Wilson LS, et al. Healthc Inform Res. 2015 Oct;21(4):213-222. http://dx.doi.org/10.4258/hir.2015.21.4.213

Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet

Background: Tuberculosis (TB) is a major global public health problem and one of the leading causes of death among infectious diseases. Although TB can be cured with first-line antibiotics treatment of 6 months regimen, non-adherence to the treatment remains the main challenge for TB prevention and control. Interventions to promote adherence need to address multiple underlying factors linked to non-adherence, which requires a synthesis of studies to understand these factors in the local context. Our review accordingly examines these factors for TB treatment in Ethiopia. Methods: Articles were searched from PubMed and ScienceDirect databases, as well as manual searches through Google and Google Scholar search engines. Both quantitative and qualitative studies that showed factors associated with or reasons for non-adherence, default or loss to follow up from TB treatment were included. A total of 276 articles were screened, and 29 articles were ultimately included in the review. Findings: The extracted factors were synthesized thematically into seven dimensions of patient-centred, social, economic, health system, therapy, lifestyle, and geographic access factors. More than 20 distinct factors were identified under these headings. Some of these factors may also apply quite widely in other settings, with greater or lesser influence, but some are particularly applicable to the Ethiopian setting. Conclusion: Helping patients to achieve full adherence to TB medication is a complex problem as it is influenced by interplay between many factors. Healthcare managers, providers, and researchers need to consider and address multiple underlying factors when designing adherence interventions. This work provides a reference set of such factors for Ethiopian interventions.

In this paper we describe an algorithm for clustering multivariate time series with variables taking both categorical and continuous values. Time series of this type are frequent in health care, where they represent the health trajectories of individuals. The problem is challenging because categorical variables make it difficult to define a meaningful distance between trajectories. We propose an approach based on Hidden Markov Models (HMMs), where we first map each trajectory into an HMM, then define a suitable distance between HMMs and finally proceed to cluster the HMMs with a method based on a distance matrix. We test our approach on a simulated, but realistic, data set of 1,255 trajectories of individuals of age 45 and over, on a synthetic validation set with known clustering structure, and on a smaller set of 268 trajectories extracted from the longitudinal Health and Retirement Survey. The proposed method can be implemented quite simply using standard packages in R and Matlab and may be a good candidate for solving the difficult problem of clustering multivariate time series with categorical variables using tools that do not require advanced statistic knowledge, and therefore are accessible to a wide range of researchers.