This paper compares two popular R packages for structural equation modelling (SEM) – lavaan and seminr – to help researchers understand not only how they work, but also when each is most appropriate. Although both tools allow users to estimate the same structural models, they are grounded in different methodological traditions and are designed to support different research goals. Using an identical model, we estimated results with both covariance-based SEM ( lavaan ) and variance-based SEM ( seminr ) and compared their outputs, including model specification syntax, evaluation criteria, and reporting conventions. The results show that both approaches lead to substantively similar conclusions regarding the relationships between constructs, while differing in emphasis: lavaan provides richer global model-fit diagnostics, whereas seminr places greater emphasis on prediction-oriented assessment and convenient access to latent variable scores. The contribution of this study lies in its practical, hands-on demonstration rather than in a theoretical or simulation-based comparison. The findings reinforce that there is no universally “better” SEM approach; instead, methodological choice should be guided by the research objective. Researchers focussed on theory testing may benefit more from lavaan , while those prioritising prediction or exploratory analysis may find seminr more suitable. Ultimately, considering both perspectives can support more transparent, robust, and methodologically appropriate SEM applications.
Objective We performed a microsimulation analysis predicting the societal cost of antimicrobial resistance (AMR), which represents the potential cost savings if Ghana eliminates AMR.Design This study combined bacterial resistance epidemiology and cost data from Ghana to perform a microsimulation analysis focusing on sociodemographic groups, predicting the potential societal cost savings should Ghana eliminate AMR. The nationally representative data were collected from 12 reference laboratories across Ghana’s three geographical belts between June 2021 and December 2023. Case definition was enterobacterial third-generation cephalosporin (3GC) resistant infections, methicillin-resistant Staphylococcus aureus (MRSA) and multidrug-resistant Mycobacterium tuberculosis. Using an adapted microsimulation framework, the simulation incorporated four integrated data modules: population demographics, infection epidemiology, healthcare resource use and expenditure and labour market characteristics. This approach allowed for the construction of synthetic individuals from national data sets and the projection of annual outcomes over a 7-year horizon. Costs were calculated from a societal perspective under a status quo scenario, assuming that admission rates, resistant infection probabilities and mortality rates remain the same. This analysis also considers a 2.1% annual population growth rate, a 5% discount rate for future costs and age-specific resistance risk profile. We stratified the outcome of interest by age groups, sex and wealth quintiles to account for distributional effects and reported the costs in purchasing power parity equivalent in international US dollars.Setting Ghana in West Africa.Participants A simulated population of AMR patients of all ages and sex.Main outcome measures Societal cost attributable to AMR in Ghana.Results Assuming probabilities of all-cause hospital admissions of 0.102 for females and 0.093 for males, along with probabilities of AMR infections of 0.239 and 0.193, we predicted nearly 78 000 (95% CI 72 000 to 83 520) annual AMR infections and approximately 6300 (95% CI 3900 to 8638) attributable deaths. MRSA and 3GC-resistant infections made up 20.2% and 79.2% of the predicted annual infections, corresponding to an estimated mean societal cost of about US$435 million. In decreasing order of magnitude, the estimated mean annual cost of productivity loss due to AMR attributable mortality accounted for 40.6% of the mean annual societal cost, followed by the cost to healthcare providers (24.1%), direct medical cost to patients and caregivers (22.4%), productivity loss for surviving patients and caregivers (10.4%) and direct non-medical costs to patients and caregivers (2.6%). Resistant infections in children under 5 and adults over 60 years contributed 48.2% and 26.9% of the estimated annual societal cost, respectively. Except for the number of resistant infections, the estimated mean annual costs between wealth quintile groups were significantly different (p=0.03) due to differences in productivity costs between wealth quintile groups.Conclusion The study shows that the societal cost implications attributable to AMR are enormous, requiring a concerted effort by society to mitigate the development and spread of AMR organisms.
