A survey of hand hygiene practices among Indian medical undergraduates.

Background: Hand hygiene is the simplest and most effective method to prevent hospital acquired infections (HAI). Compliance with hand hygiene among health providers is, however, widely variable due to significant knowledge gaps and behavioral issues. The knowledge and practices of hand hygiene among medical undergraduate students, the future health workforce in the current COVID-19 pandemic will enable a reality check and ensure course correction at an early stage of their learning. Methods: An online questionnaire-based survey based on a standard, validated WHO 25-point (multiple-choice question and Yes/No answers) hand hygiene survey was undertaken among medical undergraduate students in an urban city from April to June 2020. Each correct response was awarded 1 point, and an incorrect response was awarded 0 points. The maximum achievable score was 25, and the minimum score was 0. Results: A total of 457 students with a mean age 20.24 ± 1.27 years completed the survey. A total of 415 (90.8%) students received hand hygiene training in the past three years. Overall 27.6% (n = 126) students had good knowledge, 65.4% (n = 299) had moderate and 7% (n = 32) had poor knowledge of hand hygiene practices. Conclusions: There were significant knowledge gaps regarding hand hygiene among medical undergraduate students in the midst of the COVID-19 pandemic. There is a need to strengthen existing teaching methods to positively impact behavioral change and potentially translate into better hand hygiene compliance among the future health workforce of the country.

Computed tomography severity score as a predictor of disease severity and mortality in COVID-19 patients: A systematic review and meta-analysis.

BACKGROUND: Prediction of outcomes in severe COVID-19 patients using chest computed tomography severity score (CTSS) may enable more effective clinical management and early, timely ICU admission. We conducted a systematic review and meta-analysis to determine the predictive accuracy of the CTSS for disease severity and mortality in severe COVID-19 subjects. METHODS: The electronic databases PubMed, Google Scholar, Web of Science, and the Cochrane Library were searched to find eligible studies that investigated the impact of CTSS on disease severity and mortality in COVID-19 patients between 7 January 2020 and 15 June 2021. Two independent authors looked into the risk of bias using the Quality in Prognosis Studies (QUIPS) tool. RESULTS: Seventeen studies involving 2788 patients reported the predictive value of CTSS for disease severity. The pooled sensitivity, specificity, and summary area under the curve (sAUC) of CTSS were 0.85 (95% CI 0.78-0.90, I2 =83), 0.86 (95% CI 0.76-0.92, I2 =96) and 0.91 (95% CI 0.89-0.94), respectively. Six studies involving 1403 patients reported the predictive values of CTSS for COVID-19 mortality. The pooled sensitivity, specificity, and sAUC of CTSS were 0.77 (95% CI 0.69-0.83, I2 = 41), 0.79 (95% CI 0.72-0.85, I2 = 88), and 0.84 (95% CI 0.81-0.87), respectively. DISCUSSION: Early prediction of prognosis is needed to deliver the better care to patients and stratify them as soon as possible. Because different CTSS thresholds have been reported in various studies, clinicians are still determining whether CTSS thresholds should be used to define disease severity and predict prognosis. CONCLUSION: Early prediction of prognosis is needed to deliver optimal care and timely stratification of patients.  CTSS has strong discriminating power for the prediction of disease severity and mortality in patients with COVID-19.

Hydroxychloroquine/Chloroquine Prophylaxis among Health-care Workers: Was it Really Preventive? - Evidence from a Multicentric Cross-sectional Study.

Background: Despite the low level of clinical evidence in hydroxychloroquine (HCQ) favor, it was prescribed for pre- and post-exposure prophylaxis in India and worldwide. In absence of a large randomized control trial, the evidence needs to be generated through observation study, hence the study was conducted to find the evidence for prophylaxis of HCQ. Materials and Methods: A multi-centric cross-sectional study involving government hospitals was chosen for serosurvey conducted from August 21, 2020, to November 20, 2020. Questionnaire was adopted from WHO. Data about chloroquine (CQ) use among health-care workers (HCWs) were added and the duration of CQ intake was also noted. Results: A total of 2,224 HCWs were recruited. The mean duration of time of taking HCQ was 7.1 weeks (standard deviation ± 6.1 weeks, median = 4 weeks with IQR, 3-10 weeks). Training on personal protective equipment (PPE), knowledge of handwashing, direct care to the patient, availability of alcohol hand rub, close contact with the patient, duration of contact, and usage of PPE were associated with HCQ intake. The antibody formation in group taking HCQ was 16.9% compared to 19.8% not taking it (P = 0.08). The Chi-square for linear trend for weeks of HCQ intake and antibody formation. However, the same was not statistically significant (Chi-square = 3.61, P = 0.06). Conclusion: Our study did not find a statistically significant association in the large multicentric study. The absolute difference of 2.9% in the two groups may not be sufficient to warrant its widespread use for prophylaxis.

