Automating blood pressure measurements: The division of labour and the transformation of method

Social Studies of Science

Volumn 26, Issue 1, 1996, Pages 71-94

  Hartland, Joanne ^a  

^a NONE

Author keywords

[No Author keywords available]

Indexed keywords

EID: 30244544989     PISSN: 03063127     EISSN: None     Source Type: Journal    
DOI: 10.1177/030631296026001005     Document Type: Article

References (51)

1. See, for example, Edward A. Feigenbaum and Pamela McCorduck, The Fifth Generation: Artificial Intelligence and Japan's Computer Challenge to the World (London & Sydney: Pan Books, 1984);
2. Jerrold S. Maxmen, The Post Physician Era (New York: John Wiley, 1976);
3. and Maxmen, 'Long Term Trends in Health Care: The Post Physician Era Reconsidered', in Detlef Schwefel (ed.), Indicators and Trends in Health Care (Berlin: Springer Verlag, 1987), 109-15.
4. Joseph Weizenbaum, Computer Power and Human Reason: From Judgement to Calculation (San Francisco, ÇA: W.H. Freeman, 1976).
5. Hubert L. Dreyfus, What Computers Can't Do (New York: Harper & Row, 1979; reprinted Cambridge, MA: MIT Press, 1992).
6. H.M. Collins, Artificial Experts: Social Knowledge and Intelligent Machines (Cambridge, MA: MIT Press, 1990).
7. Critikon, A Guide to Blood Pressure Measurement Techniques (Bracknell, Berks.: Critikon, 1992), 4. This publication is available from Critikon pic (a Division of Johnson & Johnson Professional Products Ltd), The Braccans, London Road, Bracknell, Berkshire RG12 2AT, UK (fax +44 1344 864091).
8. The medical 'firm' is the term used to refer to all the staff working for one consultant. The firm may consist of one or more house officers and senior house officers, a registrar, a senior registrar, the consultant and any medical students attached to that group.
9. See H.M. Collins, 'Reproducing the Past: Three Methods for Assembling a Cultural Inventory', paper presented to the International Workshop on Replications of Historical Experiments in Physics (Oldenberg, Germany, 24-26 August 1992).
10. The observer in these 'proxy stranger' experiments should be an expert in the task at hand. In this respect, the techniques differ from the 'stranger' techniques described by Bruno Latour and Steve Woolgar in Laboratory Life: The Social Construction of Scientific Facts (Beverly Hills, CA: Sage Publications, 1979): in the latter account, the observer was both a novice and a stranger, and was largely ignorant of science. Achieving complete understanding, or 'going native', was treated as a danger (ibid., 29-30). As such, the LatourAVooIgar and Collins methods differ, and cannot be directly compared.
11. A problem with this method lies in deciding how 'strange' the stranger should be. An extreme choice of stranger would be one so strange that she did not speak the same language as the natives. This would illustrate all the expertise to do with understanding English that is involved in the task at hand. The stranger need be as strange as the experimenter chooses. In this case, my choice was based on the need for someone unfamiliar with the medical task under scrutiny, but familiar with other aspects (such as English language knowledge) which are needed for the job.
12. These instructions were largely my own creation, based on my own experience. I also consulted several textbooks and spoke to a physician/ diabetologist before completing the first draft.
13. In retrospect, it may seem that I was naïve to have expected novices to the medical world to understand these anatomical terms, and that including them in the instructions was disingenuous. But at the time I formulated the instructions, these terms seemed perfectly appropriate. This may signify how invisible the extent of my medical socialization had become to me. As such it reinforces my claim that a 'step back' needs to be taken by those over-familiar with any task.
14. This point is often overlooked. It is assumed that because a technique can be described using a set of rules, its execution is straightforward.
15. See, for example, H.M. Collins, Changing Order (London: Sage Publications, 1985).
16. James C. Pétrie et al., 'British Hypertension Society Recommendations on Blood Pressure Measurement', British Medical Journal, Vol. 239 (6 September 1986), 611-15.
17. Benjamin F. Miller and Claire B. Keane, The Encyclopedia and Dictionary of Medicine, Nursing and Allied Health (London: W.U. Saundcrs & Co., 1978), ISO32.
18. This doctor was suggesting that staff produce variations in readings, and are prone to 'observer bias'. He reinforced this view in a subsequent quote, when he pointed out that some staff fail to record the 4th and 5th Korotkoff sounds when the difference between them is significant. He felt this dual recording to be essential in such cases, but other staff did not see it as necessary.
19. Critikon, op. cit. note 5, 7.
20. Michael J. Jamieson et al., 'The Measurement of Blood Pressure: Sitting or Supine, Once or Twice?', Journal of Hypertension, Vol. 8 (1990), 6435-40, at 6435.
21. Edna O'Brien and Kevin O'Malley, 'Twenty-Four Hour Ambulatory Blood Pressure Monitoring: A Review of Validation Data', Journal of Hypertension, Vol. 8, Supplement 16 (1990), S11-S16.
22. Pétrie et al., op. cit. note 14.
23. See Gill Halt, Disciplining the Body: The Production of Order in Medical Diagnosis (unpublished MSc dissertation, University of Bath, 1992). A copy of this dissertation is available at the University of Bath Library.
