International Journal of Health Care Quality Assurance Key performance measures to control maintenance-associated HAIs Stanley Njuangang Champika Liyanage Akintola Akintoye
Article information:
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
To cite this document: Stanley Njuangang Champika Liyanage Akintola Akintoye , (2015),"Key performance measures to control maintenance-associated HAIs", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 690 - 708 Permanent link to this document: http://dx.doi.org/10.1108/IJHCQA-12-2014-0117 Downloaded on: 31 January 2016, At: 05:07 (PT) References: this document contains references to 24 other documents. To copy this document: [email protected] The fulltext of this document has been downloaded 170 times since 2015*
Users who downloaded this article also downloaded: Sohaib Aleem, William C Torrey, Mathew S Duncan, Shoshana J Hort, John N Mecchella, (2015),"Depression screening optimization in an academic rural setting", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 709-725 http://dx.doi.org/10.1108/ IJHCQA-01-2015-0012 Pia Hannele Jansson von Vultée, (2015),"Healthy work environment – a challenge?", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 660-666 http://dx.doi.org/10.1108/ IJHCQA-11-2014-0108 Mohammadkarim Bahadori, Mehdi Raadabadi, Ramin Ravangard, Donia Baldacchino, (2015),"Factors affecting dental service quality", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 678-689 http://dx.doi.org/10.1108/IJHCQA-12-2014-0112
Access to this document was granted through an Emerald subscription provided by emeraldsrm:126209 []
For Authors If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services. Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
*Related content and download information correct at time of download.
The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0952-6862.htm
IJHCQA 28,7
690 Received 23 December 2014 Revised 12 March 2015 22 April 2015 Accepted 9 May 2015
Key performance measures to control maintenance-associated HAIs Stanley Njuangang, Champika Liyanage and Akintola Akintoye Grenfell-Baines School of Architecture, Construction and Environment, University of Central Lancashire, Preston, UK Abstract
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Purpose – The purpose of this paper is to improve overall healthcare maintenance (HM) service performance in NHS infection control (IC). Hence, the authors identify critical success factors (CSFs) and key performance measures in maintenance-associated infections. These infections occur because of the poor performance of HM service in IC. Design/methodology/approach – In the first Delphi exercise, complete CSFs and performance measures were presented to the Delphi participants for refinement and modification. Delphi round 1 data were analysed manually and used to refine the rounds 2 and 3 Delphi instruments. In subsequent Delphi rounds, the results were analysed through descriptive statistics. Findings – In total, eight CSFs and 53 key performance measures were identified for reducing maintenance-associated infections in hospitals. For example, establishing clear communication between the infection control team (ICT) and HM unit is important for preventing maintenanceassociated HAIs. Dust prevention is also identified by the healthcare experts as an important measure to prevent maintenance-associated HAIs in high-risk patient areas. Originality/value – The findings provide CSFs and key performance measures for measuring performance in HM in IC. Reducing the rate of maintenance-associated infections will have important socio-economic and health ramifications for hospitals. It will reduce cost and free up additional resources for alternative projects. It will also raise confidence among healthcare users about the quality of services provided by hospitals. Keywords Critical success factors, Delphi, Infection control, Maintenance, National health service, Performance measures, HAIs Paper type Research paper
International Journal of Health Care Quality Assurance Vol. 28 No. 7, 2015 pp. 690-708 © Emerald Group Publishing Limited 0952-6862 DOI 10.1108/IJHCQA-12-2014-0117
Introduction Hospital-acquired infections (HAIs) are a major problem. The European Centre for Disease Control and Prevention (ECDC) estimate the UK HAI rate to be 6 per cent and that 1,602 patients in UK acute care acquire HAIs every year (ECDC, 2013). Although progress is being made to reduce HAIs, the UK lags behind other western European countries. Figures released by the ECDC (2013) show England’s C. difficile rate was higher than in the Netherlands, France, Spain and Italy. High HAI rates are a financial burden for the NHS and costs £1 billion yearly (National Audit Office, 2004). Money spent treating HAIs could be used productively; e.g., in clinical services. Better infection control (IC) practices could reduce HAIs by 15 to 30 per cent (National Audit Office (NAO), 2004). Epidemiological evidence suggests that HAIs can also be caused by poorly performing facilities management (FM) staff; i.e., cleaning, maintenance, laundry and catering have a high impact on IC. A thorough literature review revealed healthcare maintenance (HM) as an areas lacking attention. The Woodbine Report (1970, cited in Allen, 1993, p. 7) defines the term HM as “work undertaken to keep or restore hospital premises to acceptable standards of safety and efficiency having due regard to the needs of patients and staff within the immediate environment, the requirements of the NHS and the resources available”.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Modern healthcare buildings are complex facilities that have different uses and functions (Balaras et al., 2007). Hospital staff need to pay special attention to indoor healing environments (Cheng and Streifel, 2001). Continuous occupancy, upgrading utilities and installing new medical technology challenges may result in potential disruptions, which expose patients to HAIs (Cheng and Streifel, 2001). All these issues pose challenges for healthcare managers who are responsible for rendering a safe environment for patients, visitors and staff. Aspergillosis is an infection caused by construction-related work like HM (Streifel and Hendrickson, 2002; Bartley, 2000; Streifel, 2003). The American Thoracic Society (ATS) (2012) describes aspergillosis as an infection that is caused by a fungus that lives in soil, decaying plants, rotting material and dust, etc. The fungal size (2-3 µm) means they can remain suspended in the air for a significant time (Royal Liverpool Children’s NHS Trust, 2004). Aspergillus enters a susceptible host’s body by inhaled fungal spores (conidia), which may also enter the lungs by inhaling water droplets contaminated with Aspergillus conidia (Thompson and Patterson, 2008). Although Aspergilla are a natural part of the biological ecosystem (Burrill, 2008), they pose a significant risk to patients with compromised immunity through age, underlying illness, medical or surgical treatment ( Joseph, 2006). Hospital maintenance has been implicated in spreading conidia through the airborne route (Hoffman et al., 1999). According to Tabbara and Jabarti (1998), old hospitals (termed sick buildings) are more likely to harbour fungal spores including Aspergillus. In Canada, about 50 per cent of patient deaths were caused by Aspergillus fumigatus (Health Canada, 2001). Such figures have led the CDC (2005, cited in Burrill, 2008, p. 56) to state that “HAIs may be associated with dust exposure during building renovation [maintenance] or construction”. Despite HM’s role controlling HAIs, it has not recieved the attention it deserves from healthcare staff. According to Streifel and Hendrickson (2002), managers generally overlook the risk associated with construction-induced hospital air pollution. They do not spontaneously respond to mechanical ventilation deficiencies especially during construction (Cheng and Streifel, 2001). Additionally, most contractors working on construction-related projects in hospitals are unaccustomed to taking special precaution when tearing down, maintaining or renovating hospitals (Kidd et al., 2007). As a result, many inpatients are exposed to aspergillosis. Even where special precautions are taken, there is doubt whether staff actually manage special ventilation areas to designed parameters specified in various guidelines (Cheng and Streifel, 2001). Because the HM department is under pressure to cut cost (Quayle, 1997 cited in Riley et al., 2004), staff often do not bother to measure IC performance. Where performance is measured, it is mainly ad hoc to meet legislative compliance. Our aim therefore was to identify the critical success factors (CSFs) and key performance measures to control maintenance-associated HAIs in England’s acute hospitals. This was achieved through a Delphi approach. Research method The Delphi technique is a qualitative, long-range forecasting technique that elicits, refines and draws upon expert collective opinion and expertise (Gupta and Clarke, 1996). Our literature review suggests that total rounds in most Delphi studies are variable. Since a three-round Delphi appears ideal for most studies (Delbecq et al., 1975, cited in Skulmoski et al., 2007), we also used three Delphi rounds. Since Delphi relies on
Key performance measures 691
IJHCQA 28,7
expert opinion for credibility, stringent criteria were used to select prospective Delphi participants who were considered eligible if: (1) they occupied HM manager or IC positions (IC doctors, nurses and microbiologists) in an acute NHS trust and had at least five year’s work experience; and
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
692
