-
The Lancet. Healthy Longevity Jul 2024Together with environmental factors, intrinsic capacity (the composite of all the physical and mental capacities of an individual) has been proposed as a marker of... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Together with environmental factors, intrinsic capacity (the composite of all the physical and mental capacities of an individual) has been proposed as a marker of healthy ageing. However, whether intrinsic capacity predicts major clinical outcomes is unclear. We aimed to explore the association of intrinsic capacity with functional decline and mortality in older adults.
METHODS
In this systematic review and meta-analysis, we conducted a systematic search in MEDLINE (via PubMed), Scopus, and Web of Science from database inception to Feb 14, 2024, of observational longitudinal studies conducted in older adults (age ≥60 years) assessing the association of intrinsic capacity with impairment in basic activities of daily living (BADL) or instrumental activities of daily living (IADL) or risk of mortality. Estimates were extracted by two reviewers (JLS-S and W-HL) and were pooled using three-level meta-analytic models. The quality of each study was independently assessed by two authors (JLS-S and PLV) using the Newcastle-Ottawa Scale for longitudinal studies. Heterogeneity was evaluated using the I indicator at two levels: within-study (level 2) and between-study (level 3) variation. For associations between intrinsic capacity and IADL and BADL, we transformed data (standardised β coefficients and odds ratios [ORs]) into Pearson product moment correlation coefficients (r) using Pearson and Digby formulas to allow comparability across studies. For associations between intrinsic capacity and risk of mortality, hazard ratios (HRs) with 95% CIs were extracted from survival analyses. This study is registered with PROSPERO, CRD42023460482.
FINDINGS
We included 37 studies (206 693 participants; average age range 65·3-85·9 years) in the systematic review, of which 31 were included in the meta-analysis on the association between intrinsic capacity and outcomes; three studies (2935 participants) were included in the meta-analysis on the association between intrinsic capacity trajectories and longitudinal changes in BADL or IADL. Intrinsic capacity was inversely associated with longitudinal impairments in BADL (Pearson's r -0·12 [95% CI -0·19 to -0·04]) and IADL (-0·24 [-0·35 to -0·13]), as well as with mortality risk (hazard ratio 0·57 [95% CI 0·51 to 0·63]). An association was also found between intrinsic capacity trajectories and impairment in IADL (but not in BADL), with maintained or improved intrinsic capacity over time associated with a lower impairment in IADL (odds ratio 0·37 [95% CI 0·19 to 0·71]). There was no evidence of publication bias (Egger's test p>0·05) and there was low between-study heterogeneity (I=18·4%), though within-study (I=63·2%) heterogeneity was substantial.
INTERPRETATION
Intrinsic capacity is inversely associated with functional decline and mortality risk in older adults. These findings could support the use of intrinsic capacity as a marker of healthy ageing, although further research is needed to refine the structure and operationalisation of this construct across settings and populations.
FUNDING
None.
TRANSLATIONS
For the Spanish and French translations of the abstract see Supplementary Materials section.
Topics: Humans; Aged; Longitudinal Studies; Activities of Daily Living; Mortality; Geriatric Assessment; Aged, 80 and over; Female; Male
PubMed: 38945130
DOI: 10.1016/S2666-7568(24)00092-8 -
The Lancet. Healthy Longevity Jul 2024Parkinson's disease is the second most common neurodegenerative disorder, exhibiting an upward trend in prevalence. We aimed to investigate the prevalence of Parkinson's... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Parkinson's disease is the second most common neurodegenerative disorder, exhibiting an upward trend in prevalence. We aimed to investigate the prevalence of Parkinson's disease, temporal trends between 1980 and 2023, and variations in prevalence by location, age, sex, survey period, sociodemographic index (SDI), human development index (HDI), and study characteristics (sample size, diagnostic criteria, and data source).
