Validation of diagnostic coding for interstitial lung diseases in an electronic health record system in Hong Kong

Prenatal antidepressant use and risk of congenital malformations: A population-based cohort study

Previous studies examining antidepressants and congenital-malformations were primarily conducted in western countries, and many were constrained by important methodological limitations. This population-based study identified 465,069 women (including 1,705 redeemed ≥1 prescription of antidepressants during first-trimester) aged 15-50 years who delivered their first and singleton child between 2003 and 2018 in a predominantly-Chinese population in Hong Kong, using territory-wide medical-record database of public-healthcare services, and employed propensity-score fine-stratification-weighted logistic-regression analyses to evaluate risk of any major and organ/system-specific congenital-malformations following first-trimester exposure to antidepressants. Major malformation overall was not associated with any antidepressant (weighted-odds-ratio wOR, 0.88 [95 %CI, 0.44-1.76]), specific drug-class, or individual antidepressants. Exposure to any antidepressant was associated with increased risk of cardiac (wOR, 1.82 [95 %CI, 1.07-3.12]) and respiratory anomalies (wOR,4.11 [95 %CI, 1.61-10.45]). Exposure to selective-serotonin-reuptake-inhibitors (SSRI) and multiple-AD-classes were associated with respiratory and cardiac anomalies, respectively. However, these identified associations were not consistently affirmed across sensitivity analyses, precluding firm conclusion. Observed associations of specific cardiac defects with serotonin-norepinephrine-reuptake-inhibitors (SNRI), tricyclic-antidepressants (TCA) and multiple-AD-classes were noted with wide confidence-intervals, suggesting imprecise estimation. Overall, our findings suggest that first-trimester antidepressant exposure was not robustly associated with increased risk of congenital-malformations. Further research clarifying comparative safety of individual antidepressants on specific malformations is warranted.

Projecting the 10-year costs of care and mortality burden of depression until 2032: a Markov modelling study developed from real-world data

Background

Based on real-world data, we developed a 10-year prediction model to estimate the burden among patients with depression from the public healthcare system payer's perspective to inform early resource planning in Hong Kong.

Methods

We developed a Markov cohort model with yearly cycles specifically capturing the pathway of treatment-resistant depression (TRD) and comorbidity development along the disease course. Projected from 2023 to 2032, primary outcomes included costs of all-cause and psychiatric care, and secondary outcomes were all-cause deaths, years of life lived, and quality-adjusted life-years. Using the territory-wide electronic medical records, we identified 25,190 patients aged ≥10 years with newly diagnosed depression from 2014 to 2016 with follow-up until 2020 to observe the real-world time-to-event pattern, based on which costs and time-varying transition inputs were derived using negative binomial modelling and parametric survival analysis. We applied the model as both closed cohort, which studied a fixed cohort of incident patients in 2023, and open cohort, which introduced incident patients by year from 2014 to 2032. Utilities and annual new patients were from published sources.

Findings

With 9217 new patients in 2023, our closed cohort model projected the 10-year cumulative costs of all-cause and psychiatric care to reach US$309.0 million and US$58.3 million, respectively, with 899 deaths (case fatality rate: 9.8%) by 2032. In our open cohort model, 55,849-57,896 active prevalent cases would cost more than US$322.3 million and US$60.7 million, respectively, with more than 943 deaths annually from 2023 to 2032. Fewer than 20% of cases would live with TRD or comorbidities but contribute 31-54% of the costs. The greatest collective burden would occur in women aged above 40, but men aged above 65 and below 25 with medical history would have the highest costs per patient-year. The key cost drivers were relevant to the early disease stages.

Interpretation

A limited proportion of patients would develop TRD and comorbidities but contribute to a high proportion of costs, which necessitates appropriate attention and resource allocation. Our projection also demonstrates the application of real-world data to model long-term costs and mortality, which aid policymakers anticipate foreseeable burden and undertake budget planning to prepare for the care need in alternative scenarios.

Funding

Research Impact Fund from the University Grants Committee, Research Grants Council with matching fund from the Hong Kong Association of Pharmaceutical Industry (R7007-22).

Statins and risks of dementia among patients with heart failure: a population-based retrospective cohort study in Hong Kong

Background

Heart failure (HF) and dementia frequently co-exist with shared pathological mechanisms and risk factors. Our study aims to investigate the association between statin therapy and the risks of dementia and its subtypes among patients with HF.