(2) they participated in the Delphi study and did not delegate responsibility to someone else. The Delphi round 1 instrument was designed to elicit qualitative responses from participants. The first section in the round 1 instrument was about the participants. In the second section, participants were provided with performance measures grouped under eight CSFs. Participants were then asked to identify new ones. The first round exercise (mainly section two) results were used to modify the second round Delphi instrument. Participants provided comments and suggestions that led to re-wording and in some instances re-structuring sections. Results were analysed manually. In the second round, participants were asked to rate different performance measures in HM IC. Rating was based on a four-point Likert scale (1 ¼ very important and 2 ¼ important) represented the positive category and scales 3 and 4 (unimportant and very unimportant) the negative category. Participants were provided with clear instructions on how to complete round 2. The Delphi round 2 elements were assigned the same unique numbers as round 1. They were entered into SPSS vs21 and analysed. For a performance measure to be retained in a Delphi round, there had to be consensus among participants, achieved using means. Unlike other central tendency measures, the mean accounts for every number in the data set (McDonald, 2009). Thus, for a performance measure to be retained in a Delphi round, participants needed a group mean score at least 3.28. Any performance measure with a group mean less than 3.28 was re-submitted to the Delphi participants for re-rating. There is no standard criterion for defining and determining Delphi consensus (Boote et al., 2006). According to Boote et al. (2006, p. 283), the criterion for determining consensus appears “to be an issue for the research team and their advisors”. Performance measures were retained when there was consensus in the second round. However, performance measures with low-level consensus were re-submitted to the Delphi participants for re-rating in round 3, which contained 25 performance measures. For each performance measure, participants were provided with their responses and the overall percentage score in round 2. They were asked to either maintain or re-rate the performance measures on a 1-4 Likert scale. The third round Delphi exercise lasted two weeks. Since participants were the same as those who rated round 2 Delphi questions, they were assigned the same unique numbers. As there were two Delphi groups; i.e., HM managers and IC staff, it was necessary to investigate how each rated the performance measures using the Mann-Whitney U test (statistical significance was set at p ¼ o 0.05). Delphi round 1 Out of 320 invitations posted to prospective Delphi participants, only 40 (13 per cent) acceptances were returned. However, issues with the returned forms meant that only 27 (8.4 per cent) Delphi participants were accepted into the study, including 14 (52 per cent) IC staff and 13 (48 per cent) HM managers. Among the remaining 13 Delphi nominees, four did not have the required work experience. Three individuals had retired or no longer worked at the trust. The last six forms contained e-mail addresses that could not be read.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Attempts to match e-mail addresses with names on the inventory did not help. Although 27 NHS professionals agreed to take part, not all returned the first round Delphi instrument. In total, 20 (74 per cent) participants returned the first round instrument; 11 (55 per cent) were IC staff and nine (45 per cent) HM managers. On average, IC staff and HM manager work experiences were ten and nine years, respectively. One HM manager was also a facilities head in an acute NHS trust. The Delphi participants’ professional experience was more than the five years initially set for this research study. Thus, logically, participants had the required professional experience and knowledge to participate. In the first round, participants were presented with fifty-six performance measures grouped under eight CSFs. They were asked to identify new CSFs and performance measures. Although the Delphi participants did not identify any new CSFs, they identified eleven new performance measures. Among eleven performance measures, only six (identified as R2 in Table I) were added to the second round Delphi questions. Some round 1 Delphi instruments were received after the second Delphi exercise started. The Delphi instruments revealed five new performance measures. Since the second round Delphi exercise had already started, these performance measures (R3 in Table I) could only be included in the third round questions. As shown in Table I, Delphi participants only identified four new CSFs. Delphi round 2 In the second round, participants fell from 20 to 15 (25 per cent attrition). Among fifteen responses, nine came from IC staff and six from HM managers. For a performance measure to be retained in a Delphi round, the two Delphi groups needed a combined 3.28 or above score. In the second Delphi instrument, there were 62 performance measures. However, as shown in Tables II-VI, only 42 were retained in the second round. Tables II-VI clearly show those performance measures retained under eight CSFs. The remaining 20 performance measures with low-level consensus in round 2 featured in round 3 (Tables VII-IX). As shown in Table II, Delphi participants agreed that securing adequate resources is important for mandatory and operational IC compliance. Hospital buildings and infrastructures are supposed to be reviewed and results fed into the IC investment programme. Additionally, a process ought to be in place for introducing new and quality maintenance equipment/fabrics to prevent maintenance-associated HAIs. Among seven performance measures categorized under maintenance strategies, only three were retained in round 2. Prioritizing and timely executing planned maintenance work posing HAI risk and introducing a computer system to coordinate maintenance staff and equipment around the hospital were also important. Table VII shows Delphi participants disagreeing significantly on several performance measures; e.g., there was disagreement ( p ¼ 0.006) about applying a computer-based system to control maintenance-associated HAIs (Table VII). Disagreement ( p ¼ 0.029) also occurred on daily checking all critical maintenance equipment posing HAI risk. On both performance measures, HM managers achieved higher consensus levels than IC staff. Not all performance measures with significant difference in opinion between HM managers and IC members were retained in the second Delphi exercise. The IC practice CSF was divided into three sections: cleaning, transport and administration, containing 18 performance measures. For eight performance measures categorized under cleaning requirements, high-level consensus was achieved on six. Participants agreed on preventing airborne dust dispersal into patient areas. Other important performance measures agreed by the participants are hand hygiene compliance and personal protective equipment. Three important performance measures were also
Key performance measures 693
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
694
CSFs and new performance measures A. Maintenance resources availability 1. Develop processes to control the introduction of new equipment/fabric that can be maintained efficiently to reduce the risk of HAIs 2. Use risk assessment to direct maintenance resources to highest risk activities 3. Involve the HMU and IC department in the purchase of maintenance materials and products B. Maintenance strategies 4. Prioritize and respond to building defects on time to minimize the risk of HAIs 5. Introduce computer system that promotes mobility and allows maintenance staff to carry all the information they require, and communicate back to coordinators when job cannot be completed first time (so that parts/people can be planned in swiftly for revisit) 6. Both the HM and IC team staff to develop a water safety plan (reviewed annually) to identify, manage and control risks of waterborne infections associated with maintenance activities C. Infection control practices Administrative requirement 7. Ensure in-house staff and contractors work on the same clear guidelines 8. Have an agreed HAI plan to control all construction on site. This needs to be reviewed annually to monitor/review/assess level of compliance and provide annual improvement action plan based on benchmark findings 9. Develop a work culture that supports prioritization of maintenance work in infection control
Table I. Delphi round one performance measures
D. Customer satisfaction 10. Ensure visual display of response to complaints 11. Measure the number of completed maintenance jobs that fail to meet the required standard in infection control Notes: R.2, round 2; R.3, round 3
Delphi round included
R.2 R.3 R.3 R.2
R.2 R.3
R.2 R.3 R.3 R.2 R.2
agreed under the transport requirement: health and safety signage, transporting waste, transporting clean and sterile equipment via routes that avoid contamination. Five important performance measures were agreed under the administrative requirement. Participants agreed that HM staff should inform the nurses in charge about any work posing HAI risk. Participants also agreed that in-house and contracted staff should work under the same IC standards and that policies and guidelines be reviewed regularly. However, regarding HM managers obtaining IC permits from the IC department and assessing patients at risk from maintenance-associated HAIs, participants disagreed significantly ( p ¼ 0.028). As shown in Table III, the IC staff mean score was 3.89 (highlevel consensus) and for HM managers, 3.12 (medium consensus). Further disagreement ( p ¼ 0.066) also occurred about adopting safe working system for IC maintenance staff. Consensus for this performance measure was higher for HM managers than IC staff. One performance measure was not retained under the administrative requirement – HM staff pre-employment health check and immunization programme. The risk assessment CSF contained four performance measures and participants agreed that all HM stakeholders should be involved in risk identification and response. Besides educating HM staff on risk identification and responsibility, it was also agreed that reporting, managing and analysing complaints and incidents involving the HM unit in IC are important.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
CSFs and performance measures A. Maintenance resource availability 1. Secure adequate resources for mandatory and operational compliance of the healthcare maintenance unit in infection control 2. Review condition of hospital building services and infrastructure to feed into investment programme 3. Develop processes to control the introduction of new equipment/fabric that can be maintained efficiently and reduce the risk of HAIs 4. Quality maintenance materials and products to be purchased from reliable suppliers B. Maintenance strategies 5. Ensure the timely execution of all planned maintenance work posing risk of infection 6. Prioritize and respond to building defects within time-critical period to minimize the risk of HAIs 7. Introduce computer system to promote mobility and allow maintenance staff to carry all the information they require, and communicate back to coordinators when job cannot be completed first time Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
n
MannMean Mean Combined Whitney U-test ( p) HMM ICM mean
15 3.8333 4.000
3.9333
0.221
15 4.0000 3.7778
3.8667
0.231
15 3.8333 3.8889
3.8667
0.765
15 3.3333 3.3333
3.3333
0.842
15 3.8333 3.6250
3.7143
0.411
15 3.5000 3.6000
3.6000
0.533