24. See also Vincent Dawe, 'One-Third of Juniors in Dark on BP Tests', Hospital Doctor, Vol. C13, No. 2 (14 January 1993), 2;
25. David Short, 'The Diastolic Dilemma', British Medical Journal (1976), Vol. 2 (18 September 1976), 685-86; Jamieson et al., op. cit. note 18.
26. That these variations occur is borne out by Dawe (op. cit. note 22). His article highlights the discrepancies in the techniques doctors use to measure BP.
27. This falls into the category of action described by Collins as 'behaviourspecific' action: see Collins, op. cit. note 4, 34.
28. Robert H. Preisen and J. Lance Lichtor, 'Indirect Measurement of Blood Pressure in Neonates and Infants Utilizing an Automatic Noninvasive Oscillometric Monitor', Anaesthesia and Analgesia, Vol. 60, No. 10 (October 1981), 74245,
29. quoted in Critikon, Survey of Studies Involving the Dinamap Vital Signs Monitor (Bracknell, Berks.: Critikon, 1986), at 14.
30. Maynard Ramsay, 'Noninvasive Automatic Determination of Mean Arterial Pressure', Medical & Biological Engineering & Computing, Vol. 17, No 1 (1979), 11-18, quoted in Critikon, op. cit. note 25, at 22.
31. Tetsuo Inoue, 'The Use of An Automated Blood Pressure Device: The Dinamap Monitor', The Clinical Report, Vol. 22, No. 14 (1988), 5151-54,
32. quoted in Critikon, Survey of Studies Involving the Dinamap Vital Signs Monitor (Ascot, Berks.: Critikon, 1991), at 8.
33. Richard H. Epstein, Susan Kaplan, Barbara L. Leighton, Mark C. Norris and Cheryl A. DeSimone, 'Evaluation of a Continuous Non-Invasive Blood Pressure Monitor in Obstetric Patients Undergoing Spinal Anaesthesia', Journal of Clinical Monitoring, Vol. 5, No. 3 (July 1989), 157-63, quoted in Critikon, op. cit. note 27, at 6.
34. Anonymous, 'Automatic Sphygmomanometers', Health Devices, Vol. 15, No. 7 (July 1986), 187-208, quoted in Critikon, op. cit. note 27, at 7.
35. Betty L. Grundy et al., 'Non-Invasive vs. Invasive Blood Pressure Measurements', original presentation to the Society of Critical Care Medicine (Washington, DC: SCCM, 1981), quoted in Critikon, op. cit. note 25, at 15.
36. Critikon, Dinamap Monitors: You Know the Name . . . (Ascot, Berks.: Critikon, 1990), no page numbers.
37. Donna Seaman, 'Should You Trust Automatic Blood Pressure Monitors?', Nursing (January 1985), quoted in Critikon, op. cit. note 25, at 9.
38. Gaining native competence and understanding during this case study was easy. I drew on my own experience as a cardiology technician familiar with both methods of BP measuring and with the significance of the readings, and I drew on the experience of experts involved with the new technology. When full participation was not possible, I used conversations with users and periods of observation of the machine in use to see how the machine performed and how well it fitted into the routine of the department.
39. See H.M. Collins, 'Researching Spoonbending: Concepts and Methods of Participatory Research', in Colin Bell and Helen Roberts (eds), Social Researching: Politics, Problems, Practice (London: Routledge & Kegan Paul, 1984), 54-69.
40. Seaman, op. cit. note 32, quoted in Critikon, op, cit. note 25, at 9.
41. F.K. Forster and D. Turney, 'Oscillometric Determination of Diastolie, Mean and Systolic Blood Pressure: A Numerical Model', Journal of Biomédical Engineering, Vol. 108 (1986), 359-64, at 359.
42. The original publication is E.J. Marey, 'Pression et Vitesse du Sang', Physiologie Expérimentale, Vol. 2, Chapter VIII (1879), 307-43.
43. Also see Mary K. Davis, 'Anaesthesia and the Neurological Patient: Part I: Monitoring Needs', American Association of Nurse Anaesthetists (AANA) Journal, Vol. 50, No. 2 (April 1982), 143-48, quoted in Critikon, op. cit. note 25, at 12.
44. Critikon, op. cit. note 5, 10.
45. Ibid., 11.
46. See H.M. Collins and Martin Kusch, 'Two Kinds of Action: A Phenomenological Study', Philosophy and Phenomenological Research, Vol. 55 (1995, forthcoming).
47. Critikon, op. cit. note 5, 5.
48. Ibid.
49. We may be sure that none of the four proxy strangers who tried the manual method would have been able to use the automatic machine without help, because they had all experienced difficulties doing parts of the task that remain human parts when the new technology is introduced.
50. A senior registrar had suggested that attaching the cuff to the arm was one of the very skilled parts of the task. He asked me: 'Well, would you know how to fix it to the arm? Would anybody else know? Would my mother know how to fix it to the arm over the brachial artery?' He would presumably argue that this part could not . be delegated to a machine. My suggestion is that anticipating the likely variations associated with this part of the procedure is more feasible than anticipating those linked to other other parts of the task. An acceptable mechanically produced outcome, using a pre-specified method, is more easily envisaged in this case. The necessity for an experienced human to stand by and intervene when a patient with an arm type that has not been anticipated (or without a left arm) is still necessary. Whether a task is skilled or unskilled is a different issue to whether a job could be accomplished using a pre-specified or rule-based approach.
51. See J. Hartland, The Machinery of Medicine: An Analysis of Algorithmic Approaches to Medical Knowledge and Practice (unpublished PhD thesis, University of Bath, 1993).