METHODS
In this systematic review and meta-analysis we searched PubMed, Cochrane, Web of Science, Embase, Scopus, and Global Health for observational studies that reported Parkinson's disease prevalence in the general population from database inception to Nov 1, 2023. We included studies if they were original observational investigations, had participants from the general population or community-based datasets, and provided numerical data on the prevalence of Parkinson's disease either with 95% CIs or with sufficient information to calculate 95% CIs. Studies were excluded if they were conducted in a specific population, had a sample size smaller than 1000, or were review articles, case reports, protocols, meeting abstracts, letters, comments, short communications, posters, and reports. The publication characteristics (first author and publication year), study location (countries, WHO regions, SDI, and HDI), survey period, study design, diagnostic criteria, data source, participant information, and prevalence data were extracted from articles using a standard form. Two authors independently evaluated eligibility, and discrepancies were resolved through discussion with the third author. We used random effect models to pool estimates with 95% CIs. Estimated annual percentage change (EAPC) was calculated to assess the temporal trend in prevalence of Parkinson's disease. The study was registered with PROSPERO, CRD42022364417.
FINDINGS
83 studies from 37 countries were eligible for analysis, with 56 studies providing all-age prevalence, 53 studies reporting age-specific prevalence, and 26 studies providing both all-age and age-specific prevalence. Global pooled prevalence of Parkinson's disease was 1·51 cases per 1000 (95% CI 1·19-1·88), which was higher in males (1·54 cases per 1000 [1·17-1·96]) than in females (1·49 cases per 1000 [1·12-1·92], p=0·030). During different survey periods, the prevalence of Parkinson's disease was 0·90 cases per 1000 (0·48-1·44; 1980-89), 1·38 cases per 1000 (1·17-1·61; 1990-99), 1·18 cases per 1000 (0·77-1·67; 2000-09), and 3·81 cases per 1000 (2·67-5·14; 2010-23). The EAPC of Parkinson's disease prevalence was significantly higher in the period of 2004-23 (EAPC 16·32% [95% CI 6·07-26·58], p=0·0040) than in the period of 1980-2003 (5·30% [0·82-9·79], p=0·022). Statistically significant disparities in prevalence were observed across six WHO regions. Prevalence increased with HDI or SDI. Considerable variations were observed in the pooled prevalence of Parkinson's disease based on different sample sizes or diagnostic criteria. Prevalence also increased with age, reaching 9·34 cases per 1000 (7·26-11·67) among individuals older than 60 years.
INTERPRETATION
The global prevalence of Parkinson's disease has been increasing since the 1980s, with a more pronounced rise in the past two decades. The prevalence of Parkinson's disease is higher in countries with higher HDI or SDI. It is necessary to conduct more high-quality epidemiological studies on Parkinson's disease, especially in low SDI countries.
FUNDING
National Nature Science Foundation of China.
TRANSLATION
For the Chinese translation of the abstract see Supplementary Materials section.
Topics: Parkinson Disease; Humans; Prevalence; Female; Male; Global Health
PubMed: 38945129
DOI: 10.1016/S2666-7568(24)00094-1 -
Journal of Gastrointestinal and Liver... Jun 2024Mammoplasty, a common cosmetic procedure involving breast augmentation and reduction surgeries, has gained global popularity. Recently, attention has shifted towards...
BACKGROUND AND AIMS
Mammoplasty, a common cosmetic procedure involving breast augmentation and reduction surgeries, has gained global popularity. Recently, attention has shifted towards understanding the prevalence and significance of gastrointestinal (GI) symptoms following mammoplasty. This systematic review aims to consolidate existing literature to provide a comprehensive overview of the type and frequency of GI problems associated with various mammoplasty procedures.
METHODS
A systematic search of PubMed and Scopus databases was conducted until January 22, 2024, identifying observational and interventional studies examining GI symptoms post-mammoplasty. Inclusion criteria covered human studies, while exclusion criteria ensured specificity. Two independent investigators performed screening, and data extraction included study characteristics, surgical procedures, anesthesia methods, and interventions.
RESULTS
Nineteen studies, involving 2,487 subjects, were included in the review. Breast reconstruction emerged as the most studied procedure, followed by breast reduction, augmentation, mastectomy, and breast cancer surgery. Predominant GI symptoms included nausea and vomiting, with varying rates across mammoplasty types. Anesthesia modality influenced symptomatology, with general, local, and combined anesthesia associated with GI disturbances. Antiemetics, notably ondansetron and droperidol, showed variable efficacy. Non-pharmacological approaches, such as preoperative hypnosis, were explored for symptom management.