15 3.6667 3.1111 3.3333 0.084 infection control member. Results are
Under “liaison and communication”, participants achieved high-level consensus on five performance measures (Table IV); agreeing that early consultation and authorization by IC staff on IC issues was important. It was also agreed that HM workers liaise with individuals managing work areas; i.e., doctors, nurses and domestic staff regarding cleaning during and on completing maintenance work. Because managers in most hospitals now contract-out HM work, communication channels are also needed between in-house and contracted maintenance staff on IC issues. Other important performance measures include accurate record keeping, mandatory conduct codes and contractors taking responsibility for unsafe equipment or practice posing HAI risk. The staff education CSF was divided into two sections: staff training and staff development. Table V shows that five important performance measures were agreed under staff education. Participants agreed on IC statutory and technical guidance, and HM staff training. Besides employing skilled and competent maintenance staff, site induction should be provided to HM staff on IC. Under staff development, participants agreed on representing the HM unit in infection, prevention and control, risk/governance committee. Continuously developing HM staff on risk assessment and management was also accepted as important performance measure to control HAIs. Among six performance measures categorized under “customer satisfaction”, only three achieved high-level consensus (Table V). Participants agreed about measuring maintenance work that fails to meet IC standards. The other two important performance measures concern reviewing and analysing complaints and HM staff response to work request with potential HAI risk. In Delphi round 2, 17 performance measures achieved consensus in only one Delphi group (Table VI). Among 14 performance measures on which HM managers alone
Key performance measures 695
Table II. Round two CSFs and performance measures
IJHCQA 28,7 n
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
696
Table III. Infection control practices and risk assessment
C. Infection control practices Cleaning requirement 8. Provide active means to prevent airborne dust from dispersing into high risk patient areas 9. Compliance with hand hygiene whilst working in clinical areas 10. Compliance with the use of personal protective equipment as required 11. Report any injury especially if “sharp” related, cover wounds or sores 12. Maintenance staff must not work in clinical areas if any symptoms of infection exist, i.e. diarrhoea or vomiting (seek advice from the ICT) 13. Conduct maintenance work in a manner that eases cleaning Transport requirement 14. Health and safety signage used 15. Contain construction waste before transport in tightly covered containers 16. Transport clean and sterile equipment to storage areas via route that minimizes contamination Administrative requirement 17. Inform charge nurse before commencement of maintenance work 18. Ensure in-house and contractors work to same clear guidelines 19. Maintain and review infection control policies and procedures 20. Before commencement of maintenance work, obtain infection control permit, and assess patients for risk of maintenance-associated HAIs 21. Put in place safe working system for maintenance staff in infection prevention D. Risk assessment 22. Involve all stakeholders in risks identification and response (i.e. the ICT) 23. Educate staff and set clear lines of individual responsibility in managing the risk of maintenancerelated infections 24. Process for reporting, managing, and analysing complaints and incidents in infection control Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
MannMean Mean Combined Whitney U-test ( p) HMM ICM mean
15 4.0000 4.0000
4.0000
1.000
15 4.0000 3.7778
3.8667
0.231
15 4.0000 3.7778
3.8667
0.231
15 3.8333 3.5556
3.6667
0.280
15 3.2000 3.8889
3.6429
0.193
14 3.3333 3.4444
3.4000
0.598
15 3.4444 3.4667
3.4667
1.000
15 3.5556 3.4000
3.4000
0.146
15 3.5556 3.4000
3.4000
0.678
15 4.0000 3.8889
3.9333
0.414
15 3.8333 3.7778
3.8000
0.799
15 3.6667 3.5556
3.6000
0.595
15 3.1667 3.8889
3.6000
0.028*
15 3.8333 3.3333
3.5333
0.066*
15 3.5000 3.7778
3.6667
0.280
15 3.5000 3.4444
3.4667
0.837
15 3.5000 3.3333 3.4000 0.533 infection control member. Results are
achieved consensus, six were retained in the second round. Conversely, regarding the remaining three performance measures, high-level consensus was achieved by IC members alone. Overall, in the second Delphi exercise, HM managers achieved consensus on 47 performance measures (39 were retained) while IC members achieved consensus on 36 (all were retained). In total 42 performance measures were retained in the second
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
E. Liaison and communication with ICT 25. Early consultation and authorization from the Infection Control Team before commencement of any maintenance work posing the risk of HAIs 26. Seek the advice of the Infection Control Team (ICT) on such matters concerning infections 27. Liaise with person in charge of area where maintenance is to be carried 28. A system for maintenance staff to liaise with domestic staff regarding cleaning during and on completion of work 29. Set communication channel between maintenance staff and contracted staff F. Service level agreement Contract requirements 30. Contractor should have safe record keeping, and adhere to mandatory code of conduct in infection control 31. Contractor should have arrangement to response to emergency calls 32. Contractor should have procedure to supervise maintenance work and variables, i.e. spares, etc. 33. Select contractors on their strong technical, resource, managerial, and communication capabilities Contracted staff requirements 34. Contractors have to take responsibility for any unsafe equipment, or practice posing risk of infection Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
Mean N HMM
MannMean Combined Whitney U-test ( p) ICM mean
15 4.000
4.0000
4.0000
1.000
15 3.8333 4.0000
3.9333
0.221
15 3.8333 3.6667
3.7333
0.490
15 3.3333 3.6667
3.5333
0.221
15 3.1667 3.4444
3.3333
0.465
15 3.8333 3.6667
3.7333
0.490
15 3.6667 3.4444
3.5333
0.586
15 3.5000 3.2222
3.3333
0.280
15 3.5000 3.1250
3.2857
0.139
15 3.8333 3.7143 3.7692 0.626 infection control member. Results are
Delphi exercise. All performance measures in round 2 with low-level consensus were resubmitted to the Delphi participants for re-rating. Delphi round 3 In the third round, there were 15 participants the same as round 2. Among 25 performance measures contained in the third round Delphi instrument, 20 were re-introduced from the second round. The remaining five performance measures were re-introduced from the first round. Some round 1 Delphi instruments were submitted late, after the second round. Among 25 performance measures contained in the third round, consensus was achieved on 11. As shown in Tables VII-XI, round 3 results are presented in four sections according to the CSFs. Of three performance measures contained under “maintenance resource availability”, two were newly introduced from round 1. As shown in Table VII, the only performance measure in which participants achieved high-level consensus was newly introduced from round 1; i.e., risk assessment to direct resources to maintenance activities posing an HAI risk. The Delphi participants did not agree about involving the HM unit and IC staff in purchasing maintenance materials and products. They also failed to achieve high-level consensus on matching monthly expenditure against maintenance budget in IC.
Key performance measures 697
Table IV. Liaison and communication and service level agreement
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
698
Table V. Staff education and customer satisfaction
MannMean Mean Combined Whitney U-test ( p) N HMM ICM mean G. Staff education Staff training 35. Provide all maintenance staff with information on statutory and technical guidance on infection control 36. Employ skilled and competent staff to ensure safe and efficient maintenance operations 37. Conduct site induction on infection control within few weeks of employment Staff development 38. The maintenance department should be represented in infection prevention and control, risk/governance committees 39. Educate maintenance staff on the assessment and management of risk in maintenance-associated hospital-acquired infections (HAIs) H. Customer satisfaction 40. Measure the number of completed maintenance jobs that failed to meet the required standard in infection control 41. System to review, analyse complaints against maintenance services, and recommend improvement 42. Measure the speed to response to maintenance request Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
15 3.6667 3.5556
3.6000
0.678
14 3.8333 3.2500
3.5000
0.91*
15 3.5000 3.2222
3.3333
0.426
15 3.6667 3.6667
3.6667
0.761
14 3.6000 3.3333
3.4286
0.352
14 3.3333 3.6250
3.5000
0.298
15 3.5000 3.4444
3.4667
0.838
15 3.6667 3.2222 3.4000 0.188 infection control member. Results are
Of five performance measures in the maintenance strategies category, four were re-introduced from the second round. High-level consensus was achieved on three. An important performance measure under this category was the water safety plan to identify, manage, and control waterborne infections risk in maintenance. As shown in Table VII, the grouped mean score for HM managers and IC members was 3.92. Two other important performance measures were agreed under maintenance strategies. The first is about the HM staff accounting for critical maintenance equipment/assets that may cause HAIs. In round 3, the combined mean score for HM managers and IC staff increased from 3.27 to 3.4. The Delphi participants also achieved high-level consensus on applying computer-based maintenance system (i.e. reliability-centred maintenance) to coordinate IC maintenance work. Although the IC staff mean score went up by 0.11, they only attained medium-level consensus. In contrast, the HM managers’ mean score went up by 0.72 and they were able to achieve high-level consensus. The Mann-Whitney U test shows a significant difference ( p ¼ 0.007) between HM managers and IC staff. Despite the difference, participants achieved a combined mean of 3.4. The performance measure on daily checking all critical maintenance systems posing HAI risk did not achieve high-level consensus. The IC practice CSF was divided into three categories containing seven performance measures (Table VIII). Two performance measures presented under cleaning requirements were from the second round and no performance measures achieved high-level consensus.
