1
|
Potts S, Bergherr E, Reinke C, Griesbach C. Prediction-based variable selection for component-wise gradient boosting. Int J Biostat 2024; 20:293-314. [PMID: 38000054 DOI: 10.1515/ijb-2023-0052] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/24/2023] [Accepted: 09/18/2023] [Indexed: 11/26/2023]
Abstract
Model-based component-wise gradient boosting is a popular tool for data-driven variable selection. In order to improve its prediction and selection qualities even further, several modifications of the original algorithm have been developed, that mainly focus on different stopping criteria, leaving the actual variable selection mechanism untouched. We investigate different prediction-based mechanisms for the variable selection step in model-based component-wise gradient boosting. These approaches include Akaikes Information Criterion (AIC) as well as a selection rule relying on the component-wise test error computed via cross-validation. We implemented the AIC and cross-validation routines for Generalized Linear Models and evaluated them regarding their variable selection properties and predictive performance. An extensive simulation study revealed improved selection properties whereas the prediction error could be lowered in a real world application with age-standardized COVID-19 incidence rates.
Collapse
Affiliation(s)
- Sophie Potts
- Chair of Spatial Data Science and Statistical Learning, University of Goettingen, Goettingen, Germany
| | - Elisabeth Bergherr
- Chair of Spatial Data Science and Statistical Learning, University of Goettingen, Goettingen, Germany
| | - Constantin Reinke
- Chair of Empirical Methods in Social Science and Demography, University of Rostock, Rostock, Germany
| | - Colin Griesbach
- Chair of Spatial Data Science and Statistical Learning, University of Goettingen, Goettingen, Germany
| |
Collapse
|
2
|
Okogbenin EO, Seb-Akahomen OJ, Edeawe O, Ehimigbai M, Eboreime H, Odike A, Obagaye MO, Aweh BE, Erohubie P, Eriyo W, Inogbo C, Akhideno P, Eifediyi G, Eifediyi R, Asogun DA, Okogbenin S. Psychiatric manifestations and associated risk factors among hospitalised patients with COVID-19 in Edo State, Nigeria: a cross-sectional study. BMJ Open 2022; 12:e058561. [PMID: 35523500 PMCID: PMC9082729 DOI: 10.1136/bmjopen-2021-058561] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 11/04/2022] Open
Abstract
OBJECTIVE To estimate the prevalence of depression and anxiety and identify associated risk factors in hospitalised persons with confirmed COVID-19 in Edo, Nigeria. DESIGN A multicentre cross-sectional survey. SETTING Patients with COVID-19 hospitalised at the three government-designated treatment and isolation centres in Edo State, Nigeria. PARTICIPANTS The study was conducted from 15 April to 11 November 2020 among 489 patients with confirmed COVID-19 and in treatment and isolation centres in Edo State, Nigeria. The mean age of participants was 43.39 (SD=16.94) years. Male participants were 252 (51.5%) and female were 237 (48.5%). MAIN OUTCOME MEASURES The nine-item Patient Health Questionnaire for depression, (total score: 0-27, depression ≥10), Generalized Anxiety Disorder-7 for anxiety (total score: 0-21, anxiety ≥10), and social demographic and clinical characteristics for associated risk factors. RESULTS Of the 489 participants, 49.1% and 38.0% had depressive and anxiety symptoms, respectively. The prevalence rates of depression, anxiety and combination of both were 16.2%, 12.9% and 9.0%, respectively. Moderate-severe symptoms of COVID-19, ≥14 days in isolation, worrying about the outcome of infection and stigma increased the risk of having depression and anxiety. Additionally, being separated/divorced increased the risk of having depression and having comorbidity increased the risk of having anxiety. CONCLUSION A substantial proportion of our participants experienced depression, anxiety and a combination of both especially in those who had the risk factors we identified. The findings underscore the need to address modifiable risk factors for psychiatric manifestations early in the course of the disease and integrate mental health interventions and psychosocial support into COVID-19 management guidelines.