CONCLUSIONS
Our systematic review reveals insights into GI symptoms post-mammoplasty, emphasizing the common occurrence of symptoms such as nausea and vomiting, alongside less frequent manifestations such as constipation, dry mouth, retching, abdominal pain, and tightness. Variations in symptom prevalence were noted across diverse mammoplasty surgeries, anesthesia methods, and the use of antiemetics, underscoring the complex nature of post-mammoplasty GI disturbances.
Topics: Humans; Mammaplasty; Female; Postoperative Nausea and Vomiting; Gastrointestinal Diseases; Adult; Prevalence
PubMed: 38944853
DOI: 10.15403/jgld-5598 -
Epidemics Jun 2024Antibiotic-resistant Enterobacterales (ARE) are a public health threat worldwide. Dissemination of these opportunistic pathogens has been largely studied in hospitals....
BACKGROUND
Antibiotic-resistant Enterobacterales (ARE) are a public health threat worldwide. Dissemination of these opportunistic pathogens has been largely studied in hospitals. Despite high prevalence of asymptomatic colonization in the community in some regions of the world, less is known about ARE acquisition and spread in this setting. As explaining the community ARE dynamics has not been straightforward, mathematical models can be key to explore underlying phenomena and further evaluate the impact of interventions to curb ARE circulation outside of hospitals.
METHODS
We conducted a systematic review of mathematical modeling studies focusing on the transmission of AR-E in the community, excluding models only specific to hospitals. We extracted model features (population, setting), formalism (compartmental, individual-based), biological hypotheses (transmission, infection, antibiotic impact, resistant strain specificities) and main findings. We discussed additional mechanisms to be considered, open scientific questions, and most pressing data needs.
RESULTS
We identified 18 modeling studies focusing on the human transmission of ARE in the community (n=11) or in both community and hospital (n=7). Models aimed at (i) understanding mechanisms driving resistance dynamics; (ii) identifying and quantifying transmission routes; or (iii) evaluating public health interventions to reduce resistance. To overcome the difficulty of reproducing observed ARE dynamics in the community using the classical two-strains competition model, studies proposed to include mechanisms such as within-host strain competition or a strong host population structure. Studies inferring model parameters from longitudinal carriage data were mostly based on models considering the ARE strain only. They showed differences in ARE carriage duration depending on the acquisition mode: returning travelers have a significantly shorter carriage duration than discharged hospitalized patient or healthy individuals. Interestingly, predictions across models regarding the success of public health interventions to reduce ARE rates depended on pathogens, settings, and antibiotic resistance mechanisms. For E. coli, reducing person-to-person transmission in the community had a stronger effect than reducing antibiotic use in the community. For Klebsiella pneumoniae, reducing antibiotic use in hospitals was more efficient than reducing community use.
CONCLUSIONS
This study raises the limited number of modeling studies specifically addressing the transmission of ARE in the community. It highlights the need for model development and community-based data collection especially in low- and middle-income countries to better understand acquisition routes and their relative contribution to observed ARE levels. Such modeling will be critical to correctly design and evaluate public health interventions to control ARE transmission in the community and further reduce the associated infection burden.
PubMed: 38944024
DOI: 10.1016/j.epidem.2024.100783 -
Journal of Evidence-based Medicine Jun 2024The effect of hormone replacement therapy (HRT) on colorectal cancer (CRC) mortality and all-cause mortality remains unclear. We conducted a systematic review and... (Meta-Analysis)
Meta-Analysis
Impact of hormone replacement therapy on all-cause and cancer-specific mortality in colorectal cancer: A systematic review and dose‒response meta-analysis of observational studies.
OBJECTIVE
The effect of hormone replacement therapy (HRT) on colorectal cancer (CRC) mortality and all-cause mortality remains unclear. We conducted a systematic review and dose-response meta-analysis to determine the effects of HRT on CRC mortality and all-cause mortality.
METHODS
We searched the electronic databases of PubMed, Embase, and The Cochrane Library for all relevant studies published until January 2024 to investigate the effects of HRT exposure on survival rates for patients with CRC. Two reviewers independently extracted individual study data and evaluated the risk of bias between the studies using the Newcastle‒Ottawa Scale. We performed a two-stage random-effects dose-response meta-analysis to examine a possible nonlinear relationship between the year of HRT use and CRC mortality.