1. Introduce computer system that promotes mobility and allows maintenance staff to carry all the information they require, and communicate back to coordinators when job cannot be completed first time 3.6667 2. Keep account of the effectiveness of all critical maintenance equipment/assets that may cause HAI 3.5000 3. Use a computer-based maintenance system (i.e. reliability-centred maintenance) to coordinate all maintenance work 3.6667 4. Conduct daily check of all critical maintenance systems posing the risk of HAIs 3.6667 5. Maintenance staff must not work in clinical areas if any symptoms of infection exist, i.e., diarrhoea or vomiting 3.2000 6. Wash and sanitize drainage equipment after use 3.5000 7. Before commencement of maintenance work, obtain infection control permit, and assess patients for risk of maintenance-associated HAIs 3.1667 8. Pre-employment health check and immunization programme for all in-house and contracted maintenance staff 3.3333 9. Set communication channel between maintenance staff and contracted staff 3.1667 10. Contractor should have procedure to supervise maintenance work and variables, i.e., spares, etc. 3.5000 11. Select contractors on their strong technical, resource, managerial, and communication capabilities 3.5000 12. Educate staff and set clear lines of individual responsibility in managing the risk of maintenance-related infections 3.5000 13. Employ skilled and competent staff to ensure safe and efficient maintenance operations 3.8333 14. Conduct site induction on infection control within few weeks of employment 3.5000 15. Conduct annual review of staff training 3.3333 16. Maintenance staff team briefings and appraisal schemes in infection control 3.3333 17. Measure the speed to response to maintenance request 3.6667 Consensus 14 Retained 6 Notes: HMM, healthcare maintenance manager; ICM, infection control member. Results are statistically
Performance measures
3.0000 3.0000 3.8889 3.0000 3.8889 3.2222 3.4444 3.2222 3.1250
Yes Yes No Yes No Yes No Yes Yes
No
No
No Yes
Yes
Yes No
No No
No
No
3.0000 No 3.2500 No 3.2222 No 3.2222 No 3.2143 No 3.2222 No 03 03 significant at 0.05 level
3.1111
Yes
Yes Yes Yes Yes Yes Yes
3.1111
Yes
Yes No No No Yes No Yes No Yes Yes Yes No Yes Yes No No Yes
0.084* 0.107 0.006* 0.029* 0.193 0.400 0.028* 0.699 0.465 0.280 0.139 0.328 0.091* 0.426 0.785 0.524 0.188
MannWhitney HMM ICM Mean Consensus Mean Consensus U-test Retention
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Key performance measures 699
Table VI. Round two performance measures with consensus in only one Delphi group
Table VII. Maintenance resource availability and strategies Medium/ NRT Medium/ NRT
0.926 0.678
High/R
1.000
2.6000
–
0.678
0
–
–
MannWhitney Difference U-test (P) in mean (R3 –R2) R2
B. Maintenance strategies 15 4.0000 3.8571 3.9167 0.398 High/RT – – 4. The development of a water safety plan (reviewed annually) by maintenance and infection control teams, to identify, manage and control risks of waterborne infections associated with maintenance activities 5. Keep account of the effectiveness of all critical maintenance 15 3.5000 3.3333 3.4000 0.533 High/RT 3.2667 0.107 0.1333 equipment/assets that may cause HAI 6. Use a computer-based maintenance system (i.e. reliability-centred 15 3.8333 3.1111 3.4000 0.007* High/RT 3.2667 0.006 0.1333 maintenance) to coordinate all maintenance work 7. Daily check of all critical maintenance systems posing the risk 15 3.5000 3.1111 3.2667 0.221 Medium/ 3.2667 0.029 0 of HAIs NRT 8. Categorize hospital assets, and maintenance equipment into 15 2.8333 3.0000 2.9333 0.500 Medium/ 2.9333 0.697 0 significant and non-significant items in infection control NRT Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
A. Maintenance res. availability 1. Use risk assessment to direct maintenance resources to highest risk 11 3.4000 3.5000 3.4545 activities 2. Involve the HMU and IC department in the purchase of maintenance 12 3.2000 3.2857 3.2500 materials and products 3. Conduct monthly review of expenditure against budget in IC 15 2.6000 2.5556 2.6000
n
MannComb. Whitney U-test (P) Consensus/ mean R2 R3 retention
700
CSFs and performance measures
Comb. Mean Mean mean HMM ICM (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
Medium/ NRT
1.000
–
3.2667
2.8667
3.0000
3.2143
0.3333
0.673
0.699
0.2
0.486
–
0.1619
–
−0.0143 0.400
D. Risk assessment 15 3.0000 3.3333 3.2000 0.286 Medium/ 3.2000 0.943 0 15. Use a recognized risk assessment tool (i.e. infection control risk NRT assessment – ICRA) to match the level of risk associated with maintenance work Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
High/RT
0.652
Medium/ NRT High/RT
0.572
15 3.1667 3.0667 3.2000
Medium/ NRT Medium/ NRT
0.558
0.569
0.255
15 3.3333 3.1111 3.2000
15 3.5000 3.0000 3.2000
Administrative requirements 12. Develop a work culture that supports prioritization of maintenance 12 3.4000 3.7143 3.5833 work in infection control 13. Pre-employment health check and immunization programme for all 14 3.5000 3.3750 3.4286 in-house and contracted maintenance staff 13 3.0000 3.0000 3.0000 14. Have an agreed HAI plan to control all contract works on site. Review plan annually to see level of compliance and provide annual improvement action plan based on previous year’s findings
10. Provide temporal hand washing facilities for maintenance staff working in high risk patient areas Transport requirements 11. Redirect pedestrian traffic from work area
C. Infection control practices Cleaning requirement 9. Wash and sanitize drainage equipment after use
Mean Mean n HMM ICM
MannMannComb. Whitney Difference Comb. Whitney mean U-test (P) Consensus/ mean U-test (P) in mean (R3 –R2) R2 R2 R3 retention (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Key performance measures 701
Table VIII. Infection control practices and risk assessment
Table IX. Liaison and communication and service level agreement 0.894
High/RT
3.2000
0.601
0.1333
MannWhitney Difference U-test (P) in mean (R3 – R2) R2
F. Service level agreement Contract requirements 17. Take into account changes in assets and legislation when 15 3.8333 3.2222 3.4667 0.025* High/RT 3.1333 0.840 0.3334 renewing contracts 18. Customer satisfaction surveys should be part of service level 15 3.0000 2.8889 2.9333 0.673 Medium/ 2.8667 0.840 0.0444 agreement with contractors NRT Contracted staff requirements 19. Contracted workers must attend all mandatory induction and 15 3.8333 3.6250 3.7143 0.653 High/RT 3.2143 0.328 0.5 training on infection control Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
3.3333
MannComb. Whitney U-test (P) Consensus/ mean R2 R3 retention
702
E. Liaison communication with ICT 16. Regularly meet with infection control and clinical representatives 15 3.3333 3.3333 to ensure maintenance processes complement clinical care
n
Comb. Mean Mean mean HMM ICM (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Participants did not consider washing and sanitising drainage equipment and hand washing facilities important. Under transport requirements, the performance measure on re-directing pedestrian traffic from maintenance work areas did not achieve consensus. Although consensus increased slightly in round 3, the two Delphi groups only arrived at a medium consensus. The combined mean score increased from 2.87 in round 2 to 3.2 in round 3. There were three performance measures under administrative requirements. Two were newly introduced in round 3. One newly introduced performance measures on developing work culture that supports prioritizing IC maintenance achieved high-level consensus. The second performance measure on developing a construction HAI plan to manage IC contracted staff did not achieve high-level consensus. The pre-employment health check and immunization programme for maintenance staff, re-introduced from round 2, achieved high-level consensus. Consensus for both groups increased from 3.27 to 3.43. There was only one performance measure under the risk assessment CSF – applying a recognized tool to minimize maintenance-associated HAI risk. In round 3, the mean score for HM managers fell from 3.33 to 3. The IC staff mean score increased from 3.22 to 3.33 (+11). However, with a combined mean score of 3.2 (medium-level consensus), the performance measure was not added to the key performance measures. In Table IX, under the liaison and communication CSF, there was only one performance measure: “holding regular meetings between HM managers, IC and clinical representatives to ensure maintenance work complements clinical care”. As the combined mean score for both Delphi groups went up from 3.2 in round 2 to 3.3, the performance measure was included in the HM IC key performances. Under the service level agreement (SLA) CSF, there were three performance measures. Participants agreed that taking into account changes in assets and legislation when renewing contracts with external providers is important. The HM managers’ mean score increased from 3.17 to 3.87 in round 3. The IC staff mean score also increased, from 3.11 to 3.22 (medium-level consensus) in round 3. In Table IX, there was a significant difference ( p ¼ 0.025) in the agreement between HM managers and IC staff on this performance measure. Despite this difference, the combined mean score for two Delphi groups increased from 3.13 to 3.47 in round 3. Mandatory induction and training for contracted staff also achieved high-level consensus. In Tables VII-XI, between rounds 2 and 3, the combined mean score for HM managers and IC staff increased from 3.21 to 3.71. Therefore, the performance measure is considered important and is included in the key performance measures. On the other performance measure - requiring HM unit staff to have customer satisfaction surveys in the SLA – did not achieve high-level consensus. The staff education CSF contained three performance measures. Participants achieved high-level consensus on annually reviewing staff training, HM staff team briefings and IC appraisal schemes. However, they did not achieve consensus about equal access and improving working lives for HM staff. Delphi participants did not achieve high-level consensus in all three performance measures categorized under customer satisfaction. The first had to do with measurement and total maintenance products that fail to conform to request. The IC staff mean score stayed the same for the two Delphi rounds. In Table X, the combined mean score for both Delphi groups increased slightly (+0.07) between the Delphi rounds. Nevertheless, in round 3, the combined mean for the Delphi participants was only 3.13. Therefore, this performance measure was not included. The other two performance measures not achieving consensus included “visually displaying response to complaints” and “making available complaints boxes and leaflets for people to raise issues about maintenance quality”.