Methods

The Hong Kong Clinical Data Analysis and Reporting System database was interrogated to identify patients with incident HF diagnosis from 2004 to 2018, using ICD 9/ICD 10 codes. Inverse probability of treatment weighting (IPTW) was used to balance baseline covariates between statin users (N = 54,004) and non-users (N = 50,291). The primary outcomes were incident all-cause dementia, including subtypes of Alzheimer's disease, vascular dementia, and unspecified dementia. Cox proportional-hazard model with competing risk regression was performed to estimate the sub-distribution hazards ratio (SHR) with corresponding 95% confidence intervals (CI) of the risks of all-cause dementia and its subtypes that are associated with statin use.

Findings

Of all eligible patients with HF (N = 104,295), the mean age was 74.2 ± 13.6 years old and 52,511 (50.3%) were male. Over a median follow-up of 9.9 years (interquartile range [IQR]: 6.4-13.0), 10,031 (9.6%) patients were diagnosed with dementia, among which Alzheimer's disease (N = 2250), vascular dementia (N = 1831), and unspecified dementia (N = 5950) were quantified separately. After IPTW, statin use was associated with a 20% lower risk of incident dementia compared with non-use (multivariable-adjusted SHR 0.80, 95% CI 0.76-0.84). Stratified by subtypes of dementia, statin use was associated with a 28% lower risk of Alzheimer's disease (SHR 0.72, 95% CI 0.63-0.82), 18% lower risk of vascular dementia (SHR 0.82, 95% CI 0.70-0.95), and a 20% lower risk of unspecified dementia (SHR 0.80, 95% CI 0.75-0.85).

Interpretation

In patients with HF, statin use was associated with a significantly lower risk of all-cause dementia and its subtypes, including Alzheimer's disease, vascular dementia, and unspecified dementia. Both randomized trials and experimental studies to validate the potential neuroprotective effect of statin are warranted.

Funding

No funding was provided for this study.

Factors influencing long-term outcomes in fibrotic interstitial lung disease (F-ILD) diagnosed through multidisciplinary discussion (MDD): a prospective cohort study

BACKGROUND

The diagnostic process for fibrotic interstitial lung disease (F-ILD) is notably intricate, necessitating a multidisciplinary discussion to achieve consensus based on both clinical and radiological features. This study investigated the shared and distinctive long-term mortality predictors among the two primary phenotypes of F-ILD, namely idiopathic pulmonary fibrosis (IPF) and connective tissue disease-associated interstitial lung disease (CTD-ILD).

METHODS

We included patients with F-ILD diagnosed from December 2018 to December 2019 and conducted follow-up assessments until February 2023. Age, gender, usual interstitial pneumonia (UIP) pattern, gender-age-physiology (GAP) score, modified Medical Research Council (mMRC) dyspnea score, antifibrotic agent use, pulmonary function test parameters, and six-minute walking test (6MWT) parameters were recorded at baseline and used as mortality predictors in a multivariate Cox regression model.

RESULTS

We enrolled 104 ILD patients. The survival rate of non-IPF patients was more than twice that of IPF patients (78.9% vs. 34%, p < 0.001), and the survival rate of patients with a GAP score of 0-2 was more than twice that of patients with a score of > 2 (93.2% vs. 36.6%, p < 0.001). Older age, male gender, definite UIP pattern, higher GAP score, higher mMRC dyspnea score, lower forced expiratory volume in one second/forced vital capacity (FEV1/FVC), shorter 6MWT distance, and lower initial and final SpO2 were also associated with higher long-term mortality (p < 0.05). In multivariable analysis, only a GAP score of > 2 (hazard ratio [HR]:16.7; 95% confidence interval [CI] 3.28-85.14; p = 0.001) and definite UIP pattern (HR: 4.08; 95% CI 1.07-15.5; p = 0.039) were significantly associated with overall mortality.

CONCLUSION

The long-term mortality rate of IPF patients was higher than that of CTD-ILD patients. The GAP score and UIP patterns were significant mortality predictors for both IPF and CTD-ILD patients.

