Hospital Acquired Infection:
A Framework for a National System of Surveillance for the NHS in Scotland

III. SURVEY OF HOSPITAL ACQUIRED INFECTION SURVEILLANCE IN SCOTLAND

Introduction and Methodology

42. A series of questionnaires was prepared and distributed to the Service for completion by a range of healthcare professionals to provide baseline information on HAI surveillance in Scotland. These comprised a Management Questionnaire (for completion by Trust Chief Executives), an Activities Questionnaire (for completion by Consultant Medical Microbiologists/Infection Control Nurses, etc), a Health Board Questionnaire, (for completion by Board General Managers) and a GP Questionnaire, (for completion by General Practitioners). Copies of these questionnaires are reproduced in Appendices III-VI.

Summary of Responses

The Management Questionnaire

45. The Management Questionnaire was addressed to Trust Chief Executives and sought information on arrangements for the control and monitoring of HAI at the highest managerial level in the Trust. With the exception of the Scottish Ambulance Service, the questionnaire was sent to all NHS Trusts in Scotland as listed in the NHS Directory. The questionnaire was also sent to the State Hospital and to the three Island Health Units. One Community Trust did not respond and a further two Trusts did not consider the questionnaire to be applicable and did not complete it (one had no in-patient beds and the other was a Community Trust). A total of 47 respondents were therefore included in the final analysis.

46. Few Chief Executives completed the questionnaires themselves and only one third were completed by Directors of Medical or Nursing Services.

47. From the responses received, it would appear that most Trusts have structures in place for planning, undertaking, monitoring and reporting HAI control and surveillance. However, even though the majority of Trusts have established Infection Control Committees and Teams, only two-thirds of Chief Executives receive routine reports from them each year..

48. Approximately 70% of Trusts indicated that HAI was "lncluded in the quality assurance programme" but detailed information was not sought on what this meant in practice. Only half of Trusts reported that HAI, in particular, wound infections and "alert organism" monitoring, was included in clinical specifications. Very few Trusts had a specific budget for the control of HAI and even fewer had one for HAI surveillance. The survey strongly suggests that in general HAI control and surveillance are not high priorities for Trust senior management and that there is a lack of clarity and consensus over what activities should be undertaken in this area.

The Activities Questionnaire

49. The Activities Questionnaire was sent to named Infection Control Doctors (ICDs) and sought information on HAI control and surveillance activities within Trust hospitals. Copies of the Activities Questionnaire were sent to all NHS Trusts in Scotland as listed in the NHS Directory, with the exception of the Scottish Ambulance Service, giving a total of 50 Trusts. Two Trusts who completed the Management Questionnaire did not complete the Activities Questionnaire; one had no in-patient beds and one ICD did not reply. A total of 45 Trusts were therefore included in the final analysis.

50. The majority of the questionnaires were completed by an Infection Control Doctor and/or Nurse.

51. With regard to the distribution of beds, Teaching Hospital Trusts had on average 950 acute beds of which one fifth were in "high risk" areas/supra-area specialties. Acute Hospital Trusts had on average 550 acute beds of which 8% were ""high risk" and Acute Community Trusts had on average 200 acute beds of which 7% were ".high risk.. Only one of the Community Trusts had acute beds, with none in the ""high risk"' category

52. Eighty-nine per cent of Trusts had an Infection Control Doctor and 91 % of Trusts had at least one Infection Control Nurse (ICN) at grade "I" or "H" Ten Trusts had more than one Infection Control Nurse. Link Nurses were employed in up to 40% of Trusts, more frequently in Teaching Hospital Trusts. Eighty-three per cent of Infection Control Doctors and some 71 % of ICNs had clerical support All Infection Control Doctors and the majority of the ICNs in all types of Trust had access to a PC, but only 60% made use of these for surveillance purposes.

53. All but one (Acute/Community) Trust undertook "alert organism" continuous and/or intermittent surveillance. The most common organisms surveyed were MRSA, C. difficile, E. coli 0157, penicillin resistant pneumococci, Legionella sp., vancomycin resistant enterococci, rotaviruses, respiratory syncytial virus, influenza virus and Gp.A Streptococci.

54. Current/continuous surveillance of main types of HAI comprised (in order of frequency) all bloodstream infections, surgical wound infection, IV catheter-related infection and all urinary tract infection. Lower respiratory tract infection surveillance was infrequently performed. Six Trusts (13%) had not undertaken any form of surveillance by type of HAI during the past three years. Thirty-nine Trusts (87%)had undertaken some surveillance during the past three years; 25 Trusts (56%) were undertaking continuous surveillance; and 36 Trusts (80%) had undertaken intermittent/previous surveillance.

55. Three levels of surveillance were identified:

Level I Surveillance was "'laboratory-based'.. Positive culture results were identified in the laboratory but there was no "formal liaison" with clinicians/clinical areas. Definitions of infection were generally not used and any reporting was of the ""number of infections". Denominators were not ascertained. This group therefore comprised `1aboratory confirmed infections". Surveillance of positive results due to alert organisms only was not included.