Key performance measures 703
Table X. Staff education and customer satisfaction 3.1333
3.0000
15 3.3333 2.7778
15 3.3333 3.0000
3.2857
15 3.5000 3.1250
0.224
0.107
0.270
0.533
Medium/ NRT
High/RT
High/RT
3.0000
3.2143
3.2667
0.486
0.524
0.785
0
0.0714
0.1333
Medium/ 3.0667 0.724 0.0666 NRT 15 3.1667 2.8889 3.0000 0.324 Medium/ 3.0000 1.000 0 NRT 25. Make available complaint boxes/ leaflets to enable people to 15 2.8333 2.8667 2.8667 0.673 Medium/ 2.8000 0.422 0.0667 raise issues related to quality of maintenance services NRT Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
H. Customer satisfaction 23. Measure the number of maintenance products that do not conform to request 24. Ensure visual display of response to complaints
3.4000
15 3.5000 3.3333
MannMannWhitney Difference Whitney U-test (P) Consensus/ Comb. U-test (P) in mean (R3 – R2) R2 R3 retention mean R2
704
G. Staff education Staff training 20. Conduct annual review of staff training Staff development 21. Maintenance staff team briefings and appraisal schemes in infection control 22. Equal access, and improve working lives for staff
Mean Mean n HMM ICM
Comb. mean (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
From 25 performance measures in the third round, high-level consensus was achieved on 11. In the third round, there were also performance measures with high-level consensus in only one Delphi group. As shown in Table XI, there were eight performance measures on which only HM managers achieved high-level consensus. However, only two performance measures were retained as key performance measures. Neither performance measures on which only IC members achieved high-level consensus were considered as key performance measures in the third round. Discussion and conclusion Between round 2 and 3, Delphi participants identified 53 important HM IC performance measures. One important performance measure relates to preventing dust from spreading in the healthcare environment. Dust contamination in hospital wards (especially in wards with high risk patients) is an important factor in transmitting maintenance-associated HAIs. Healthcare workers’ hands remain a main route for transmitting HAIs (NAO, 2004). According to the Department of Health, poor hand hygiene is linked to hospital infection rates (National Audit Office (NAO), 2009). Where advised, maintenance staff should protect themselves by using personal protective equipment (i.e. overalls and facemasks). They should also report injuries, especially those related to sharps and cover wounds or sores. If there are symptoms; i.e., diarrhoea or vomiting, maintenance staff should report and seek IC advice. New recruits in the HM unit working in close proximity to patients should undergo pre-employment health checks and be immunized according to the same standards applied to clinical staff. Establishing close collaboration between HM and IC team is among the most important CSF. The HM staff need to consult IC staff on all maintenance activities (refurbishment, alteration, maintaining premises/equipment, etc.) with implications for IC. The consultation process must start early to give the IC staff time to assess and respond to IC issues. Basing their judgment on sound evidence, IC staff may either recommend that certain measures be put in place before starting a maintenance project, decide to set up a special committee to assess and monitor any maintenance project impact from start to completion. In the worst-case scenario, IC staff should be allowed to delay or not approve a maintenance project on IC grounds. Despite the benefits of HM working closely with IC staff, the two groups appear to function separately, with HM only requesting ad hoc help from IC staff. A survey conducted by the NAO (2004) found that 17 per cent of NHS Trust managers did not always consult IC staff on issues regarding theatre ventilation or air conditioning/air pressure control systems. A further 22 per cent did not consult IC staff when reviewing plans for alterations and additions to clinical buildings. In our study, HM managers disagreed significantly with IC staff on obtaining IC permits before maintenance work. HM staff that fail to liaise and establish clear communication with IC staff are more likely to perform poorly in IC. Communication between IC and HM staff (in-house and contracted) is central to good IC practices. One underdeveloped HM CSF is customer satisfaction. Customer here refers to anyone (patients, doctors, nurses, etc.) in a healthcare establishment. The only healthcare facilities management (HFM) services that is well known to healthcare users is cleaning (May and Clark, 2009). In the past, because of poor IC, cleaning services faced strong public scrutiny and criticisms. As a result, not enough attention was directed at other HFM services; i.e., maintenance performance in IC. Therefore, to raise standards and give HM staff credence, NHS managers must take maintenance-associated HAIs seriously. It is also important that they raise HM staff profile among healthcare users, who should
Key performance measures 705
Table XI. Round three performance measures with consensus in only one Delphi group
1. Involve the HMU and IC department in the purchase of maintenance materials and products 3.2000 No 3.2857 Yes 2. Use a computer-based maintenance system (i.e. reliability-centred maintenance) to coordinate all maintenance work 3.8333 Yes 3.1111 No 3. Conduct daily check of all critical maintenance systems posing the risk of HAIs 3.5000 Yes 3.1111 No 4. Wash and sanitize drainage equipment after use 3.5000 Yes 3.0000 No 5. Provide temporal hand washing facilities for maintenance staff working in high risk patient areas 3.3333 Yes 3.1111 No 6. Use a recognized risk assessment tool (i.e. infection control risk assessment – ICRA) to match the level of risk associated with maintenance work 3.0000 No 3.3333 Yes 7. Take into account changes in assets and legislation when renewing contracts 3.8333 Yes 3.2222 No 8. Maintenance staff team briefings and appraisal schemes in infection control 3.5000 Yes 3.1250 No 9. Equal access, and improve working lives for staff 3.3333 Yes 2.7778 No 10. Measure the number of maintenance product that do not conform to the request 3.3333 Yes 3.0000 No Consensus 8 2 Retained 2 0 Notes: HMM, healthcare maintenance manager; ICM, infection control member. Results are statistically significant at
Performance measures
706 0.006* 0.029* 0.400 0.486 0.943 0.840 0.524 0.486 0.724
0.007* 0.221 0.255 0.569 0.286 0.025* 0.270 0.107 0.224
0.05 level
– 0.926
No No No Yes No No
−0.657 −0.815 −0.254 −0.379 −0.5
Yes No No
No
0.083
0.001 0.192 −0.145
–