RESULTS
Ten cohort studies with 480,628 individuals were included. HRT was inversely associated with the risk of CRC mortality (hazard ratios (HR) = 0.77, 95% CI (0.68, 0.87), I = 69.5%, p < 0.05). The pooled results of seven cohort studies revealed a significant association between HRT and the risk of all-cause mortality (HR = 0.71, 95% CI (0.54, 0.92), I = 89.6%, p < 0.05). A linear dose-response analysis (p for nonlinearity = 0.34) showed a 3% decrease in the risk of CRC for each additional year of HRT use; this decrease was significant (HR = 0.97, 95% CI (0.94, 0.99), p < 0.05). An additional linear (p for nonlinearity = 0.88) dose-response analysis showed a nonsignificant decrease in the risk of all-cause mortality for each additional year of HRT use.
CONCLUSIONS
This study suggests that the use of HRT is inversely associated with all-cause and colorectal cancer mortality, thus causing a significant decrease in mortality rates over time. More studies are warranted to confirm this association.
Topics: Humans; Colorectal Neoplasms; Hormone Replacement Therapy; Observational Studies as Topic; Dose-Response Relationship, Drug; Cause of Death
PubMed: 38943605
DOI: 10.1111/jebm.12622 -
International Breastfeeding Journal Jun 2024Despite global public health organizations endorsing breastfeeding or human milk (HM) as the optimal source of nutrition for infants, detailed knowledge of how HM...
BACKGROUND
Despite global public health organizations endorsing breastfeeding or human milk (HM) as the optimal source of nutrition for infants, detailed knowledge of how HM composition influences infant growth is lacking. In this commentary we summarize and interpret the key findings of a large systematic review on HM components and child growth (N = 141 articles included). We highlight the most consistent associations, discuss study quality issues, explore socio-economic and time trends in this body of research, and identify gaps and future research directions.
KEY FINDINGS OF SYSTEMATIC REVIEW
We grouped HM components into three categories: micronutrients (28 articles), macronutrients (57 articles), and bioactives (75 articles). Overall, we struggled to find consistent associations between HM components and infant growth. The majority of studies (85%) were of moderate or low-quality, with inconsistent HM collection and analysis strategies being identified as the most substantial quality concerns. Additional quality issues included failing to account for potential confounding by factors such as breastfeeding exclusivity and maternal body mass index.
CONSIDERATIONS FOR FUTURE HUMAN MILK RESEARCH
Many opportunities exist for the future of HM research. Using untargeted metabolomics will expand our understanding of HM components beyond previously defined and well-understood components. Machine learning will allow researchers to investigate HM as an integrated system, rather than a collection of individual components. Future research on HM composition should incorporate evidence-based HM sampling strategies to encompass circadian variation as well as infant consumption. Additionally, researchers need to focus on developing high quality growth data using consistent growth metrics and definitions. Building multidisciplinary research teams will help to ensure that outcomes are meaningful and clinically relevant.
CONCLUSION
Despite a large body of literature, there is limited quality evidence on the relationship between HM composition and infant growth. Future research should engage in more accurate collection of breastfeeding data, use standardized HM collection strategies and employ assays that are validated for HM. By systematically evaluating the existing literature and identifying gaps in existing research methods and practice, we hope to inspire standardized methods and reporting guidelines to support robust strategies for examining relationships between HM composition and child growth.
Topics: Humans; Milk, Human; Infant; Breast Feeding; Infant, Newborn; Infant Nutritional Physiological Phenomena; Anthropometry; Female; Child Development
PubMed: 38943170
DOI: 10.1186/s13006-024-00652-x -
Critical Care (London, England) Jun 2024Gut colonization with multidrug-resistant organisms (MDRO) frequently precedes infection among patients in the intensive care unit (ICU), although the dynamics of... (Meta-Analysis)
Meta-Analysis
BACKGROUND
Gut colonization with multidrug-resistant organisms (MDRO) frequently precedes infection among patients in the intensive care unit (ICU), although the dynamics of colonization are not completely understood. We performed a systematic review and meta-analysis of ICU studies which described the cumulative incidence and rates of MDRO gut acquisition.
METHODS
We systematically searched PubMed, Embase, and Web of Science for studies published from 2010 to 2023 reporting on gut acquisition of MDRO in the ICU. MDRO were defined as multidrug resistant non-Pseudomonas Gram-negative bacteria (NP-GN), Pseudomonas spp., and vancomycin-resistant Enterococcus (VRE). We included observational studies which obtained perianal or rectal swabs at ICU admission (within 48 h) and at one or more subsequent timepoints. Our primary outcome was the incidence rate of gut acquisition of MDRO, defined as any MDRO newly detected after ICU admission (i.e., not present at baseline) for all patient-time at risk. The study was registered with PROSPERO, CRD42023481569.