Collapse
Affiliation(s)
- Esther Osemudiamen Okogbenin
- Department of Mental Health, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Department of Mental Health, Ambrose Alli University College of Medicine, Ekpoma, Edo State, Nigeria
| | | | - Osahogie Edeawe
- Institute of Lassa Fever Research and Control, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
| | - Mary Ehimigbai
- Department of Mental Health, University of Benin Teaching Hospital, Benin City, Edo State, Nigeria
| | - Helen Eboreime
- Department of Mental Health, Edo University Iyamho, Iyamho, Edo State, Nigeria
| | - Angela Odike
- Department of Child Health, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Department of Child Health, Ambrose Alli University College of Medicine, Ekpoma, Edo State, Nigeria
| | | | - Benjamin Efi Aweh
- Department of Mental Health, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
| | - Paul Erohubie
- Department of Mental Health, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
| | - Williams Eriyo
- Department of Mental Health, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Department of Mental Health, Ambrose Alli University College of Medicine, Ekpoma, Edo State, Nigeria
| | - Chinwe Inogbo
- Department of Clinical Services, Federal Neuro Psychiatric Hospital, Benin City, Edo State, Nigeria
| | - Peter Akhideno
- Institute of Lassa Fever Research and Control, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Department of Internal Medicine, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
| | - Gloria Eifediyi
- Institute of Lassa Fever Research and Control, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
| | - Reuben Eifediyi
- Department of Obstetrics and Gynaecology, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Department of Obstetrics and Gynaecology, Ambrose Alli University College of Medicine, Ekpoma, Edo State, Nigeria
| | - Danny A Asogun
- Institute of Lassa Fever Research and Control, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Community Health, Ambrose Alli University College of Medicine, Ekpoma, Edo State, Nigeria
| | - Sylvanus Okogbenin
- Department of Obstetrics and Gynaecology, Irrua Specialist Teaching Hospital, Irrua, Edo State, Nigeria
- Department of Obstetrics and Gynaecology, Ambrose Alli University College of Medicine, Ekpoma, Edo State, Nigeria
| |
Collapse
|
3
|
Cereda A, Toselli M, Palmisano A, Vignale D, Leone R, Nicoletti V, Gnasso C, Mangieri A, Khokhar A, Campo G, Scoccia A, Bertini M, Loffi M, Sergio P, Andreini D, Pontone G, Iannopollo G, Nannini T, Ippolito D, Bellani G, Patelli G, Besana F, Vignali L, Sverzellati N, Iannaccone M, Vaudano PG, Sangiorgi GM, Turchio P, Monello A, Tumminello G, Maggioni AP, Rapezzi C, Colombo A, Giannini F, Esposito A. The hidden interplay between sex and COVID-19 mortality: the role of cardiovascular calcification. GeroScience 2021; 43:2215-2229. [PMID: 34260010 PMCID: PMC8278366 DOI: 10.1007/s11357-021-00409-y] [Citation(s) in RCA: 12] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/12/2021] [Accepted: 06/23/2021] [Indexed: 02/05/2023] Open
Abstract
Recent clinical and demographical studies on COVID-19 patients have demonstrated that men experience worse outcomes than women. However, in most cases, the data were not stratified according to gender, limiting the understanding of the real impact of gender on outcomes. This study aimed to evaluate the disaggregated in-hospital outcomes and explore the possible interactions between gender and cardiovascular calcifications. Data was derived from the sCORE-COVID-19 registry, an Italian multicentre registry that enrolled COVID-19 patients who had undergone a chest computer tomography scan on admission. A total of 1683 hospitalized patients (mean age 67±14 years) were included. Men had a higher prevalence of cardiovascular comorbidities, a higher pneumonia extension, more coronary calcifications (63% vs.50.9%, p<0.001), and a higher coronary calcium score (391±847 vs. 171±479 mm3, p<0.001). Men experienced a significantly higher mortality rate (24.4% vs. 17%, p=0.001), but the death event tended to occur earlier in women (15±7 vs. 8±7 days, p= 0.07). Non-survivors had a higher coronary, thoracic aorta, and aortic valve calcium score. Female sex, a known independent predictor of a favorable outcome in SARS-CoV2 infection, was not protective in women with a coronary calcification volume greater than 100 mm3. There were significant differences in cardiovascular comorbidities and vascular calcifications between men and women with SARS-CoV2 pneumonia. The differences in outcomes can be at least partially explained by the different cardiovascular profiles. However, women with poor outcomes had the same coronary calcific burden as men. The presumed favorable female sex bias in COVID-19 must therefore be reviewed in the context of comorbidities, especially cardiovascular ones.
Collapse
Affiliation(s)
- Alberto Cereda
- GVM Care & Research Maria Cecilia Hospital, Cotignola, Italy.
- Cardiovascular Department, ASST Santi Paolo e Carlo, Milan, Italy.