Multidisciplinary-derived clinical score for accurate prediction of long-term mortality in fibrotic lung disease patients

BACKGROUND

Idiopathic pulmonary fibrosis (IPF) stands out as one of the most aggressive forms of interstitial lung diseases (ILDs), currently without a definitive cure. Multidisciplinary discussion (MDD) is now considered a cornerstone in diagnosing and differentiating ILD subtypes. The Gender-Age-Physiology (GAP) score, developed to assess IPF prognosis based on sex, age, forced vital capacity, and diffusion capacity for carbon monoxide (DLCO), is limited in not considering dyspnea and functional impairment during the walking test. We proposed a MDD-based clinical score for mortality prediction among those patients.

METHODS

From December 2018 to December 2019, we enrolled ILD patients with IPF and non-IPF and followed-up them till December 2020. Based on DLCO, modified Medical Research Council (mMRC) Dyspnea Scale, and six-minute walking test (6MWT) distance, a functional score was developed for mortality prediction.

RESULTS

We enrolled 104 ILD patients, 12 (11.5%) died by the one-year follow-up. In receiver operating characteristic (ROC) curve analysis, DLCO (% predicted) was the most accurate variable predicting one-year mortality with an area under curve (AUC) of 0.88 (95% confidence interval [CI] = 0.80-0.94), followed by mMRC Dyspnea Score (AUC = 0.82 [95% CI = 0.73-0.89]), 6MWT distance (AUC = 0.80 [95% CI = 0.71-0.88]), and GAP score (AUC = 0.77 [95% CI = 0.67-0.84]). Only the GAP score (hazard ratio [HR] = 1.55, 95% CI = 1.03-2.34, p = 0.0.37) and functional score (HR = 3.45, 95% CI = 1.11-10.73, p = 0.032) were significantly associated with one-year mortality in multivariable analysis.

CONCLUSION

The clinical score composite of DLCO, mMRC Dyspnea Scale, and 6MWT distance could provide an accurate prediction for long-term mortality in ILD patients, laying out a helpful tool for managing and following these patients.

Real-World Effectiveness Study of Nirmatrelvir-Ritonavir or Molnupiravir in Hospitalized Unvaccinated Patients with Chronic Respiratory Diseases and Moderate COVID-19 at Presentation

Introduction

Nirmatrelvir-ritonavir (NMV-r) and molnupiravir (MOL) were developed as out-patient anti-viral for mild COVID-19. There was limited data on their role in treating COVID-19 for hospitalized patients, especially among adult patients who are unvaccinated and had chronic respiratory diseases.

Methods

A territory-wide retrospective study was conducted in Hong Kong to compare the efficacy of NMV-r and MOL against COVID-19 in unvaccinated adult patients with asthma, chronic obstructive pulmonary disease, bronchiectasis and interstitial lung diseases presenting with moderate COVID-19 from 16th February 2022 to 15th March 2023.

Results

A total of 1354 patients were included, 738 received NMV-r and 616 received MOL. NMV-r was more effective in reducing 90-day mortality with adjusted hazard ratios (aHR) of 0.508 (95% confidence interval [CI] = 0.314-0.822, p = 0.006). Patients who received NMV-r also had significantly shorter length of stay (LOS) than those receiving MOL, with median LOS of 4 (Interquartile range [IQR] = 2-7) for NMV-r and 6 (IQR = 3-10) for MOL (p-value < 0.001). There was no statistically significant difference in the development of respiratory failure and severe respiratory failure in the two groups.

Discussion

NMV-r was more effective than MOL among unvaccinated adults with chronic respiratory diseases who were hospitalized for moderate COVID-19 without hypoxaemia on admission.

Prevalence, incidence, and survival analysis of interstitial lung diseases in Hong Kong: a 16-year population-based cohort study

Background

Published data on the epidemiology of interstitial lung disease (ILD) in Asia is scarce. Understanding the epidemiology is important for authorities in the health management planning. This study aimed to estimate the prevalence, incidence, and survival of ILD in Hong Kong from 2005 to 2020 and evaluate the change of trend over time.

Methods

In this retrospective cohort study, we identified ILD patients between 2005 and 2020 using a territory-wide electronic health record database. Prevalence, incidence rates, and age- and sex-standardised incidence rates with United Nations population in 2020 as a reference were estimated. Trends in prevalence and incidence were analysed using joinpoint regression and the average annual percent change (AAPC) was estimated. Median survival, and risk factors of mortality were evaluated using Cox proportional hazard regression.