RESULTS
Of 482 studies initially identified, 14 studies with 37,305 patients met criteria for inclusion. The pooled incidence of gut acquisition of MDRO during ICU hospitalization was 5% (range: 1-43%) with a pooled incidence rate of 12.2 (95% CI 8.1-18.6) per 1000 patient-days. Median time to acquisition ranged from 4 to 26 days after ICU admission. Results were similar for NP-GN and Pseudomonas spp., with insufficient data to assess VRE. Among six studies which provided sufficient data to perform curve fitting, there was a quasi-linear increase in gut MDRO colonization of 1.41% per day which was stable through 30 days of ICU hospitalization (R = 0.50, p < 0.01).
CONCLUSIONS
Acquisition of gut MDRO was common in the ICU and increases with days spent in ICU through 30 days of follow-up. These data may guide future interventions seeking to prevent gut acquisition of MDRO in the ICU.
Topics: Humans; Intensive Care Units; Drug Resistance, Multiple, Bacterial; Vancomycin-Resistant Enterococci; Incidence
PubMed: 38943133
DOI: 10.1186/s13054-024-04999-9 -
BMC Public Health Jun 2024The implementation of digital disease surveillance systems at national levels in Africa have been challenged by many factors. These include user applicability, utility...
Availability of published evidence on coverage, cost components, and funding support for digitalisation of infectious disease surveillance in Africa, 2003-2022: a systematic review.
BACKGROUND
The implementation of digital disease surveillance systems at national levels in Africa have been challenged by many factors. These include user applicability, utility of IT features but also stable financial support. Funding closely intertwines with implementations in terms of geographical reach, disease focus, and sustainability. However, the practice of evidence sharing on geographical and disease coverage, costs, and funding sources for improving the implementation of these systems on the continent is unclear.
OBJECTIVES
To analyse the key characteristics and availability of evidence for implementing digital infectious disease surveillance systems in Africa namely their disease focus, geographical reach, cost reporting, and external funding support.
METHODS
We conducted a systematic review of peer-reviewed and grey literature for the period 2003 to 2022 (PROSPERO registration number: CRD42022300849). We searched five databases (PubMed, MEDLINE over Ovid, EMBASE, Web of Science, and Google Scholar) and websites of WHO, Africa CDC, and public health institutes of African countries. We mapped the distribution of projects by country; identified reported implementation cost components; categorised the availability of data on cost components; and identified supporting funding institutions outside Africa.
RESULTS
A total of 29 reports from 2,033 search results were eligible for analysis. We identified 27 projects implemented in 13 countries, across 32 sites. Of these, 24 (75%) were pilot projects with a median duration of 16 months, (IQR: 5-40). Of the 27 projects, 5 (19%) were implemented for HIV/AIDs and tuberculosis, 4 (15%) for malaria, 4 (15%) for all notifiable diseases, and 4 (15%) for One Health. We identified 17 cost components across the 29 reports. Of these, 11 (38%) reported quantified costs for start-up capital, 10 (34%) for health personnel compensation, 9 (31%) for training and capacity building, 8 (28%) for software maintenance, and 7(24%) for surveillance data transmission. Of 65 counts of external funding sources, 35 (54%) were governmental agencies, 15 (23%) foundations, and 7 (11%) UN agencies.
CONCLUSIONS
The evidence on costing data for the digitalisation of surveillance and outbreak response in the published literature is sparse in quantity, limited in detail, and without a standardised reporting format. Most initial direct project costs are substantially donor dependent, short lived, and thus unsustainable.
Topics: Humans; Africa; Communicable Diseases; Population Surveillance
PubMed: 38943132
DOI: 10.1186/s12889-024-19205-2 -
BMC Public Health Jun 2024Owing to the introduction of highly active antiretroviral therapy (HAART), the trajectory of mortality and morbidity associated with human immunodeficiency virus (HIV)... (Meta-Analysis)
Meta-Analysis
Highly active antiretroviral therapy is necessary but not sufficient. A systematic review and meta-analysis of mortality incidence rates and predictors among HIV-infected adults receiving treatment in Ethiopia, a surrogate study for resource-poor settings.