Findings of second multicentric follow-up serosurvey among Health Care Workers in government hospitals.

Background: The change in serological status of community may be used as input for guiding the public health policy. Hence, the present study was conducted to determine change in seroprevalence of COVID-19 among healthcare workers (HCWs). Methods: From the baseline multicentric study sample, a subsample was followed up, and a seroepidemiological study was conducted among them between 6 and 22 weeks after the second dose of the vaccination. Multistage population proportion to size sampling was performed for the selection of subsample of HCWs. The serosurvey was conducted using the enzyme-linked immunosorbent assay-based IgG antibody test (COVID KAVACH). Results: Follow-up serological testing was done in subsample of 1122 participants of original 3253 participants. The mean age of the participants was 34.6 (8.13) years. A total of 300 (26.7%) participants were females. The seroprevalence was 78.52, (95%CI:76-80.1). Among those who were seronegative at initial test, 708 (77.04%) were seroconverted. Those who were not seroconverted (241 (21.5%)) have longer duration from the second dose of the vaccination (93 (31.4) vs. 56 (38.4); p value < 0.001). The COVID-19 infection was significantly associated with seropositive status and being a medical staff was associated with remaining seronegative on follow-up. The higher age (≥50 years) was found to be significantly associated with seroreversion. Conclusion: Four in five HCWs had detectable antibodies. Seroepidemiological studies carry vital information to control the public health response in the course of the pandemic. The study can also further help as a platform to study the seroconversion and effect of vaccination among HCWs for newer variants of SARS-CoV-2.

Epidemiological analysis of SARS-COV-2 B.1.617.2 (delta variant) transmission in an educational institute.

Background: Despite having an effective COVID-19 vaccine, the COVID-19 pandemic is far from over and the delta variant continues to cause havoc across several continents. The present study was conducted to analyze and describe the occurrence of COVID-19 cases among completely vaccinated individuals. Methods: In an educational institute in Western Maharashtra, we analyzed a cluster of RTPCR positive COVID-19 cases among fully vaccinated students which occurred in 12 days. The cases were linked to a series of curricular and co-curricular events in the institute. A detailed epidemiological investigation and genome sequencing of cases were conducted. IgG antibodies against S1 protein of novel SARS-CoV-2 were estimated for cases and age, sex, and vaccination status matched controls. Results: All 37 identified cases were mild COVID. 188 high risk (HR) contacts of the cases were identified. The overall secondary attack was 9.5%. Out of 31 cases and 50 controls, 09 (29%) cases and 08 (16%) controls were found to have IgG antibodies against S1 protein of novel SARS-CoV-2 titer of more than 60 U/ml. Whole-genome sequencing of 15 samples of the cluster showed the presence of the Delta variant of SARS-CoV-2. No correlation was observed between Ct value and IgG S1 antibody titers. Conclusion: The study provides significant evidence that only vaccination alone does not completely protect against SARS-CoV-2 B.1.617.2 (Delta) variant infection. An all-encompassing multicomponent strategy involving implementation of NPIs, robust contact tracing, early identification and isolation of cases, and high vaccination coverage is the way forward for the prevention of COVID-19.

COVISHIELD (AZD1222) VaccINe effectiveness among healthcare and frontline Workers of INdian Armed Forces: Interim results of VIN-WIN cohort study.

