Version classiqueVersion mobile
OpenEdition Books

Proceedings of the fourth Resilience Engineering Symposium

 | 
Erik Hollnagel
, 
Éric Rigaud
, 
Denis Besnard

From Hazards To Resilience In Socio-Technical Healthcare Systems

Siri Andersen, P. Fairbrother, M. Felici, J. Hanley, B. McKinstry et J. Ure

Résumé

Our earlier work on socio-technical hazards identified some key new hazard classes: Boundary Hazards, Evolutionary Hazards, and Performativity Hazards. These classes help ground an account of emergent risk in socio-technical systems. From the insights underpinning these new hazard classes we consider how these contribute to flexibility or brittleness in organizational responses to events. We exemplify these ideas in the healthcare domain.

Note de l’auteur

Note portant sur l’auteur1

Note portant sur l’auteur2

Note portant sur l’auteur3, 4

Note portant sur l’auteur5

Note portant sur l’auteur6

Note portant sur l’auteur7

Texte intégral

1 Introduction

1The Social Sciences have made a significant contribution to the analysis of hazards in socio-technical systems that, so far, have not been adopted widely in the deployment of such systems. Normal Accident Theory (NAT) analyzes application domains and their technologies, to provide a characterization of their complexity-coupling risk space (Perrow, 1999). Organizations, hence technologies, that exhibit high complexity high coupling are usually characterized by complex interactions, that is, “those of unfamiliar sequences, or unplanned and unexpected sequences, and either not visible or not immediately comprehensible” (Perrow, 1999). This stresses the limitations faced by the ‘calculability of risk’ (MacKenzie, 2001). That is, simply, it is possible to calculate, hence, quantify, limited risk aspects by narrowing the analysis to technical argumentations of risk. From an organizational viewpoint, High Reliability Theory (HRT) Theory characterises those organizations that are able to achieve high performance by the successful deployment of new technologies.

2From a social perspective, Cultural Theory (CT) of risk highlights how different organizational mechanisms and groups select and perceive technological risks (Douglas, Wildavsky, 1983). Furthermore, the Risk Society Theory stresses how technological risks often cross organizational as well as social boundaries (Beck, 1992). These broad perspectives on the risk of technology drove our attempt at synthesis in our analysis of emergent technological risk (Anderson, Felici, 2009)1.

3Drawing on research collaborations in Dependability, we have identified classes of socio-technical hazards that are usually overlooked, or misinterpreted, by a narrow, technology based assessment, rather than on a wider perspective (e.g. involving social, cultural, organizational aspects) (Anderson, Felici, 2009). We have identified three main classes of socio-technical hazards: Boundary Hazards, Evolutionary Hazards and Performativity Hazards. Boundary Hazards characterize technology that ‘supports’ different communities of practice. Technological integration strategies often undermine differences between communities of practices giving rise to tensions resulting into ‘failures’. Evolutionary Hazards characterize a lack of understanding of the evolutionary nature of technology. Technology innovation involves an extent of evolving work practice. Assessing technology and its impact involves dealing with knowledge uncertainty. Unfortunately, engineering methodologies often struggle to cope with uncertainty. Performativity Hazards, finally, characterize the interplay between technology and social behaviour. This stresses the differences between local and global aspects of technology. Technology supports local as well as global social mechanisms. Unfortunately, this triggers either positive or negative feedback due to social behaviour. These socio-technical classes of hazards, we believe, embrace the lessons learned in looking at technology failures by multi-disciplinary lenses.