Background: The growth of high-throughput sequencing and multi-omics research has intensified the need for secure, interoperable and transparent data management infrastructures. Blockchain technology has been widely proposed as a potential solution; however, its feasibility, empirical maturity and comparative performance in bioinformatics remain unclear.Objective: This systematic review analyses blockchain applications in bioinformatics, highlighting claimed security and governance benefits, comparing them with traditional data security approaches, discussing implementation challenges and assessing the empirical rigor of existing studies using a structured quality assessment framework.Methods: This overview was conducted using Scopus, ScienceDirect, IEEE Xplore, ACM Digital Library and SpringerLink for publications from 2014 to 2024. Search strings combined blockchain, bioinformatics and security-related terms. Sixty-five studies met the inclusion criteria. Each study was evaluated using five equally weighted quality dimensions: application specificity, clarity of benefits, empirical evaluation, challenge articulation and reproducibility.Results: Most studies focused on blockchain use cases in genomic data sharing, provenance tracking and access control, with a strong emphasis on conceptual benefits such as immutability and auditability. Fewer studies provided empirical evaluations or direct comparisons with traditional security mechanisms. Quality assessment results revealed a predominance of conceptual and prototype-level contributions; over half of the studies lacked empirical benchmarking and reproducibility was frequently limited. Heterogeneity in blockchain architectures and the absence of standardized genomic benchmarking environments hindered cross-study comparison. No study demonstrated deployment within a production-scale genomic pipeline.Conclusion: Blockchain demonstrates conceptual potential for enhancing provenance, decentralized governance and tamper-resistant auditing in bioinformatics data management. However, empirical validation remains limited and significant technical, regulatory and organizational challenges persist. The current evidence base is insufficient to support large-scale adoption. Future research should prioritize benchmarking using realistic genomic workloads, hybrid architectures that integrate off-chain storage, consent-aware governance models and alignment with regulatory frameworks such as GDPR and HIPAA.
This study examines the applicability of traditional corporate failure prediction models to rural banking institutions in developing economies through an in-depth case analysis of a Ghanaian rural bank exhibiting severe financial distress. Using descriptive-comparative and temporal precedence methods and adopting a multi-dimensional analytical framework incorporating Altman's Z-Score derivatives, banking-specific distress indicators, and operational efficiency metrics, we document systematic failures across financial, operational, and governance dimensions that traditional models inadequately capture. Our analysis reveals that rural banks in developing economies exhibit unique distress patterns characterized by simultaneous technical insolvency, operational control breakdowns, and governance failures that compound traditional financial metrics. The study contributes to corporate failure literature by proposing an enhanced predictive framework specifically calibrated for rural banking institutions, emphasizing the critical role of operational efficiency indicators, transparency metrics, and governance quality measures in early distress detection. Our findings have policy and practical implications as regulatory intervention frameworks require substantial modification to address the multi-dimensional nature of rural banking failures, with implications for deposit insurance schemes and financial system stability in emerging markets.
This paper compares two popular R packages for structural equation modelling (SEM) – lavaan and seminr – to help researchers understand not only how they work, but also when each is most appropriate. Although both tools allow users to estimate the same structural models, they are grounded in different methodological traditions and are designed to support different research goals. Using an identical model, we estimated results with both covariance-based SEM ( lavaan ) and variance-based SEM ( seminr ) and compared their outputs, including model specification syntax, evaluation criteria, and reporting conventions. The results show that both approaches lead to substantively similar conclusions regarding the relationships between constructs, while differing in emphasis: lavaan provides richer global model-fit diagnostics, whereas seminr places greater emphasis on prediction-oriented assessment and convenient access to latent variable scores. The contribution of this study lies in its practical, hands-on demonstration rather than in a theoretical or simulation-based comparison. The findings reinforce that there is no universally “better” SEM approach; instead, methodological choice should be guided by the research objective. Researchers focussed on theory testing may benefit more from lavaan , while those prioritising prediction or exploratory analysis may find seminr more suitable. Ultimately, considering both perspectives can support more transparent, robust, and methodologically appropriate SEM applications.