BACKGROUND: On 16 Jan 2021, India launched its immunization program against COVID-19. Among the first recipients were 1.59 million Health Care Workers (HCWs) and Frontline Workers (FLWs) of the Indian Armed Forces, who were administered COVISHIELD (Astra Zeneca). We present an interim analysis of vaccine effectiveness (VE) estimates till 30 May 2021. METHODS: The VIN-WIN cohort study was carried out on anonymized data of HCWs and FLWs of Indian Armed Forces. The existing surveillance system, enhanced for COVID-19 monitoring, was sourced for data. The cohort transitioned from Unvaccinated (UV) to Partially Vaccinated (PV) to Fully Vaccinated (FV), serving as its own internal comparison. Outcomes studied in the three groups were breakthrough infections and COVID related deaths. Incidence Rate Ratio (IRR) was used to compare outcomes among the three groups to estimate VE. RESULTS: Data of 1,595,630 individuals (mean age 27.6 years; 99% male) over 135 days was analysed. Till 30 May 21, 95.4% and 82.2% were partially and fully vaccinated. The UV, PV and FV compartments comprised 106.6, 46.7 and 58.7 million person-days respectively. The number of breakthrough cases in the UV, PV and FV groups were 10061, 1159 and 2512; while the deaths were 37, 16 and 7 respectively. Corrected VE was 91.8-94.9% against infections. CONCLUSION: Interim results of the VIN-WIN cohort study of 1.59 million HCWs and FLWs of Indian Armed Forces showed a ∼93% reduction in COVID-19 breakthrough infections with COVISHIELD vaccination.

Seroconversion among COVID-19 patients admitted in a dedicated COVID hospital: A longitudinal prospective study of 1000 patients.

BACKGROUND: The immune response after SARS-CoV-2 is complex and may be affected by severity of the disease, race, etc. The present study was conducted to assess the serial antibody response among the COVID-19 patients admitted in the hospital. METHODS: The study was conducted between July and October 2020 in a dedicated COVID-19 hospital. All consented patients underwent serial testing of antibodies using a rapid chromatographic immunoassay-based qualitative IgG/IgM kit every third day until their discharge or death. The data about age, sex, severity of disease, length of stay, onset of symptoms, date of molecular testing were also collected. Appropriate statistical tests were used. RESULTS: The mean age of 1000 COVID-19 patients was 47.5 ± 17.9 years. Out of the total, 687 (68.7%) were males. With respect to severity, 682 (68.2%) were asymptomatic/mild, 200 (20%) were moderate and 118 (11.8%) were severe cases. The seroconversion percentage increased from 12.8% to 97.9% and 16.3% to 80.9% for IgG and IgM respectively in 21 days. The median time for seroconversion was 10 days (IQR:6-12 days) for IgG and eight days (IQR: 6-11 days) for IgM. At the time of discharge (median nine days), detectable IgG and IgM antibodies were present in 502 (52.46%) and 414 (43.26%) participants respectively. Seroconversion was associated with days after the symptoms, increasing severity of the disease and the presence of co-morbidity. CONCLUSION: Seroconversion increased during the period of observation. The severe/moderate cases of COVID-19 tend to have an early seroconversion as compared to the asymptomatic/mild cases. Only half of the patients were seroconverted at discharge.

Mental health impact of COVID-19 among health-care workers: An exposure-based cross-sectional study.

BACKGROUND: Health-care workers (HCWs) in COVID-19 pandemic have faced the major impact in providing care to infected persons. Most of the studies on mental health impact among HCW have not incorporated paramedical staff. Furthermore, they have not compared psychological morbidity among HCW on the basis of high COVID exposure (HCE) and low COVID exposure (LCE). To address the above gap, this study aimed at evaluating mental health impact among HCW and its associated risk factors. MATERIALS AND METHODS: Consecutive 200 HCWs each in the HCE and LCE groups (between the age group of 18 and 60 years) were enrolled from two tertiary care hospitals providing COVID-19 treatment from August 1, 2020. After collection of sociodemographic data, participants were administered Impact of Event Scale-Revised, Patient Health Questionnaire-9, and Generalized Anxiety Disorder-7 Scale. RESULTS: The difference in stress, depression, and anxiety symptom scores between the HCE and LCE groups was not statistically significant. Multiple regression analysis revealed that female sex was associated with higher scores in all the three domains. The effect remained significant even after adjusting for effect of other risk factors. CONCLUSION: HCE or LCE was associated with similar impact in terms of stress, depression, and anxiety among HCWs. Female HCWs had a higher prevalence of stress, depression, and anxiety. Doctors, nurses, and paramedics had a similar prevalence of stress, depression, and anxiety when odds were adjusted.

Deep Learning Approach for Analyzing the COVID-19 Chest X-Rays.