Findings

We identified 5924 patients and included 5884 of them for analysis. The prevalence of ILD increased from 24.7 to 33.6 per 100,000 population from 2005 to 2020 with an AAPC of 1.94 (95% confidence interval, CI: 1.69-2.34). The standardized incidence rate decreased from 5.36 to 2.57 per 100,000 person from 2005 to 2020 (AAPC -3.56, 95% CI, -4.95 to -1.78). The median survival of ILD was 2.50 (95% CI, 2.32-2.69) years. Male, older age, higher Charlson comorbidity index, and IIP subtype were associated with increased mortality with statistical significance.

Interpretation

This study provided the first epidemiological evaluation of ILD in Hong Kong. Further studies on ILD in multiple Asian cities and countries are warranted.

Funding

None.

SGLT2i reduces risk of developing HCC in patients with co-existing type 2 diabetes and hepatitis B infection: A territory-wide cohort study in Hong Kong

BACKGROUND AND AIMS

Type 2 diabetes (T2D) and chronic hepatitis B infection (CHB) are risk factors of HCC. Sodium glucose co-transporter 2 inhibitors (SGLT2i) inhibit HCC oncogenesis in preclinical studies. However, clinical studies are lacking. This study aimed to evaluate the impact of SGLT2i use on incident HCC using a territory-wide cohort of exclusively patients with co-existing T2D and CHB.

APPROACH AND RESULTS

Patients with co-existing T2D and CHB between 2015 and 2020 were identified from the representative electronic database of the Hong Kong Hospital Authority. Patients with and without SGLT2i use were 1:1 matched by propensity score for their demographics, biochemistry results, liver-related characteristics, and background medications. Cox proportional hazards regression model was used to assess the association between SGLT2i use and incident HCC. A total of 2,000 patients with co-existing T2D and CHB (1,000 in each SGLT2i and non-SGLT2i group; 79.7% on anti-HBV therapy at baseline) were included after propensity-score matching. Over a follow-up of 3,704 person-years, the incidence rates of HCC were 1.39 and 2.52 cases per 100 person-year in SGLT2i and non-SGLT2i groups, respectively. SGLT2i use was associated with a significantly lower risk of incident HCC (HR 0.54, 95%CI: 0.33-0.88, p =0.013). The association remained similar regardless of sex, age, glycemic control, diabetes duration, presence of cirrhosis and hepatic steatosis, timing of anti-HBV therapy, and background antidiabetic agents including dipeptidyl peptidase-4 inhibitors, insulin, or glitazones (all p interaction>0.05).

CONCLUSIONS

Among patients with co-existing T2D and CHB, SGLT2i use was associated with a lower risk of incident HCC.

Validation of diagnostic coding for bronchiectasis in an electronic health record system in Hong Kong

BACKGROUND AND OBJECTIVE

Electronic medical record (EMR) databases can facilitate epidemiology research in various diseases including bronchiectasis. Given the diagnostic challenges of bronchiectasis, the validity of the coding in EMR requires clarification. We aimed to assess the validity of International Classification of Diseases, 9th Revision (ICD-9) code algorithms for identifying bronchiectasis in the territory-wide electronic medical health record system of Clinical Data Analysis and Reporting System (CDARS) in Hong Kong.

MATERIALS AND METHODS

Adult patients who had the diagnosis of bronchiectasis input from Queen Mary Hospital in 2011-2020 were identified using the ICD-9 code of 494 by CDARS. All patients who had high resolution computed tomography (HRCT) were reviewed by respiratory specialists to confirm the presence of bronchiectasis on HRCT.

RESULTS

A total of 19 617 patients who had the diagnostic code of bronchiectasis among all public hospitals in Hong Kong and 1866 in Queen Mary Hospital in the same period. Six hundred and forty-eight cases were randomly selected and validated using medical record and HRCT review by a respiratory specialist. The overall positive predictive value (PPV) was 92.7% (95% CI 90.7-94.7).

CONCLUSIONS

This was the first ICD-9 coding validation for bronchiectasis in Hong Kong CDARS. Our study demonstrated that using ICD-9 code of 494 was reliable to support utility of CDARS database for further clinical research on bronchiectasis.