| | - Marco Toselli
- GVM Care & Research Maria Cecilia Hospital, Cotignola, Italy
| | - Anna Palmisano
- IRCCS San Raffaele Scientific Institute, Milan, Italy
- Vita-Salute San Raffaele University, Milan, Italy
| | - Davide Vignale
- IRCCS San Raffaele Scientific Institute, Milan, Italy
- Vita-Salute San Raffaele University, Milan, Italy
| | - Riccardo Leone
- IRCCS San Raffaele Scientific Institute, Milan, Italy
- Vita-Salute San Raffaele University, Milan, Italy
| | - Valeria Nicoletti
- IRCCS San Raffaele Scientific Institute, Milan, Italy
- Vita-Salute San Raffaele University, Milan, Italy
| | - Chiara Gnasso
- IRCCS San Raffaele Scientific Institute, Milan, Italy
- Vita-Salute San Raffaele University, Milan, Italy
| | | | - Arif Khokhar
- GVM Care & Research Maria Cecilia Hospital, Cotignola, Italy
| | - Gianluca Campo
- Azienda Ospedaliero-Universitaria di Ferrara, Cona, FE, Italy
| | | | - Matteo Bertini
- Azienda Ospedaliero-Universitaria di Ferrara, Cona, FE, Italy
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Claudio Rapezzi
- GVM Care & Research Maria Cecilia Hospital, Cotignola, Italy
- Azienda Ospedaliero-Universitaria di Ferrara, Cona, FE, Italy
| | - Antonio Colombo
- GVM Care & Research Maria Cecilia Hospital, Cotignola, Italy
| | | | - Antonio Esposito
- IRCCS San Raffaele Scientific Institute, Milan, Italy
- Vita-Salute San Raffaele University, Milan, Italy
| |
Collapse
|
4
|
Nehal KR, Steendam LM, Campos Ponce M, van der Hoeven M, Smit GSA. Worldwide Vaccination Willingness for COVID-19: A Systematic Review and Meta-Analysis. Vaccines (Basel) 2021; 9:1071. [PMID: 34696179 PMCID: PMC8540052 DOI: 10.3390/vaccines9101071] [Citation(s) in RCA: 92] [Impact Index Per Article: 30.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/07/2021] [Revised: 09/20/2021] [Accepted: 09/21/2021] [Indexed: 12/19/2022] Open
Abstract
Countries across the globe are currently experiencing a third or fourth wave of SARS-CoV-2 infections; therefore, the need for effective vaccination campaigns is higher than ever. However, effectiveness of these campaigns in disease reduction is highly dependent on vaccination uptake and coverage in susceptible populations. Therefore, this systematic review and meta-analysis estimated the vaccination intention and identified determinants of willingness and hesitancy. This study updates the existing body of literature on vaccination willingness, and was conducted according to the PRISMA guidelines. PubMed was searched for publications, selecting only studies published between 20 October 2020 and 1 March 2021, in English, with participants aged >16 years of age. The search identified 411 articles, of which 63 surveys were included that accounted for more than 30 countries worldwide. The global COVID-19 vaccination willingness was estimated at 66.01% [95% CI: 60.76-70.89% I2 = 99.4% [99.3%; 99.4%]; τ2 = 0.83]. The vaccination willingness varied within as well as between countries. Age, gender, education, attitudes and perceptions about vaccines were most frequently observed to be significantly associated with vaccine acceptance or refusal.
Collapse
Affiliation(s)
| | | | - Maiza Campos Ponce
- Department of Health Sciences, Infectious Disease and Public Health, Vrije Universiteit, 1081 HV Amsterdam, The Netherlands; (K.R.N.); (L.M.S.); (M.v.d.H.); (G.S.A.S.)
| | | | | |
Collapse
|
5
|
Olusola-Makinde OO, Makinde OS. COVID-19 incidence and mortality in Nigeria: gender based analysis. PeerJ 2021; 9:e10613. [PMID: 33614262 PMCID: PMC7883696 DOI: 10.7717/peerj.10613] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/22/2020] [Accepted: 11/30/2020] [Indexed: 11/20/2022] Open
Abstract
BACKGROUND Coronavirus Disease 2019 (COVID-19) has been surging globally. Risk strata in medical attention are of dynamic significance for apposite assessment and supply distribution. Presently, no known cultured contrivance is available to fill this gap of this pandemic. The aim of this study is to develop a predictive model based on vector autoregressive moving average (VARMA) model of various orders for gender based daily COVID-19 incidence in Nigeria. This study also aims to proffer empirical evidence that compares incidence between male and female for COVID-19 risk factors. METHODS Wilcoxon signed-rank test is employed to investigate the significance of the difference in the gender distributions of the daily incidence. A VARMA model of various orders is formulated for the gender based daily COVID-19 incidence in Nigeria. The optimal VARMA model is identified using Bayesian information criterion. Also, a predictive model based on univariate autoregressive moving average model is formulated for the daily death cases in Nigeria. Fold change is estimated based on crude case-fatality risk to investigate whether there is massive underreporting and under-testing of COVID-19 cases in Nigeria. RESULTS Daily incidence is higher in males on most days from 11 April 2020 to 12 September 2020. Result of Wilcoxon signed-rank test shows that incidence among male is significantly higher than female (p-value < 2.22 × 10-16). White neural network test shows that daily female incidence is not linear in mean (p-value = 0.00058746) while daily male incidence is linear in mean (p-value = 0.4257). McLeod-Li test shows that there is autoregressive conditional heteroscedasticity in the female incidence (Maximum p-value = 1.4277 × 10-5) and male incidence (Maximum p-value = 9.0816 × 10-14) at 5% level of significance. Ljung-Box test (Tsay, 2014) shows that the daily incidence cases are not random (p-value=0.0000). The optimal VARMA model for male and female daily incidence is VARMA (0,1). The optimal model for the Nigeria's daily COVID-19 death cases is identified to be ARIMA (0,1,1). There is no evidence of massive underreporting and under-testing of COVID-19 cases in Nigeria. CONCLUSIONS Comparison of the observed incidence with fitted data by gender shows that the optimal VARMA and ARIMA models fit the data well. Findings highlight the significant roles of gender on daily COVID-19 incidence in Nigeria.
Collapse
Affiliation(s)
| | - Olusola S. Makinde
- Department of Statistics, Federal University of Technology, Akure, Akure, Ondo, Nigeria
| |
Collapse
|