PURPOSE: The purpose of this study is to analyze the utility of Convolutional Neural Network (CNN) in medical image analysis. In this study, deep learning (DL) models were used to classify the X-ray into COVID, viral pneumonia, and normal categories. MATERIALS AND METHODS: In this study, we have compared the results 9 layers CNN model (9 LC) developed by us with 2 transfer learning models (Visual Geometry Group) 16 and VGG19. Two different datasets used in this study were obtained from the Kaggle database and the Radiodiagnosis department of our institution. RESULTS: In our study, VGG16 yields the highest accuracy among all three models for different datasets as the Kaggle dataset-94.96% and the department of Radiodiagnosis dataset 85.71%. Although, the precision was found better while using 9 LC and VGG19 for both datasets. CONCLUSIONS: DL can help the radiologists in the speedy prediction of diseases and detecting minor features of the disease which may be missed by the human eye. In the present study, we have used three models, i.e.,, CNN with 9 LCs, VGG16, and VGG19 transfer learning models for the classification of X-ray images with good accuracy and precision. DL may play a key role in analyzing the medical image dataset.

ROX index as a good predictor of high flow nasal cannula failure in COVID-19 patients with acute hypoxemic respiratory failure: A systematic review and meta-analysis.

PURPOSE: Prediction of high flow nasal cannula (HFNC) failure in COVID-19 patients with acute hypoxemic respiratory failure (AHRF) may improve clinical management and stratification of patients for optimal treatment. We performed a systematic review and meta-analysis to determine performance of ROX index as a predictor of HFNC failure. MATERIALS AND METHODS: Systematic search was performed in electronic databases (PubMed, Google Scholar, Web of Science and Cochrane Library) for articles published till 15 June 2021 investigating ROX index as a predictor for HFNC failure. Quality In Prognosis Studies (QUIPS) tool was used to analyze risk of bias for prognostic factors, by two independent authors. RESULTS: Eight retrospective or prospective cohort studies involving 1301 patients showed a good discriminatory value, summary area under the curve (sAUC) 0.81 (95% CI, 0.77-0.84) with sensitivity of 0.70 (95% CI, 0.59-0.80) and specificity of 0.79 (95% CI, 0.67-0.88) for predicting HNFC failure. The positive and negative likelihood ratio were 3.0 (95% CI, 2.2-5.3) and 0.37 (95% CI, 0.28-0.50) respectively, and was strongly associated with a promising predictive accuracy (Diagnostic odds ratio (DOR) 9, 95% CI, 5-16). CONCLUSION: This meta-analysis suggests ROX index has good discriminating power for prediction of HFNC failure in COVID-19 patients with AHRF.

Vaccination status and COVID-19 related mortality: A hospital based cross sectional study.

BACKGROUND: Efficacy of vaccines studied in clinical trial settings are likely to be different from their effectiveness in a real-world scenario. Indian Armed Forces launched its vaccine drive against COVID-19 on 16 Jan 2021. This study evaluated the effect of vaccination on mortality amongst hospitalized COVID patients. METHODS: A cross sectional study was done on all admitted moderate to severe COVID-19 patients at a designated COVID hospital in New Delhi. The primary outcome assessed the association of being fully vaccinated with mortality. Unadjusted odds ratios (OR) (with 95% CI) was performed for each predictor. Logistic regression was used for multivariable analysis and adjusted odds ratios obtained. RESULTS: The 1168 patients included in the study had a male preponderance with a mean age of 54.6 (± 17.51) years. A total of 266 (23%) patients were partially vaccinated with COVISHIELD® and 184 (16%) were fully vaccinated. Overall, 518 (44.3%) patients had comorbidities and 332 (28.4%) died. Among those fully vaccinated, there was 12.5% (23/184) mortality while it was 31.45 % (309/984) among the unvaccinated (OR 0.3, 95% CI 0.2 to 0.5, p<0.0001). In a logistic regression model, complete vaccination status and younger age were found to be associated with survival. CONCLUSIONS: Vaccination with two doses of COVISHIELD® was associated with lower odds of mortality among hospitalized patients with moderate to severe COVID.

Conundrum of re-positive COVID-19 cases: A systematic review of case reports and case series.