Evaluating the Patient Boarding during Omicron Surge in Hong Kong: Time Series Analysis

The fifth wave of COVID-19 outbreaks in Hong Kong (HK) from January to March 2022 has the highest confirmed cases and deaths compared with previous waves. Severe hospital boarding (to inpatient wards) was noted in various Emergency Departments (EDs). Our objective is to identify factors associated with hospital boarding during Omicron surge in HK. We conducted a retrospective cohort study including all ED visits and inpatient (IP) ward admissions from January 1st to March 31st, 2022. Vector Autoregression model evaluated the effects of a single variable on the targeted hospital boarding variables. Admissions from elderly homes with 6 lag days held the highest positive value of statistical significance (t-stat = 2.827, P < .05) caused prolonged admission waiting time, while medical patients with 4 lag days had the highest statistical significance (t-stat = 2.530, P < .05) caused an increased number of boarding patients. Within one week after impulses, medical occupancy's influence on the waiting time varied from 0.289 on the 1st day to -0.315 on the 7th day. While occupancy of medical wards always positively affected blocked number of patients, and its response was maximized at 0.309 on the 2nd day. Number of confirmed COVID-19 cases was not the sole significant contributor, while occupancy of medical wards was still a critical factor associated with patient boarding. Increasing ward capacity and controlling occupancy were suggested during the outbreak. Moreover, streamlining elderly patients in ED could be an approach to relieve pressure on the healthcare system.

Association of Molnupiravir and Nirmatrelvir-Ritonavir with reduced mortality and sepsis in hospitalized omicron patients: a territory-wide study

This study evaluates the association between antivirals (Molnupiravir and Nirmatrelvir-Ritonavir) and all-cause and respiratory mortality and organ dysfunction among high-risk COVID-19 patients during an Omicron outbreak. Two cohorts, Nirmatrelvir-Ritonavir versus control and Molnupiravir versus control, were constructed with inverse probability treatment weighting to balance baseline characteristics. Cox proportional hazards models evaluated the association of their use with all-cause mortality, respiratory mortality, and all-cause sepsis (a composite of circulatory shock, respiratory failure, acute liver injury, coagulopathy, and acute liver impairment). Patients recruited were hospitalized and diagnosed with the COVID-19 Omicron variant between February 22, 2022 and April 15, 2022, and followed up until May 15, 2022. The study included 17,704 patients. There were 4.67 and 22.7 total mortalities per 1000 person-days in the Nirmatrelvir-Ritonavir and control groups respectively before adjustment (weighted incidence rate ratio, - 18.1 [95% CI - 23.0 to - 13.2]; hazard ratio, 0.18 [95% CI, 0.11-0.29]). There were 6.64 and 25.9 total mortalities per 1000 person-days in the Molnupiravir and control groups respectively before adjustment (weighted incidence rate ratio per 1000 person-days, - 19.3 [95% CI - 22.6 to - 15.9]; hazard ratio, 0.23 [95% CI 0.18-0.30]). In all-cause sepsis, there were 13.7 and 35.4 organ dysfunction events per 1000 person-days in the Nirmatrelvir-Ritonavir and control groups respectively before adjustment (weighted incidence rate ratio per 1000 person-days, - 21.7 [95% CI - 26.3 to - 17.1]; hazard ratio, 0.44 [95% CI 0.38-0.52]). There were 23.7 and 40.8 organ dysfunction events in the Molnupiravir and control groups respectively before adjustment (weighted incidence ratio per 1000 person-days, - 17.1 [95% CI, - 20.6 to - 13.6]; hazard ratio, 0.63 [95% CI 0.58-0.69]). Among COVID-19 hospitalized patients, use of either Nirmatrelvir-Ritonavir or Molnupiravir compared with no antiviral use was associated with a significantly lower incidence of 28-days all-cause and respiratory mortality and sepsis.

Validation of Diagnostic Coding for Asthma in an Electronic Health Record System in Hong Kong

Background

Electronic health record (EHR) databases can facilitate epidemiology research into various diseases including asthma. Given the diagnostic challenges of asthma, the validity of the coding in EHR requires clarification. We aimed to assess the validity of International Classification of Diseases, 9th Revision (ICD-9) code algorithms for identifying asthma in the territory-wide electronic medical health record system of the Clinical Data Analysis and Reporting System (CDARS) in Hong Kong.

