer and the conventional spirometer were all higher than 0.94.

CONCLUSION

The Air-Smart Spirometer is a simple and very precise instrument for detecting obstructive airway diseases. It is easy to use, which could make it especially useful non-specialized care and in other areas.

[Spirometry in a population of coal miners in Paipa, Colombia]

INTRODUCTION

Respiratory diseases resulting from exposure to particulate matter such as in coal mining remains a research challenge in this country and a public health issue. Spirometry is a basic test of fundamental respiratory function for the diagnosis and monitoring of these types of chronic lung diseases.

OBJECTIVE

To determine spirometric values in the coal mining municipality of Paipa and their association with age and occupational exposure times.

MATERIALS AND METHODS

We conducted a descriptive cross-sectional study. The occupational respiratory disease questionnaire of the American Thoracic Society (ATS) was completed while spirometric measurements were performed and interpreted in accordance with international recommendations for conducting the test.

RESULTS

The sample consisted of 226 coal mining workers of the municipality of Paipa. Twenty-eight subjects (12.3%) of the sample showed patterns of obstructive and restrictive respiratory disease with mild degrees of severity. Eighty subjects (35%) showed a decrease in the forced vital capacity ratio/expiratory volume in one second (FVC/FEV1). A statistically significant association between age range (p=0.002) and years of mining work (p=0.34) with the development of restrictive and obstructive disorders was found. Also, there was a statistically significant association between age range (p<0.01) and years of mining work (p<0.01) with various degrees of severity of the spirometric pattern.

CONCLUSIONS

Spirometry is a useful test for detecting the presence of respiratory disorders in the population of coal miners. The time of exposure was significantly associated with the respiratory disease exhibited by these miners.

The Danish National Database for Asthma: establishing clinical quality indicators

Asthma is one of the most common chronic diseases worldwide affecting more than 300 million people. Symptoms are often non-specific and include coughing, wheezing, chest tightness, and shortness of breath. Asthma may be highly variable within the same individual over time. Although asthma results in death only in extreme cases, the disease is associated with significant morbidity, reduced quality of life, increased absenteeism, and large costs for society. Asthma can be diagnosed based on report of characteristic symptoms and/or the use of several different diagnostic tests. However, there is currently no gold standard for making a diagnosis, and some degree of misclassification and inter-observer variation can be expected. This may lead to local and regional differences in the treatment, monitoring, and follow-up of the patients. The Danish National Database for Asthma (DNDA) is slated to be established with the overall aim of collecting data on all patients treated for asthma in Denmark and systematically monitoring the treatment quality and disease management in both primary and secondary care facilities across the country. The DNDA links information from population-based disease registers in Denmark, including the National Patient Register, the National Prescription Registry, and the National Health Insurance Services register, and potentially includes all asthma patients in Denmark. The following quality indicators have been selected to monitor trends: first, conduction of annual asthma control visits, appropriate pharmacological treatment, measurement of lung function, and asthma challenge testing; second, tools used for diagnosis in new cases; and third, annual assessment of smoking status, height, and weight measurements, and the proportion of patients with acute hospital treatment. The DNDA will be launched in 2016 and will initially include patients treated in secondary care facilities in Denmark. In the nearby future, the database aims to include asthma diagnosis codes and clinical data registered by general practitioners and specialised practitioners as well.

Screening for Chronic Obstructive Pulmonary Disease: Validity and Reliability of a Portable Device in Non-Specialized Healthcare Settings

Introduction and Objectives

The underdiagnosis of chronic obstructive pulmonary disease (COPD) could be improved through screening using portable devices simpler than conventional spirometers in specific healthcare settings to reach a higher percentage of the at-risk population. This study was designed to assess the validity and reliability of the COPD-6 portable device to screen for COPD in non-specialized healthcare settings.

Methods

Prospective cohort study to validate a diagnostic test. Three cohorts were recruited: primary care (PC), emergency services (ES) and community pharmacies (CPh). Study population: individuals with risk factors for COPD (>40 years, smoking >10 pack-years, with respiratory symptoms). The values measured using the COPD-6 were FEV1, FEV6 and the FEV1/FEV6 ratio. Subsequently, participants underwent conventional spirometry at hospital, using a post-bronchodilator FEV1/FVC value <0.7 as the gold standard criterion for the COPD diagnosis.

Results

437 participants were included, 362 were valid for the analysis. COPD was diagnosed in 114 patients (31.5%). The area under the ROC curve for the COPD-6 for COPD screening was 0.8.The best cut-off point for the FEV1/FEV6 ratio was 0.8 (sensitivity, 92.1%) using spirometry with the bronchodilator test as the gold standard. There were practically no differences in the COPD-6 performancein the different settings and also regarding age, gender and smoking status.

Conclusions

The COPD-6 device is a valid tool for COPD screening in non-specialized healthcare settings. In this context, the best cut-off point for the FEV1/FEV6 ratio is 0.8.

Economic Assessment and Budgetary Impact of a Telemedicine Procedure and Spirometry Quality Control in the Primary Care Setting

OBJECTIVE

To evaluate the economic impact of a telemedicine procedure designed to improve the quality of lung function testing (LFT) in primary care in a public healthcare system, compared with the standard method.

MATERIALS AND METHODS

The economic impact of 9,039 LFTs performed in 51 health centers (2010-2013) using telespirometry (TS) compared to standard spirometry (SS) was studied.

RESULTS

TS costs more per unit than SS (€47.80 vs. €39.70) (2013), but the quality of the TS procedure is superior (84% good quality, compared to 61% using the standard procedure). Total cost of TS was €431,974 (compared with €358,306€ for SS), generating an economic impact of €73,668 (2013). The increase in cost for good quality LFT performed using TS was €34,030 (2010) and €144,295 (2013), while the costs of poor quality tests fell by €15,525 (2010) and 70,627€ (2013).

CONCLUSION

The cost-effectiveness analysis concludes that TS is 23% more expensive and 46% more effective. Healthcare costs consequently fall as the number of LFTs performed by TS rises. Avoiding poor quality, invalid LFTs generates savings that compensate for the increased costs of performing LFTs with TS, making it a cost-effective method.

Inter-regional changes in the performance and interpretation of spirometry in Spain: 3E study

OBJECTIVES

This report shows the results of a nation-wide survey on spirometry to assess regional differences.

METHODS

Observational cross-sectional study conducted by means of a telephone survey in 805 primary care (PC) and specialized centers (SC) in all regions. The survey contacted technicians in charge of spirometry and consisted of 36 questions related to the test.

RESULTS

The results showed major differences between regions. Most centers had 1-2 spirometers. The number of spirometry tests per week ranged from 2 to 8.9 in PC and between 34.3 and 98.3 in SC. Some training had been given in most centers (63.6-100% in PC and 60.0-100% in SC) but not on a regular basis. Most centers used several short-acting bronchodilators for the bronchodilation test, but with insufficient inhalations (2.0-3.8 in PC and 2.0-3.3 in SC) and frequently incorrect waiting time (29.4-83.3% PC and 33.3-87.5% in SC). Daily calibration was not performed in all centers (0-100% in PC and 66.7-100% in SC). Significant inter-regional differences in spirometry quality criteria were observed, with 6 or more criteria met in 9.1-84.6% of PC centers and 37.5-100% in SC.

CONCLUSIONS

Our results show the current situation of spirometry in primary and specialized care in Spain, highlighting considerable variability and areas for improvement. This information should be considered by health officials to improve the quality and accessibility of such tests.