BackgroundNationally mandated information and communication technology (ICT) systems are often locally resented and little used. This problem is sometimes framed in behaviourist terms, depicting the intended user of technology as a rational actor whose resistance stems from Luddism and/or ignorance, and viewing solutions in terms of training, incentives and sanctions. The implication is that if we get the ‘rewards’ and ‘punishments’ right, people will use technologies. Previous research in the social sciences, notably sociotechnical systems theory, actor–network theory and normalisation process theory, have considered the human, social and organisational context of technology use (and non-use). However, these have all had limitations in explaining the particular phenomenon of resistance to nationally mandated ICT systems.ObjectiveTo develop a sociologically informed theory of resistance to nationally mandated ICT systems.Theoretical approachWe drew on Anthony Giddens’ notion of expert systems (comprising bureaucratic rules and classification systems delivered through technology) as well as theories of professional roles and ethical practice. A defining characteristic of expert systems is that they can produce ‘action at a distance’, allowing managerial control to be exerted over local practice. To the extent that people use them as intended, these systems invariably ‘empty out’ social situations by imposing rules and categories that are insensitive to local contingencies or the unfolding detail of social situations.Study design and settingSecondary analysis of data from case studies of three nationally mandated ICT systems in the English NHS, collected over the period 2007–10.ResultsOur analysis focused mainly on the Choose and Book system for outpatient referrals, introduced in 2004, which remained unpopular and little used throughout the period of our research (i.e. 2007–13). We identified four foci of resistance: to the policy of choice that Choose and Book symbolised and purported to deliver; to accommodating the technology’s sociomaterial constraints; to interference with doctors’ contextual judgements; and to adjusting to the altered social relations consequent on its use. More generally, use of the mandated system tended to constrain practice towards a focus on (the efficiency of) means rather than (the moral value of) ends. A similar pattern of complex sociological reasons for resistance was also seen in the other two technologies studied (electronic templates for chronic disease management and the Summary Care Record), though important differences surfaced and were explained in terms of the policy inscribed in the technology and its material features.Conclusion‘Resistance’ is a complex phenomenon with sociomaterial and normative components; it is unlikely to be overcome using atheoretical behaviourist techniques. To guide the study of resistance to ICT systems in health care, we offer a new theoretical and empirical approach, based around a set of questions about the policy that the technology is intended to support; the technology’s material properties; the balance between (bureaucratic) means and (professional) ends; and the implications for social roles, relationship and interactions.We suggest avenues for future research, including methodology (e.g. extending the scope and scale of ethnographic research in ICT infrastracture), theory development (e.g. relating to the complexities of multi-professional team working) and empirical (e.g. how our findings might inform the design and implementation of technologies that are less likely to be resisted).FundingThe National Institute for Health Research Health Services and Delivery Research programme.