BACKGROUND
Owing to the introduction of highly active antiretroviral therapy (HAART), the trajectory of mortality and morbidity associated with human immunodeficiency virus (HIV) infection has significantly decreased in developed countries. However, this remains a formidable public health challenge for people living with HIV in resource-poor settings. This study was undertaken to determine the pooled person-time incidence rate of mortality, analyze the trend, and identify predictors of survival among HIV-infected adults receiving HAART.
METHODS
Quantitative studies were searched in PubMed, Embase, Scopus, Google Scholar, African Journals Online, and Web of Science. The Joana Briggs Institute critical appraisal tool was used to assess the quality of the included articles. The data were analyzed using the random-effects Dersimonian-Laird model.
RESULTS
Data abstracted from 35 articles involving 39,988 subjects were analyzed. The pooled person-time incidence rate of mortality (all-cause) was 4.25 ([95% uncertainty interval (UI), 3.65 to 4.85]) per 100 person-years of observations. Predictors of mortality were patients aged ≥ 45 years (hazard ratio (HR), 1.70 [95% UI,1.10 to 2.63]), being female (HR, 0.82 [95% UI, 0.70 to 0.96]), history of substance use (HR, 3.10 [95% UI, 1.31 to 7.32]), HIV positive status non disclosure (HR, 3.10 [95% UI,1.31 to 7.32]), cluster of differentiation 4 + T cell - count < 200 cells/mm3 (HR, 3.23 [95% UI, [2.29 to 4.75]), anemia (HR, 2.63 [95% UI, 1.32 to 5.22]), World Health Organisation classified HIV clinical stages III and IV (HR, 3.02 [95% UI, 2.29 to 3.99]), undernutrition (HR, 2.24 [95% UI, 1.61 to 3.12]), opportunistic infections (HR, 1.89 [95% UI, 1.23 to 2.91]), tuberculosis coinfection (HR, 3.34 [95% UI, 2.33 to 4.81]),bedridden or ambulatory (HR,3.30 [95% UI, 2.29 to 4.75]), poor treatment adherence (HR, 3.37 [95% UI,1.83 to 6.22]), and antiretroviral drug toxicity (HR, 2.60 [95% UI, 1.82 to 3.71]).
CONCLUSION
Despite the early introduction of HAART in Ethiopia, since 2003, the mortality rate has remained high. Therefore, guideline-directed intervention of identified risk factors should be in place to improve overall prognosis and increase quality-adjusted life years.
Topics: Humans; HIV Infections; Antiretroviral Therapy, Highly Active; Ethiopia; Incidence; Adult; Female; Male
PubMed: 38943123
DOI: 10.1186/s12889-024-19268-1 -
BMC Ophthalmology Jun 2024The purpose of this review was to examine if dipeptidyl peptidase-4 inhibitor (DPP4i) use affects the risk of diabetic retinopathy (DR). (Meta-Analysis)
Meta-Analysis
BACKGROUND
The purpose of this review was to examine if dipeptidyl peptidase-4 inhibitor (DPP4i) use affects the risk of diabetic retinopathy (DR).
METHODS
Cohort studies published up to 20th July 2023 in the databases of PubMed, CENTRAL, Embase, Scopus, and Web of Science were searched. The adjusted effect size was pooled to calculate the odds ratio (OR).
RESULTS
Seven studies were included. Meta-analysis showed that the use of DPP4i was not associated with any significant change in the risk of DR (OR: 0.86 95% CI: 0.70, 1.06 I = 78%). The pooled analysis also found that DPP4i use was not associated with any significant risk of progression of DR (OR: 0.87 95% CI: 0.47, 1.59 I = 86%). The results did not change during sensitivity analysis.
CONCLUSION
Present evidence from a limited number of real-world studies shows that DPP4i may not affect the incidence and progression of DR. There is a need for further studies from different countries using accurate definitions of DR and its progression to validate the current results.
Topics: Humans; Diabetic Retinopathy; Dipeptidyl-Peptidase IV Inhibitors; Incidence; Diabetes Mellitus, Type 2; Risk Factors; Disease Progression
PubMed: 38943083
DOI: 10.1186/s12886-024-03535-1