Mann-Whitney U-test HMM ICM Round Round Mean Cons. Mean Cons. (R) 3 (R) 2 R3 – R2 Retained
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
appreciate HM staff contribution to safe patient care. Continuous improvement could also be ensured by putting measure in place to listen to customer views. By giving an ear to customers, HM staff will be in a better position to identify IC areas for further improvement. References Allen, D. (1993), “What is building maintenance?”, Facilities, Vol. 11 No. 3, pp. 7-12. American Thoracic Society (2012), “Aspergillosis fungal disease”, American Journal of Respiratory and Critical Care Medicine, Vol. 186 No. 4, pp. 1-2.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Balaras, C.A., Dascalaki, E. and Gaglia, A. (2007), “HVAC and indoor thermal conditions in hospital operating rooms”, Energy and Buildings, Vol. 39 No. 4, pp. 454-470. Bartley, J.M. (2000), “APIC state-of-the-art report: the role of infection control during construction in health care facilities”, American Journal of Infection Control, Vol. 28 No. 2, pp. 156-169. Boote, J., Barber, R. and Cooper, C. (2006), “Principles and indicators of successful consumer involvement in NHS research: results of a delphi study and subgroup analysis”, Health Policy, Vol. 75 No. 3, pp. 280-297. Burrill, G. (2008), “Healthcare construction under way”, available at: www.touchbriefings.com/ pdf/3202/burrill.pdf (accessed 27 September 2012). Cheng, S.M. and Streifel, A. (2001), “Infection control considerations during construction activities: land excavation and demolition”, American Journal of Infection Control, Vol. 29 No. 5, pp. 321-328. European Centre for Disease Prevention and Control (2013), “Point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals 2011–2012”, Surveillance Report No. 2011–2012, ECDC, Stockholm. Gupta, U.G. and Clarke, R.E. (1996), “Theory and applications of the Delphi technique: a bibliography (1975-1994)”, Technological Forecasting and Social Change, Vol. 53 No. 2, pp. 185-211. Health Canada (2001), “Construction-related nosocomial infections in patients in health care facilities. Decreasing the risk of Aspergillus legionella and other infections communicable Disease Report volume 27S2, Ottawa. Hoffman, P.N., Bennett, A.M. and Scott, G.M. (1999), “Controlling airborne infections”, Journal of Hospital Infection, Vol. 43 No. 1, pp. 203-210. Joseph, A. (2006), “The impact of environment on infections in healthcare facilities”, available at: www.healthdesign.org/sites/default/files/Impact%20of%20the%20Environment%20on% 20Infections%20in%20HC%20Facilities_0.pdf (accessed 22 October 2012). Kidd, F., Buttner, C. and Kressel, A.B. (2007), “Construction: a model program for infection control compliance”, American Journal of Infection Control, Vol. 35 No. 5, pp. 347-350. McDonald, J. (2009), Handbook of Biological Statistics, 3rd ed., Sparky House Publishing, Baltimore, MD. May, D. and Clark, L. (2009), “Achieving patient focussed maintenance services/systems”, Journal of Facilities Management, Vol. 7 No. 2, pp. 128-141. National Audit Office (2004), Improving Patient Care by Reducing the Risk of Hospital Acquired Infection (No. 876 2003-2004), NAO, London. National Audit Office (2009), Reducing Healthcare Associated Infections in Hospitals in England, NAO, London.
Key performance measures 707
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
708
Riley, D., Freihaut, J., Bahnfleth, W. and Karapatyan, Z. (2004), “Indoor air quality management and infection control in healthcare facilities”, available at: www.engr.psu.edu/iec/ publications/papers/indoor_air_quality.pdf (accessed 1 July 2012). Royal Liverpool Children’s NHS Trust (2004), Environmental Control/Aspergillus Policy (No. C32), Royal Liverpool Children’s NHS Trust, Liverpool. Skulmoski, G.J., Hartman, F.T. and Krahn, J. (2007), “The Delphi method for graduate research”, Journal of Information Technology Education, Vol. 6 No. 1, pp. 1-21. Streifel, A.J. (2003), “Hospital accreditation for airborne infection control: how to prepare for environment-of-care inspection for infection control by health-care accrediting organizations”, Heating/Piping/Air Conditioning Engineering, Vol. 75 No. 3, p. 46. Streifel, A. and Hendrickson, C. (2002), “Minimizing the threat of infection from constructioninduced air pollution in healthcare facilities related to construction”, available at: www. industrialairsolutions.com/contamination-control/hospital-air-purifiers-pdf/HPACConstruction-maintenance-health%20care-facilities.pdf (accessed 22 November 2012). Tabbara, K.F. and Jabarti, A.A. (1998), “Hospital construction-associated outbreak of ocular aspergillosis after cataract surgery”, Ophthalmology, Vol. 105 No. 3, pp. 522-526. Thompson, G.R. and Patterson, M. (2008), “Pulmonary aspergillosis”, Seminars in Respiratory and Critical Care Medicine, Vol. 28 No. 2, pp. 103-110.
Corresponding author Dr Stanley Njuangang can be contacted at: [email protected]
For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: [email protected]
Article information:
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
To cite this document: Stanley Njuangang Champika Liyanage Akintola Akintoye , (2015),"Key performance measures to control maintenance-associated HAIs", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 690 - 708 Permanent link to this document: http://dx.doi.org/10.1108/IJHCQA-12-2014-0117 Downloaded on: 31 January 2016, At: 05:07 (PT) References: this document contains references to 24 other documents. To copy this document: [email protected] The fulltext of this document has been downloaded 170 times since 2015*
Users who downloaded this article also downloaded: Sohaib Aleem, William C Torrey, Mathew S Duncan, Shoshana J Hort, John N Mecchella, (2015),"Depression screening optimization in an academic rural setting", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 709-725 http://dx.doi.org/10.1108/ IJHCQA-01-2015-0012 Pia Hannele Jansson von Vultée, (2015),"Healthy work environment – a challenge?", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 660-666 http://dx.doi.org/10.1108/ IJHCQA-11-2014-0108 Mohammadkarim Bahadori, Mehdi Raadabadi, Ramin Ravangard, Donia Baldacchino, (2015),"Factors affecting dental service quality", International Journal of Health Care Quality Assurance, Vol. 28 Iss 7 pp. 678-689 http://dx.doi.org/10.1108/IJHCQA-12-2014-0112
Access to this document was granted through an Emerald subscription provided by emeraldsrm:126209 []
For Authors If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services. Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
*Related content and download information correct at time of download.
