Journal of Healthcare Informatics Research最新文献

Diabetes mellitus increases the risk of adverse maternal and fetal outcomes. Preconception care is vital to minimise complications; however, preconception care service provision is hindered by inadequate knowledge, resources and care fragmentation. Mobile health technology, particularly smartphone apps, could improve preconception care and pregnancy outcomes for women with diabetes. The aim of this study is to co-create a preconception and diabetes information app with healthcare professionals and women with diabetes and explore the feasibility, acceptability and preliminary effects of the app. A mixed-methods study design employing questionnaires and semi-structured interviews was used to assess preliminary outcome estimates (preconception care knowledge, attitudes and behaviours), and user acceptability. Data analysis included thematic analysis, descriptive statistics and non-parametric tests. Improvements were recorded in knowledge and attitudes to preconception care and patient activation measure following the 3-month app usage. Participants found the app acceptable (satisfaction rating was 72%), useful and informative. The app's usability and usefulness facilitated usage while manual data input and competing priorities were barriers which participants felt could be overcome via personalisation, automation and use of daily reminders. This is the first study to explore the acceptability and feasibility of a preconception and diabetes information app for women with diabetes. Triangulated data suggest that the app has potential to improve preconception care knowledge, attitudes and behaviours. However, in order for women with DM to realise the full potential of the app intervention, particularly improved maternal and fetal outcomes, further development and evaluation is required.

Epidemic spread in a population is traditionally modeled via compartmentalized models which represent the free evolution of disease in the absence of any intervention policies. In addition, these models assume full observability of disease cases and do not account for under-reporting. We present a mathematical model, namely PolSIRD, which accounts for the under-reporting by introducing an observation mechanism. It also captures the effects of intervention policies on the disease spread parameters by leveraging intervention policy data along with the reported disease cases. Furthermore, we allow our recurrent model to learn the initial hidden state of all compartments end-to-end along with other parameters via gradient-based training. We apply our model to the spread of the recent global outbreak of COVID-19 in the USA, where our model outperforms the methods employed by the CDC in predicting the spread. We also provide counterfactual simulations from our model to analyze the effect of lifting the intervention policies prematurely and our model correctly predicts the second wave of the epidemic.

There have been many efforts in the last decade in the health informatics community to develop systems that can automatically recognize and predict disclosures on social media. However, a majority of such efforts have focused on simple topic prediction or sentiment classification. However, taboo disclosures on social media that people are not comfortable to talk with their friends represent an abstract theme dependent on context and background. Recent research has demonstrated the efficacy of injecting concept into the learning model to improve prediction. We present a vectorization scheme that combines corpus- and lexicon-based approaches for predicting taboo topics from anonymous social media datasets. The proposed vectorization scheme exploits two context-rich lexicons LIWC and Urban Dictionary. Our methodology achieves cross-validation accuracies of up to 78.1% for the supervised learning task on Facebook Confessions dataset, and 70.5% for the transfer learning task on the YikYak dataset. For both the tasks, supervised algorithms trained with features generated by the proposed vectorizer perform better than vanilla t f - i d f representation. This work presents a novel methodology for predicting taboos from anonymous emotional disclosures on confession boards.

Patients can use social media to describe their healthcare experiences. Several social media platforms, such as the Care Opinion platform, host large volumes of patient stories. However, the large number of these stories and the healthcare system's workload make exploring these stories a difficult task for healthcare providers and administrators. This study uses text mining for analyzing patient stories on the Care Opinion platform and exploring healthcare experiences described in these stories. We collected 367,573 stories, which were posted between September 2005 and September 2019. Topic modeling (Latent Dirichlet Allocation) and sentiment analysis were used to analyze the stories. Sixteen topics were identified representing five aspects of the healthcare experience: communication between patients and providers, quality of clinical services, quality of non-clinical services, human aspects of healthcare experiences, and patient satisfaction. There was also a clear sentiment in 99% of the stories. More than 55% of the stories that describe the patient's request for information, the patient's description of treatment, or the patient's making of an appointment had a negative sentiment, which represents patient dissatisfaction. The study provides insights into the content of patient stories and demonstrates how topic modeling and sentiment analysis can be used to analyze large volumes of patient stories and provide insights into these stories. The findings suggest that these stories are not general social media posts; instead, they describe elements of healthcare experiences that can be helpful for quality improvement.

Supplementary information: The online version contains supplementary material available at 10.1007/s41666-021-00097-5.