BACKGROUND: The systematic review was conducted to summarize and synthesize evidence from all available case series and case reports published on re-positive COVID-19 cases. METHODS: The systematic review was registered with Prospero (CRD42020210446). PRISMA guidelines were followed for conducting the systematic review. Inclusion criteria for studies included case reports and case series which have documented cases of positive reverse transcriptase polymerase chain reaction (RT-PCR) after a period of clinical improvement or a negative RT-PCR report. Reviews, opinions, and animal studies were excluded. Methodological quality was assessed using the modified Murad scale. RESULTS: A total of 30 case reports/case series were included in the study, wherein a total of 219 cases were included. In re-positive cases, the age range varied from 10 months to 91 years. The pooled proportion of positive cases after follow-up using random-effects was 12% (95% confidence interval [CI]: 09%-15%). Among the re-positives, a total of 57 cases (26%) had comorbidities. A total of 51 (23.3%) and 17 (7.8%) re-positive cases had been treated with antivirals and corticosteroids, respectively. Only a few studies have confirmed the presence of antibodies after the first episode. Studies that included contact tracing of re-positives did not find any positive cases among close contacts of re-positive cases. CONCLUSION: The systemic review found that reinfection is a possibility within 123 days of a negative RT-PCR test in a small number of cases of COVID-19. This has wider ramifications in framing clinical, preventive, and public health policy guidelines.

Systematic review of predictive mathematical models of COVID-19 epidemic.

BACKGROUND: Various mathematical models were published to predict the epidemiological consequences of the COVID-19 pandemic. This systematic review has studied the initial epidemiological models. METHODS: Articles published from January to June 2020 were extracted from databases using search strings and those peer-reviewed with full text in English were included in the study. They were analysed as to whether they made definite predictions in terms of time and numbers, or contained only mathematical assumptions and open-ended predictions. Factors such as early vs. late prediction models, long-term vs. curve-fitting models and comparisons based on modelling techniques were analysed in detail. RESULTS: Among 56,922 hits in 05 databases, screening yielded 434 abstracts, of which 72 articles were included. Predictive models comprised over 70% (51/72) of the articles, with susceptible, exposed, infectious and recovered (SEIR) being the commonest type (mean duration of prediction being 3 months). Common predictions were regarding cumulative cases (44/72, 61.1%), time to reach total numbers (41/72, 56.9%), peak numbers (22/72, 30.5%), time to peak (24/72, 33.3%), hospital utilisation (7/72, 9.7%) and effect of lockdown and NPIs (50/72, 69.4%). The commonest countries for which models were predicted were China followed by USA, South Korea, Japan and India. Models were published by various professionals including Engineers (12.5%), Mathematicians (9.7%), Epidemiologists (11.1%) and Physicians (9.7%) with a third (32.9%) being the result of collaborative efforts between two or more professions. CONCLUSION: There was a wide diversity in the type of models, duration of prediction and the variable that they predicted, with SEIR model being the commonest type.

Seropositivity of severe acute respiratory syndrome coronavirus 2 infection among healthcare workers of the Armed Forces medical services, India: A multicentric study.

BACKGROUND: Serosurveys provide the prevalence of infection and over time will reveal the trends. The present study was conducted to estimate the seroprevalence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) among healthcare workers (HCWs) and to analyse various characteristics (risk factors) associated with SARS CoV-2 infection. METHODS: Eight government designated Corona virus disease -19 (COVID-19) hospitals were selected based on the hospital admission of patients with COVID-19 and the local epidemiological situation in the region. Multistage population proportion to size sampling was performed for the selection of HCWs. Serosurvey was conducted using the enzyme-linked immunosorbent assay-based IgG antibody test (COVID KAVACH). Bivariate and multiple logistic regression was performed to find out the factor/factors associated with the positive antibody test. RESULTS: Out of 3255 HCWs that participated in the study, data of 3253 were analysed. The seroprevalence was 19.7% (95% confidence interval: 18.5-21.3%). Factors associated were location, category of HCWs, male sex, previously tested positive by the molecular test, training on infection prevention and control, personal protective measures, handwashing technique, close contact with a patient confirmed with COVID-19, use of personal protective equipment and symptoms in the last 30 days. However, in multiple logistic regression, only location, category, previously tested positive by the molecular test and symptoms in the last 30 days were statistically significant. CONCLUSION: HCWs are vulnerable to SARS-CoV-2 infection. One in five HCWs had detectable antibodies. The presence of antibodies among HCWs may help in their placement and triage. HCWs may be advised to report early in case of any symptoms of COVID-19. Preventive measures may be targeted based on the location, with particular emphasis on ancillary workers and nurses.