The current issue and full text archive of this journal is available on Emerald Insight at: www.emeraldinsight.com/0952-6862.htm
IJHCQA 28,7
690 Received 23 December 2014 Revised 12 March 2015 22 April 2015 Accepted 9 May 2015
Key performance measures to control maintenance-associated HAIs Stanley Njuangang, Champika Liyanage and Akintola Akintoye Grenfell-Baines School of Architecture, Construction and Environment, University of Central Lancashire, Preston, UK Abstract
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Purpose – The purpose of this paper is to improve overall healthcare maintenance (HM) service performance in NHS infection control (IC). Hence, the authors identify critical success factors (CSFs) and key performance measures in maintenance-associated infections. These infections occur because of the poor performance of HM service in IC. Design/methodology/approach – In the first Delphi exercise, complete CSFs and performance measures were presented to the Delphi participants for refinement and modification. Delphi round 1 data were analysed manually and used to refine the rounds 2 and 3 Delphi instruments. In subsequent Delphi rounds, the results were analysed through descriptive statistics. Findings – In total, eight CSFs and 53 key performance measures were identified for reducing maintenance-associated infections in hospitals. For example, establishing clear communication between the infection control team (ICT) and HM unit is important for preventing maintenanceassociated HAIs. Dust prevention is also identified by the healthcare experts as an important measure to prevent maintenance-associated HAIs in high-risk patient areas. Originality/value – The findings provide CSFs and key performance measures for measuring performance in HM in IC. Reducing the rate of maintenance-associated infections will have important socio-economic and health ramifications for hospitals. It will reduce cost and free up additional resources for alternative projects. It will also raise confidence among healthcare users about the quality of services provided by hospitals. Keywords Critical success factors, Delphi, Infection control, Maintenance, National health service, Performance measures, HAIs Paper type Research paper
International Journal of Health Care Quality Assurance Vol. 28 No. 7, 2015 pp. 690-708 © Emerald Group Publishing Limited 0952-6862 DOI 10.1108/IJHCQA-12-2014-0117
Introduction Hospital-acquired infections (HAIs) are a major problem. The European Centre for Disease Control and Prevention (ECDC) estimate the UK HAI rate to be 6 per cent and that 1,602 patients in UK acute care acquire HAIs every year (ECDC, 2013). Although progress is being made to reduce HAIs, the UK lags behind other western European countries. Figures released by the ECDC (2013) show England’s C. difficile rate was higher than in the Netherlands, France, Spain and Italy. High HAI rates are a financial burden for the NHS and costs £1 billion yearly (National Audit Office, 2004). Money spent treating HAIs could be used productively; e.g., in clinical services. Better infection control (IC) practices could reduce HAIs by 15 to 30 per cent (National Audit Office (NAO), 2004). Epidemiological evidence suggests that HAIs can also be caused by poorly performing facilities management (FM) staff; i.e., cleaning, maintenance, laundry and catering have a high impact on IC. A thorough literature review revealed healthcare maintenance (HM) as an areas lacking attention. The Woodbine Report (1970, cited in Allen, 1993, p. 7) defines the term HM as “work undertaken to keep or restore hospital premises to acceptable standards of safety and efficiency having due regard to the needs of patients and staff within the immediate environment, the requirements of the NHS and the resources available”.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Modern healthcare buildings are complex facilities that have different uses and functions (Balaras et al., 2007). Hospital staff need to pay special attention to indoor healing environments (Cheng and Streifel, 2001). Continuous occupancy, upgrading utilities and installing new medical technology challenges may result in potential disruptions, which expose patients to HAIs (Cheng and Streifel, 2001). All these issues pose challenges for healthcare managers who are responsible for rendering a safe environment for patients, visitors and staff. Aspergillosis is an infection caused by construction-related work like HM (Streifel and Hendrickson, 2002; Bartley, 2000; Streifel, 2003). The American Thoracic Society (ATS) (2012) describes aspergillosis as an infection that is caused by a fungus that lives in soil, decaying plants, rotting material and dust, etc. The fungal size (2-3 µm) means they can remain suspended in the air for a significant time (Royal Liverpool Children’s NHS Trust, 2004). Aspergillus enters a susceptible host’s body by inhaled fungal spores (conidia), which may also enter the lungs by inhaling water droplets contaminated with Aspergillus conidia (Thompson and Patterson, 2008). Although Aspergilla are a natural part of the biological ecosystem (Burrill, 2008), they pose a significant risk to patients with compromised immunity through age, underlying illness, medical or surgical treatment ( Joseph, 2006). Hospital maintenance has been implicated in spreading conidia through the airborne route (Hoffman et al., 1999). According to Tabbara and Jabarti (1998), old hospitals (termed sick buildings) are more likely to harbour fungal spores including Aspergillus. In Canada, about 50 per cent of patient deaths were caused by Aspergillus fumigatus (Health Canada, 2001). Such figures have led the CDC (2005, cited in Burrill, 2008, p. 56) to state that “HAIs may be associated with dust exposure during building renovation [maintenance] or construction”. Despite HM’s role controlling HAIs, it has not recieved the attention it deserves from healthcare staff. According to Streifel and Hendrickson (2002), managers generally overlook the risk associated with construction-induced hospital air pollution. They do not spontaneously respond to mechanical ventilation deficiencies especially during construction (Cheng and Streifel, 2001). Additionally, most contractors working on construction-related projects in hospitals are unaccustomed to taking special precaution when tearing down, maintaining or renovating hospitals (Kidd et al., 2007). As a result, many inpatients are exposed to aspergillosis. Even where special precautions are taken, there is doubt whether staff actually manage special ventilation areas to designed parameters specified in various guidelines (Cheng and Streifel, 2001). Because the HM department is under pressure to cut cost (Quayle, 1997 cited in Riley et al., 2004), staff often do not bother to measure IC performance. Where performance is measured, it is mainly ad hoc to meet legislative compliance. Our aim therefore was to identify the critical success factors (CSFs) and key performance measures to control maintenance-associated HAIs in England’s acute hospitals. This was achieved through a Delphi approach. Research method The Delphi technique is a qualitative, long-range forecasting technique that elicits, refines and draws upon expert collective opinion and expertise (Gupta and Clarke, 1996). Our literature review suggests that total rounds in most Delphi studies are variable. Since a three-round Delphi appears ideal for most studies (Delbecq et al., 1975, cited in Skulmoski et al., 2007), we also used three Delphi rounds. Since Delphi relies on
Key performance measures 691
IJHCQA 28,7
expert opinion for credibility, stringent criteria were used to select prospective Delphi participants who were considered eligible if: (1) they occupied HM manager or IC positions (IC doctors, nurses and microbiologists) in an acute NHS trust and had at least five year’s work experience; and
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
692
(2) they participated in the Delphi study and did not delegate responsibility to someone else. The Delphi round 1 instrument was designed to elicit qualitative responses from participants. The first section in the round 1 instrument was about the participants. In the second section, participants were provided with performance measures grouped under eight CSFs. Participants were then asked to identify new ones. The first round exercise (mainly section two) results were used to modify the second round Delphi instrument. Participants provided comments and suggestions that led to re-wording and in some instances re-structuring sections. Results were analysed manually. In the second round, participants were asked to rate different performance measures in HM IC. Rating was based on a four-point Likert scale (1 ¼ very important and 2 ¼ important) represented the positive category and scales 3 and 4 (unimportant and very unimportant) the negative category. Participants were provided with clear instructions on how to complete round 2. The Delphi round 2 elements were assigned the same unique numbers as round 1. They were entered into SPSS vs21 and analysed. For a performance measure to be retained in a Delphi round, there had to be consensus among participants, achieved using means. Unlike other central tendency measures, the mean accounts for every number in the data set (McDonald, 2009). Thus, for a performance measure to be retained in a Delphi round, participants needed a group mean score at least 3.28. Any performance measure with a group mean less than 3.28 was re-submitted to the Delphi participants for re-rating. There is no standard criterion for defining and determining Delphi consensus (Boote et al., 2006). According to Boote et al. (2006, p. 283), the criterion for determining consensus appears “to be an issue for the research team and their advisors”. Performance measures were retained when there was consensus in the second round. However, performance measures with low-level consensus were re-submitted to the Delphi participants for re-rating in round 3, which contained 25 performance measures. For each performance measure, participants were provided with their responses and the overall percentage score in round 2. They were asked to either maintain or re-rate the performance measures on a 1-4 Likert scale. The third round Delphi exercise lasted two weeks. Since participants were the same as those who rated round 2 Delphi questions, they were assigned the same unique numbers. As there were two Delphi groups; i.e., HM managers and IC staff, it was necessary to investigate how each rated the performance measures using the Mann-Whitney U test (statistical significance was set at p ¼ o 0.05). Delphi round 1 Out of 320 invitations posted to prospective Delphi participants, only 40 (13 per cent) acceptances were returned. However, issues with the returned forms meant that only 27 (8.4 per cent) Delphi participants were accepted into the study, including 14 (52 per cent) IC staff and 13 (48 per cent) HM managers. Among the remaining 13 Delphi nominees, four did not have the required work experience. Three individuals had retired or no longer worked at the trust. The last six forms contained e-mail addresses that could not be read.