Social aspects of pharmaceutical innovation: heart disease

This study examines the invention, innovation, introduction and use of a new drug therapy for coronary heart disease and hypertension; beta-blockade. The relationships between drug introductions and changes in medical perceptions of disease are analysed, and the development and effects of our perception of heart disease through drug treatments and diagnostic technology is described. The first section looks at the evolution of hypertension from its origin as a kidney disorder, Bright's disease, to the introduction and use of effective drugs for its treatment. It is shown that this has been greatly influenced by the introduction of new medical technologies. A medical controversy over its nature is shown both to be strongly influenced by the use of new drugs, and to influence their subsequent use. The second section reviews the literature analysing drug innovation, and examines the innovation of the beta-blocking drugs, making extensive use of participant accounts. The way in which the development of receptor theory, the theoretical basis of the innovation,was influenced by the innovation and use of drugs is discussed, then the innovation at ICI, the introduction into clinical use, and the production of similar drugs by other manufacturers are described. A study of the effects of these drugs is then undertaken, concentrating on therapeutic costs and benefits, and changes in medical perceptions of disease. The third section analyses the effects of other drugs on heart disease, looking at changes in mortality statistics and in medical opinions. The study concludes that linking work on drug innovation with that on drug effects is fruitful, that new drugs and diagnostic technology have greatly influenced medical perceptions of the nature and extent of heart disease, and that in hypertension, the improvement in drug treatment will soon result in much of the population being defined as in need of it life-long.

Adverse drug reaction classification with deep neural networks

We study the problem of detecting sentences describing adverse drug reactions (ADRs) and frame the problem as binary classification. We investigate different neural network (NN) architectures for ADR classification. In particular, we propose two new neural network models, Convolutional Recurrent Neural Network (CRNN) by concatenating convolutional neural networks with recurrent neural networks, and Convolutional Neural Network with Attention (CNNA) by adding attention weights into convolutional neural networks. We evaluate various NN architectures on a Twitter dataset containing informal language and an Adverse Drug Effects (ADE) dataset constructed by sampling from MEDLINE case reports. Experimental results show that all the NN architectures outperform the traditional maximum entropy classifiers trained from n-grams with different weighting strategies considerably on both datasets. On the Twitter dataset, all the NN architectures perform similarly. But on the ADE dataset, CNN performs better than other more complex CNN variants. Nevertheless, CNNA allows the visualisation of attention weights of words when making classification decisions and hence is more appropriate for the extraction of word subsequences describing ADRs.