Methods

Adult patients who had the diagnosis of asthma input from all public hospitals in Hong Kong and those from Queen Mary Hospital in 2011-2020 were identified using the ICD-9 code of 493 (493.0, 493.1, 493.2, and 493.9) by CDARS. Patients' clinical record and spirometry were reviewed by two respiratory specialists to confirm the presence of asthma in the randomly selected cases.

Results

There were 43,454 patients who had the diagnostic code of asthma among all public hospitals in Hong Kong and 1852 in Queen Mary Hospital in the same period. A total of 200 cases were randomly selected and validated using medical record and spirometry review by a respiratory specialist. The overall positive predictive value (PPV) was 85.0% (95% CI 80.1-89.9%).

Conclusion

This was the first ICD-9 code validation for CDARS (EHR) in Hong Kong on asthma. Our study demonstrated that using ICD-9 code (493.0, 493.1, 493.2 and 493.9) to identify asthma can result in a PPV that was reliable to support the utility of the CDARS database for further research on asthma among the Hong Kong population.

Validation of Diagnostic Codes to Identify Carbon Monoxide Poisoning in Taiwan’s Claims Data

Purpose: Previous studies identified the study cohort or outcome of carbon monoxide poisoning (COP) by using the relevant disease diagnosis codes in secondary databases, but the validity of diagnosis codes of COP is unclear in such secondary databases. This study aimed to evaluate the accuracy of case definitions for COP using the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and Tenth Revision, Clinical Modification (ICD-10-CM) diagnosis codes in Taiwan’s health insurance claims data.

Methods: We selected a 10% simple random sample from an original cohort of patients newly undergoing carboxyhemoglobin (COHb) testing under any clinical diagnosis at four Chang Gung Memorial Hospitals in Taiwan during 2011–2020. Two clinical doctors independently ascertained the COP diagnosis by reviewing the medical records as the reference standard. We estimated the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of various case definitions (e.g., ICD-9-CM code, ICD-10-CM code and the uses of hyperbaric oxygen therapy) in the claims data.

Results: We randomly selected 557 cases from the original cohort of 5,571 cases newly receiving COHb test in the study hospitals. We found 90, 35, and 9 cases were true-positive, false-positive, and false-negative episodes of COP, respectively, among 278 cases with an ICD-9-CM code of 986. A further 111, 34, and 6 cases were true-positive, false-positive, and false-negative episodes of COP, respectively, among 279 cases with an ICD-10-CM code of T58. Using ICD-9-CM codes, the sensitivity, specificity, PPV and NPV for COP were 90.9, 80.4, 72, and 94.1%, respectively. Using ICD-10-CM codes they were 94.9, 79, 76.6, and 95.5%, respectively. PPV typically increased when COP was the primary diagnosis and could reach 100% if patients with ICD-CM code 986 or T58 also received hyperbaric oxygen therapy during hospitalization.

Conclusion: The COP-related ICD-CM codes alone did not accurately identify COP patients, but accuracy improved after including oxygen therapy data with the ICD-CM codes in Taiwan’s claims data.

Risk of Severe Infection among Rheumatoid Arthritis Patients on Biological DMARDs: A Population-Based Cohort Study

Biological disease-modifying anti-rheumatic drugs (bDMARDs) are widely used for the management of rheumatoid arthritis, although their benefits are counterweight by an increased risk of infections. In the present study, we used administrative data to compare the risk of severe infections among different classes of bDMARDs. A retrospective cohort study was conducted using Administrative Health Databases of the Piedmont Region, Italy. Relevant data were obtained from: (1) the inhabitants registry, (2) hospital discharge records, and (3) the co-payment exemption registry and (4) drug claims registry. Fine and Gray competing risk models were fitted to evaluate the association between the use of different types of bDMARDs and occurrence of severe infection accounting for treatment interruption as competing risk. A total of 1780 new users of bDMARDs were identified. Among them, 50 hospitalizations for infection occurred during the study period. The use of Tocilizumab was associated with an increased risk of infection, compared to tumor necrosis factor (TNF) inhibitor drugs (sub-distribution hazard ratios-sHR: 2.510; 95% CI: 1.279–4.926), whereas no difference in the risk of severe infection was found for abatacept (sHR: 0.584; 95% CI: 0.234–1.457). bDMARDs treatment is generally safe in clinical practice with slight but important differences among classes. The increased risk of infection associated with tocilizumab use should be taken into account when balancing the risk and benefits of starting a treatment with this drug.