Perceived stigma among the patients of coronavirus disease-19 admitted at a dedicated COVID-19 hospital in Northern India: A cross-sectional study.

BACKGROUND: Coronavirus disease-related stigma started early in the pandemic with multiple media reports highlighting the discriminatory practices toward the health-care workers, patients, and survivors. MATERIALS AND METHODS: A cross-sectional study was conducted in a designated COVID -19 hospital in Delhi from 22 October 20 to 21 November 20. All patients admitted to the hospital for more than 72 h were eligible for participation in the study. Anyone without a mobile phone and not able to read was excluded from the study. The data were collected using a pretested, prevalidated questionnaire. RESULTS: One hundred and twenty-two (92.4%) patients answered the questionnaire. All patients were male. A total of 54 (44.26%: 95% confidence interval [CI] 35.3-53.5) COVID-19 patients blamed themselves for getting the infection, compared to 68 (55.74% 95% CI: 46.5-64.7) patients who believed that acquiring the disease was not their fault. There was a statistically significant association between feeling ashamed and blaming themselves for COVID-19 (P = 0.046). A total of 19 (15.6%) have reported that they have been told that getting COVID-19 is your fault. CONCLUSION: The stigma related to COVID - 19 needs to be tackled with multipronged strategy. In India, it is not a routine to assess mental health; however, the current pandemic has brought forward the importance of stigma and other related issues during the pandemic.

Anxiety level among government employees admitted to dedicated COVID-19 hospital.

INTRODUCTION: Patients of COVID-19 patients while in a hospital may have stigma, fear, and guilt among them. However, the data on anxiety among the admitted COVID-19 patients are lacking in India and elsewhere. Hence, the study was conducted among the admitted patient of COVID-19 to describe their anxiety status. METHODS: The study was conducted as a cross-sectional study in a designated COVID-19 hospital in Delhi. The data were collected from October 22, 2020, to November 21, 2020. All patients who were admitted to the hospital for more than 72 h were eligible for participation. The data collection was done using a questionnaire. The questionnaire consists of two parts. One part was sociodemographic variables, and the other part was the Anxiety Scale. The anxiety score was collected on the Zung Self-Rating Anxiety Scale. RESULTS: A total of 132 eligible patients were admitted during the period. The questionnaire was answered by 122 (92.4%) patients. All patients were male. The patients' mean age was 33.5 years (standard deviation = 8.9 years), with a range of 21 years-65 years. The mean score of the Zung Self-Rating Scale was 29.5 (7.2), with an interquartile range of 24-33. There were only five patients (4.4%; 95% confidence interval: 1.3%-9.3%) whose scores were 45 or more, indicating mild-to-moderate anxiety. There was no statistically significant association between any sociodemographic variable and Anxiety Rating Scale. CONCLUSION: The anxiety level in the specialized population was low due to social security. The level of anxiety among health-care workers may be further explored.

Prediction of mortality by age and multi-morbidities among confirmed COVID-19 patients: Secondary analysis of surveillance data in Pune, Maharashtra, India.

Maharashtra has reported the maximum number of COVID-19 cases in India. This study was conducted to describe the predictors of death among the confirmed cases of COVID-19 by carrying out a secondary analysis of surveillance data of 11,278 lab-confirmed COVID-19 cases and admitted in dedicated COVID hospitals and dedicated COVID health-care centers between April 4, 2020, and July 17, 2020, in Pune district of Maharashtra. A total of 1270 (11.2%, 95% confidence interval [CI]: 10.7-11.9) deaths out of 11,278 patients were reported. Out of the 1270 deaths, 825 (65%) were male and 788 (62%) had one or more comorbidities. Logistic regression was done for predictors of death, and males (adjusted odds ratio: 1.6, 95% CI: 1.4-1.8), those with symptoms at the time of admission (adjusted odds ratio: 2.9, 95% CI: 2.5-3.4), and those with the presence of two or more comorbidities (adjusted odds ratio: 2.7, 95% CI: 2.2-3.4) were having a higher risk of death.