4Taking these classes of socio-technical hazards as a starting point we discuss how we can provide an analysis of potential contributions to technology to brittleness or flexibility in Healthcare organizations. For instance, dependability issues in telemedicine systems give rise to mistrust in technological artefacts. Critically, they affect Healthcare practices (in particular, how different groups position themselves with respect to technologies). The deployment of such systems in large scale often faces several governance issues. Other issues concern how system developments fit organizational practices. System developments often advocate subsequent incremental deployments. However, it is often controversial how technology trajectories support evolving work practices. Finally, at an organizational level, it is interesting to look how organizations build trust by assuring scrutiny of uncertain data. This characterizes those systems that enable organizational learning. In each of these situations we provide a brief analysis of how technical deployments are shaped by surrounding organisations and cultures to contribute to flexibility or brittleness in response to new events. This paper is structure as follows. Section 2 gives an introduction to Healthcare problems that may benefit from technology support. In particular, it introduces an argument for Telemetry-Enabled Healthcare Developments. Section 3 uses the classes of socio-technical hazards in order to conduct a post-analysis classification of potential hazardous conditions highlighted by relevant clinical trials. This paper analyses the trials’ outcomes from multi-disciplinary account of risk – the one supported by the socio-technical classes of hazards – rather than from a clinical perspective. The point is to use the socio-technical classes of hazards in order to identify and analyse potential hazards due to the introduction of new technology in a critical domain such as the Healthcare. Section 4 at the end draws some concluding remarks.

2 Telemetry-Enabled Healthcare Developments

  • 2 WHO, Fact sheet N°317, February 2007. “What are cardiovascular diseases? Cardiovascular diseases (C (...)
  • 3 WHO, Fact sheet N°317, February 2007. “FACTS ABOUT CARDIOVASCULAR DISEASES: CVDs are the number one (...)
  • 4 WHO, Fact sheet N° 317, February 2007. “The causes of CVDs are well established and well known. The (...)

5Cardiovascular Diseases (CVDs)2 are major causes of death worldwide and the biggest cause of death in the EU. Figures reported by WHO (World Health Organization) highlight the importance of addressing CVDs3. While it is known that multiple factors4 (e.g. life style, social and cultural influences, etc.) increase the risk of CVDs, management strategies which include change in lifestyle (e.g. eating a healthy diet, practising physical activity, smoking cessation, reducing alcohol intake, etc.) suffer from poor uptake and compliance. Likewise compliance to drug therapies such as aspirin and statins is also poor (Benner et al., 2002) – the statins are a class of drugs that lower cholesterol levels in people with or at risk of cardiovascular disease. This remains true even of patients who have had serious cardiac events (Jackevicius et al., 2002; Gislason GH et al., 2006). Many of these seriously affected individuals need to monitor their condition. There is evidence that monitoring of patients with severe chronic heart failure is beneficial in terms of mortality and hospital admission, but such monitoring is, at best, based on observations taken at most once a day. There is little evidence of how continuous monitoring of such patients, integrated with medical management, influences disease outcome. Nor is there any evaluation of how the personal and socio-economic impact of such continuous monitoring influences compliance to lifestyle or medication advice. There is a pressing need to develop enabling technologies for continuous monitoring, to understand how such devices will perform, and how integration of new pervasive information sources with the current medical information infrastructure influences medical care, and patient self-management. In addition, any work in this area needs to integrate governance issues as a key design consideration because failure to address these issues can wreck attempts at deployment.

6Heart failure is a major health problem. UK figures from the 1990s showed that heart failure has a higher incidence than breast or lung cancer, with a 1 year survival rate of 62 % (Office for National Statistics, 2001; Cowie et al., 2000). The condition accounts for about 5 % of all medical admissions to hospital, and rates of re-admission are also among the highest for any common condition in the UK. Providing services to patients with heart failure costs the UK NHS (National Health Service) an estimated £625 million each year. Research indicates that care provided by effective multidisciplinary teams can have a positive impact on outcomes for patients and their quality of life. Early detection of deterioration of symptoms allows timely intervention and may prevent hospitalisation (Clark et al., 2007). Deterioration can be detected using relatively simple measures (e.g., history of increasing tiredness, breathlessness, ankle swelling or palpitation) and a measure of the patients’ weight, blood pressure and pulse oximetry. These parameters can currently be self-monitored by the patient with telemetry enabled support from health care professionals, and the care model is being piloted in a number of EU locations. Respiratory rate also rises as the patient’s condition deteriorates, but currently cannot readily be measured by the systems available. Telemetry supported self monitoring for those at high risk of cardiovascular events due to a combination of type 2 diabetes, hypertension and obesity is also being evaluated. The parameters being monitored include blood pressure, blood glucose and weight. Risk can effectively be reduced by increasing physical activity, but it is often difficult for the patient and clinician to estimate the amount of physical activity achieved.