Level II Surveillance was "selective surveillance", aimed at identifying a proportion of the total number of patients with infection. This included follow-up of positive culture results by discussion with clinicians, visits to the wards, or from medical records. Definitions of infection were generally used with reporting by "number of infections".

Level W Surveillance aimed to identify infections developing within a defined sub-group of patients. This included prevalence surveys and continuous incidence surveillance, `targeted surveillance" and "surveillance by objectives`.. Definitions of infection were generally used and infections were reported as "infections/all patients" or (infrequently) by "weighted rate or risk scores."

56. Continuous bloodstream surveillance was predominantly laboratory-based, followed up by the microbiologist, with 90% at Level I or II. Surgical wound infection (SW1) surveillance was predominantly performed at Level III by the ICN, sometimes with the assistance of specialist nurses or day surgery liaison nurses. One third of Trusts who had undertaken SWI surveillance with post discharge surveillance had made use of audit nurses and/or special funding. Continuous IV catheter-related surveillance tended to be performed at Level II with intermittent surveillance at Level III. This was performed by the ICN with occasional use of audit nurses and specialist TPN nurses.

57. Most Trusts perceive surgical wound infection, followed by IV catheter-related infection as being the most important areas for HAI surveillance. They also view "alert organism"', MRSA or other individual organism surveillance as being of equal importance.

58. The "Glenister" and "HIS" definitions of HAI were most frequently used but no consensus emerged on choice of definition for surgical wound and catheter-related infection surveillance. A total of six different definitions for HAI were identified as being in use in the NHSiS.

59. Three quarters of Trusts regularly reported surveillance results, most commonly to the Infection Control Team, or staff on monitored wards. One third reported surveillance results to senior managers or Clinical/Medical Directors. Twenty-seven per cent communicated their results to the Chief Executive and 29% to the Health Board (the majority as part of a contract).

60. Less than half the Trusts, irrespective of the Trust type, had a planned programme of HAI surveillance and less than half prepared an annual report on the subject.

61. The survey results reveal that the quality and quantity of HAI control and surveillance activities currently being undertaken in Scottish Trusts vary enormously. To achieve national comparability of HAI rates, a pre-requisite will be the standardisation of definitions and methodology to be used.

The Health Board Questionnaire

62. Copies of the Health Board questionnaire were sent to every Health Board in Scotland. Thirteen out of 15 responses were received. In the majority of cases, the questionnaires were completed by Consultants in Public Health Medicine. All of the respondents considered HAI to be an indicator of quality of care although none had specified this in contractual terms. Five rated HAI as "very" important, and a further 5 rated it as "moderately" important. Eleven Health Boards had a designated member of staff with specific responsibility for HAI although the scope of the responsibility was not defined in detail. In most cases this responsibility rested with the Consultant in Public Health Medicine (Communicable Disease/Environmental Health). Just over half of the respondents specified a requirement for HAI surveillance through their contracts with Trusts but very few provided any details with regard to the methodology to be adopted or the specialties, sites or organisms to be surveyed.

63. Most Health Boards do not require Trusts to provide them with HAI information and, in those that do, the extent of communication is limited to reporting on compliance with policy, procedures and standards. Very few of the respondents had adopted a standardised approach to HAI surveillance across all Trusts in their area and none had defined or set acceptable levels for HAI. However, all were unanimous in their support for standardisation at some point in the future.

64. The following organisms, specialties and sites were offered by respondents as appropriate specific candidates for HAI surveillance:

65. The survey results reveal that Health Boards, while appreciating the significance of HAI surveillance, are not clear how important or relevant it is in management terms. Responsibility for HAI surveillance within Boards is poorly defined, with the methodology and practical details left in the main to Trusts to determine of their own accord. Information provided to Health Boards is, for the most part, communicated on a voluntary basis.

The GP Questionnaire

66. Copies of the GP Questionnaire were sent to approximately 10% of all general practitioners in Scotland with representation from every Health Board area. The results of the questionnaire revealed that, although general practitioners do not routinely request information on HAI, two-thirds regard its surveillance as an important indicator of quality of care. Most would like to receive more information on HAI but are uncertain at this stage about their exact requirements. If information were to be provided to them, it would have to be clearly presented and of direct relevance to their patients. Some 64% of respondents said that they would be prepared to participate in HAI surveillance and a further 8% said that they would do so if funded.

Conclusion

67. It is clear from these survey results that a sea-change is required throughout the NHSiS with regard to the perceived relative importance of HAI. The issue needs to move higher up the Health Service's quality agenda in the new era of clinical governance. Trusts in particular need to be more aware of the benefits which follow from having in place a planned programme of HAI surveillance. One such benefit is the opportunity cost which will be associated with a reduction in HAI rates (see paragraph 41). The converse to this is the kind of scenario referred to in paragraph 32 where an HAI outbreak can result in punitive financial and other costs to an individual Trust. Implementation of a national system of HAI surveillance, developed from current local systems, will be crucial if such unfortunate incidents are to be avoided in the future.

68. The Group therefore strongly endorses the view expressed in the House of Lords Select Committee report on antimicrobial resistance¹¹ that infection control and surveillance need to move further up the commissioning agenda.