3 A Socio-Technical Risk Analysis

  • 5 “Reconfiguring Care for Chronic Disease with Home Monitoring – A Pilot Study Report on Telemetry-Su (...)
  • 6 WHO, Fact sheet N° 315, May 2008. “Chronic obstructive pulmonary disease (COPD) is a lung ailment t (...)

7Recent pilot work5 in this area in Chronic Obstructive Pulmonary Disease (COPD)6, suggests that while monitoring technologies can be popular with patients, clinicians are undecided on how best to use monitoring data. In addition, continuous monitoring can increase the number of false alarms –resulting in more frequent intervention (in the case of COPD, increased steroid and antibiotic use) and an increase in the number of GP (General Practitioner) contacts although this may reflect better care for patients. Additionally, early work from on telemetrically supervised self-monitored blood pressure suggests that this data is more easily assimilated by doctors. This section uses the identified classes of socio-technical hazards in order to classify information drawn from different trials ongoing telemetry-enabled healthcare developments. The aim is to discuss how the classes of socio-technical hazards provide valuable support as a means to identify and discuss subtle hazards. The analysis then is of classifying, that is, identifying relevant cases drawn from healthcare trials providing examples for the identified hazards.

8Table 1 summarises the classes of socio-technical hazards, i.e. Boundary Hazards, Evolutionary Hazards, and Performativity Hazards. These classes provide us lenses to analyse the findings drawn from the clinical trials. We have reviewed feedback collected during clinical trials of telemetry-enabled healthcare systems. The pilot work summarises findings according to different categories of users: Patient and Care Personnel. This is useful because it provides us feedback from different perspectives. Hence, it relates to the socialities of technologies in a healthcare domain. Some feedback collected by pilot trials relates directly to our multi-disciplinary classes of socio-technical hazards.

9From a patient perspective, it is evident how telemetry-enable systems change the interaction and access to healthcare services. This somehow changed also patient behaviour, shifted responsibilities and modified risk exposure – “The frequency with which patients experienced such symptoms during the pilot study underlined the difficulty patients had in balancing the risk of delaying treatment for an exacerbation, against concerns with bothering staff at the surgery unnecessarily.” Potential concerns and risk exposures due to the different nature of healthcare services (e.g. in terms of types of interactions) suggested that such services would be suitable for patient with moderate symptoms. Care professionals highlighted concerns how telemetry-enable systems potentially increase workload and require the redistribution of responsibilities –“Some GPs and nurses had concerns from the outset that this might add to workload, or change the balance of work from face-to-face consultations to remote monitoring and administration. This was perceived by implementation team interviewees as having important resource implications for the wider rollout.” Other feedback can be reviewed with respect to the classes of socio-technical hazards. This would provide further information how to design, validate and deploy such systems.

Table 1. Classes of Socio-Technical Hazards

10Other information can be analysed with respect to the socio-technical hazards. They provide a basis for a systematic discussion of hazards that otherwise might be overlooked. The socio-technical hazards complement and extend those risk analyses focusing mainly on technologies. Therefore, they are an useful tool that intends to support practitioners to discuss and identify hazards that my jeopardise any effort to deploy complex technologies in critical application domains.

4 Concluding Remarks

11This paper has discussed findings from a pilot study for telemetry-enabled healthcare systems. Some concerns relate to the classes of socio-technical hazards. The classes of socio-technical hazards allow us to discuss relevant hazards and to classify them. Thye would provide useful support for the design of technologies to be used in complex settings. This provides also some indication how to achieve resilience in such critical application domains.

Bibliographie

References

Anderson, S., Felici, M. (2009). Classes of socio-technical hazards: Microscopic and macroscopic scales of risk analysis. Risk Management, Palgrave Macmillan, Volume 11, Issue 3-4, pp. 208-240.

Beck, U. (1992). Risk Society: Towards a New Modernity, SAGE Publications.

Benner, J.S., et al. (2002). Long-term persistence in use of statin therapy in elderly patients. JAMA (The Journal of American Medical Association) 288(4):455-461. http://dx.doi.org/doi:10.1001/jama.288.4.455

Clark RA, Inglis SC, McAlister FA, Cleland JGF, Stewart S. (2007). Telemonitoring or structured telephone support programmes for patients with chronic heart failure: systematic review and meta-analysis. BMJ, 334: 942.

Cowie M R, Wood D A, Coats A J S, Thompson S G, Suresh V, Poole-Wilson P A and Sutton G C (2000). Survival of patients with a new diagnosis of heart failure: a population based study. Heart; 83: 505-10.

Douglas, M., Wildavsky, A. (1983). Risk and Culture: An Essay on the Selection of Technological and Environmental Dangers, University of California Press, 1983.

Gislason GH et al. (2006). Long-term compliance with beta-blockers, angiotensin-converting enzyme inhibitors, and statins after acute myocardial infarction. Eur Heart J. 27(10):1153-8.

Jackevicius, C.A., et al. (2002) Adherence with statin therapy in elderly patients with and without acute coronary syndromes. JAMA (The Journal of American Medical Association) 288(4):462-467. http://dx.doi.org/10.1001/jama.288.4.462

MacKenzie, D. (2001). Mechanizing Proof: Computing, Risk, and Trust, MIT Press. Office for National Statistics (2001). Quinn M, Babb P, Brock A, Kirby L and Jones J. Cancer trends in England and Wales 1950-1999. The Stationary Office: London;

Perrow, C. (1999). Normal Accidents: Living with High-Risk Technologies, Princeton University Press.

Notes

1 School of Informatics, The University of Edinburgh 10 Crichton Street, Edinburgh EH8 9AB, UK

2 Centre for Integrated Healthcare Research, Edinburgh Napier University Sighthill Court, Edinburgh EH11 4BN, UK

3 School of Informatics, The University of Edinburgh 10 Crichton Street, Edinburgh EH8 9AB, UK

4 Medical School, The University of Edinburgh Teviot Place, Edinburgh EH8 9AG, UK

5 Centre for Integrated Healthcare Research, Edinburgh Napier University Sighthill Court, Edinburgh EH11 4BN, UK

6 Medical School, The University of Edinburgh Teviot Place, Edinburgh EH8 9AG, UK

7 Centre for Integrated Healthcare Research, Edinburgh Napier University Sighthill Court, Edinburgh EH11 4BN, UK

Notes de fin

1 Corresponding author: mfelici@staffmail.ed.ac.uk

2 WHO, Fact sheet N°317, February 2007. “What are cardiovascular diseases? Cardiovascular diseases (CVDs) are a group of disorders of the heart and blood vessels…”

3 WHO, Fact sheet N°317, February 2007. “FACTS ABOUT CARDIOVASCULAR DISEASES: CVDs are the number one cause of death globally: more people die annually from CVDs than from any other cause; An estimated 17.5 million people died from CVDs in 2005, representing 30 % of all global deaths. Of these deaths, an estimated 7.6 million were due to coronary heart disease and 5.7 million were due to stroke. Over 80 % of CVD deaths take place in low-and middle-income countries and occur almost equally in men and women; By 2015, almost 20 million people will die from CVDs, mainly from heart disease and stroke. These are projected to remain the single leading causes of death.”

4 WHO, Fact sheet N° 317, February 2007. “The causes of CVDs are well established and well known. The most important causes of heart disease and stroke are unhealthy diet, physical inactivity and tobacco use. These are called 'modifiable risk factors'.”

5 “Reconfiguring Care for Chronic Disease with Home Monitoring – A Pilot Study Report on Telemetry-Supported Care for COPD in Lothian Region”.

6 WHO, Fact sheet N° 315, May 2008. “Chronic obstructive pulmonary disease (COPD) is a lung ailment that is characterized by a persistent blockage of airflow from the lungs. It is an under-diagnosed, life-threatening lung disease that interferes with normal breathing and is not fully reversible. The more familiar terms of chronic bronchitis and emphysema are no longer used; they are now included within the COPD diagnosis.”

© Presses des Mines, 2011

Conditions d’utilisation : http://www.openedition.org/6540