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Attempts to match e-mail addresses with names on the inventory did not help. Although 27 NHS professionals agreed to take part, not all returned the first round Delphi instrument. In total, 20 (74 per cent) participants returned the first round instrument; 11 (55 per cent) were IC staff and nine (45 per cent) HM managers. On average, IC staff and HM manager work experiences were ten and nine years, respectively. One HM manager was also a facilities head in an acute NHS trust. The Delphi participants’ professional experience was more than the five years initially set for this research study. Thus, logically, participants had the required professional experience and knowledge to participate. In the first round, participants were presented with fifty-six performance measures grouped under eight CSFs. They were asked to identify new CSFs and performance measures. Although the Delphi participants did not identify any new CSFs, they identified eleven new performance measures. Among eleven performance measures, only six (identified as R2 in Table I) were added to the second round Delphi questions. Some round 1 Delphi instruments were received after the second Delphi exercise started. The Delphi instruments revealed five new performance measures. Since the second round Delphi exercise had already started, these performance measures (R3 in Table I) could only be included in the third round questions. As shown in Table I, Delphi participants only identified four new CSFs. Delphi round 2 In the second round, participants fell from 20 to 15 (25 per cent attrition). Among fifteen responses, nine came from IC staff and six from HM managers. For a performance measure to be retained in a Delphi round, the two Delphi groups needed a combined 3.28 or above score. In the second Delphi instrument, there were 62 performance measures. However, as shown in Tables II-VI, only 42 were retained in the second round. Tables II-VI clearly show those performance measures retained under eight CSFs. The remaining 20 performance measures with low-level consensus in round 2 featured in round 3 (Tables VII-IX). As shown in Table II, Delphi participants agreed that securing adequate resources is important for mandatory and operational IC compliance. Hospital buildings and infrastructures are supposed to be reviewed and results fed into the IC investment programme. Additionally, a process ought to be in place for introducing new and quality maintenance equipment/fabrics to prevent maintenance-associated HAIs. Among seven performance measures categorized under maintenance strategies, only three were retained in round 2. Prioritizing and timely executing planned maintenance work posing HAI risk and introducing a computer system to coordinate maintenance staff and equipment around the hospital were also important. Table VII shows Delphi participants disagreeing significantly on several performance measures; e.g., there was disagreement ( p ¼ 0.006) about applying a computer-based system to control maintenance-associated HAIs (Table VII). Disagreement ( p ¼ 0.029) also occurred on daily checking all critical maintenance equipment posing HAI risk. On both performance measures, HM managers achieved higher consensus levels than IC staff. Not all performance measures with significant difference in opinion between HM managers and IC members were retained in the second Delphi exercise. The IC practice CSF was divided into three sections: cleaning, transport and administration, containing 18 performance measures. For eight performance measures categorized under cleaning requirements, high-level consensus was achieved on six. Participants agreed on preventing airborne dust dispersal into patient areas. Other important performance measures agreed by the participants are hand hygiene compliance and personal protective equipment. Three important performance measures were also
Key performance measures 693
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
694
CSFs and new performance measures A. Maintenance resources availability 1. Develop processes to control the introduction of new equipment/fabric that can be maintained efficiently to reduce the risk of HAIs 2. Use risk assessment to direct maintenance resources to highest risk activities 3. Involve the HMU and IC department in the purchase of maintenance materials and products B. Maintenance strategies 4. Prioritize and respond to building defects on time to minimize the risk of HAIs 5. Introduce computer system that promotes mobility and allows maintenance staff to carry all the information they require, and communicate back to coordinators when job cannot be completed first time (so that parts/people can be planned in swiftly for revisit) 6. Both the HM and IC team staff to develop a water safety plan (reviewed annually) to identify, manage and control risks of waterborne infections associated with maintenance activities C. Infection control practices Administrative requirement 7. Ensure in-house staff and contractors work on the same clear guidelines 8. Have an agreed HAI plan to control all construction on site. This needs to be reviewed annually to monitor/review/assess level of compliance and provide annual improvement action plan based on benchmark findings 9. Develop a work culture that supports prioritization of maintenance work in infection control
Table I. Delphi round one performance measures
D. Customer satisfaction 10. Ensure visual display of response to complaints 11. Measure the number of completed maintenance jobs that fail to meet the required standard in infection control Notes: R.2, round 2; R.3, round 3
Delphi round included
R.2 R.3 R.3 R.2
R.2 R.3
R.2 R.3 R.3 R.2 R.2
agreed under the transport requirement: health and safety signage, transporting waste, transporting clean and sterile equipment via routes that avoid contamination. Five important performance measures were agreed under the administrative requirement. Participants agreed that HM staff should inform the nurses in charge about any work posing HAI risk. Participants also agreed that in-house and contracted staff should work under the same IC standards and that policies and guidelines be reviewed regularly. However, regarding HM managers obtaining IC permits from the IC department and assessing patients at risk from maintenance-associated HAIs, participants disagreed significantly ( p ¼ 0.028). As shown in Table III, the IC staff mean score was 3.89 (highlevel consensus) and for HM managers, 3.12 (medium consensus). Further disagreement ( p ¼ 0.066) also occurred about adopting safe working system for IC maintenance staff. Consensus for this performance measure was higher for HM managers than IC staff. One performance measure was not retained under the administrative requirement – HM staff pre-employment health check and immunization programme. The risk assessment CSF contained four performance measures and participants agreed that all HM stakeholders should be involved in risk identification and response. Besides educating HM staff on risk identification and responsibility, it was also agreed that reporting, managing and analysing complaints and incidents involving the HM unit in IC are important.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
CSFs and performance measures A. Maintenance resource availability 1. Secure adequate resources for mandatory and operational compliance of the healthcare maintenance unit in infection control 2. Review condition of hospital building services and infrastructure to feed into investment programme 3. Develop processes to control the introduction of new equipment/fabric that can be maintained efficiently and reduce the risk of HAIs 4. Quality maintenance materials and products to be purchased from reliable suppliers B. Maintenance strategies 5. Ensure the timely execution of all planned maintenance work posing risk of infection 6. Prioritize and respond to building defects within time-critical period to minimize the risk of HAIs 7. Introduce computer system to promote mobility and allow maintenance staff to carry all the information they require, and communicate back to coordinators when job cannot be completed first time Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
n
MannMean Mean Combined Whitney U-test ( p) HMM ICM mean
15 3.8333 4.000
3.9333
0.221
15 4.0000 3.7778
3.8667
0.231
15 3.8333 3.8889
3.8667
0.765
15 3.3333 3.3333
3.3333
0.842
15 3.8333 3.6250
3.7143
0.411
15 3.5000 3.6000
3.6000
0.533
15 3.6667 3.1111 3.3333 0.084 infection control member. Results are
Under “liaison and communication”, participants achieved high-level consensus on five performance measures (Table IV); agreeing that early consultation and authorization by IC staff on IC issues was important. It was also agreed that HM workers liaise with individuals managing work areas; i.e., doctors, nurses and domestic staff regarding cleaning during and on completing maintenance work. Because managers in most hospitals now contract-out HM work, communication channels are also needed between in-house and contracted maintenance staff on IC issues. Other important performance measures include accurate record keeping, mandatory conduct codes and contractors taking responsibility for unsafe equipment or practice posing HAI risk. The staff education CSF was divided into two sections: staff training and staff development. Table V shows that five important performance measures were agreed under staff education. Participants agreed on IC statutory and technical guidance, and HM staff training. Besides employing skilled and competent maintenance staff, site induction should be provided to HM staff on IC. Under staff development, participants agreed on representing the HM unit in infection, prevention and control, risk/governance committee. Continuously developing HM staff on risk assessment and management was also accepted as important performance measure to control HAIs. Among six performance measures categorized under “customer satisfaction”, only three achieved high-level consensus (Table V). Participants agreed about measuring maintenance work that fails to meet IC standards. The other two important performance measures concern reviewing and analysing complaints and HM staff response to work request with potential HAI risk. In Delphi round 2, 17 performance measures achieved consensus in only one Delphi group (Table VI). Among 14 performance measures on which HM managers alone
Key performance measures 695
Table II. Round two CSFs and performance measures
IJHCQA 28,7 n
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
696
Table III. Infection control practices and risk assessment
C. Infection control practices Cleaning requirement 8. Provide active means to prevent airborne dust from dispersing into high risk patient areas 9. Compliance with hand hygiene whilst working in clinical areas 10. Compliance with the use of personal protective equipment as required 11. Report any injury especially if “sharp” related, cover wounds or sores 12. Maintenance staff must not work in clinical areas if any symptoms of infection exist, i.e. diarrhoea or vomiting (seek advice from the ICT) 13. Conduct maintenance work in a manner that eases cleaning Transport requirement 14. Health and safety signage used 15. Contain construction waste before transport in tightly covered containers 16. Transport clean and sterile equipment to storage areas via route that minimizes contamination Administrative requirement 17. Inform charge nurse before commencement of maintenance work 18. Ensure in-house and contractors work to same clear guidelines 19. Maintain and review infection control policies and procedures 20. Before commencement of maintenance work, obtain infection control permit, and assess patients for risk of maintenance-associated HAIs 21. Put in place safe working system for maintenance staff in infection prevention D. Risk assessment 22. Involve all stakeholders in risks identification and response (i.e. the ICT) 23. Educate staff and set clear lines of individual responsibility in managing the risk of maintenancerelated infections 24. Process for reporting, managing, and analysing complaints and incidents in infection control Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
MannMean Mean Combined Whitney U-test ( p) HMM ICM mean
15 4.0000 4.0000
4.0000
1.000
15 4.0000 3.7778
3.8667
0.231
15 4.0000 3.7778
3.8667
0.231
15 3.8333 3.5556
3.6667
0.280
15 3.2000 3.8889
3.6429
0.193
14 3.3333 3.4444
3.4000
0.598
15 3.4444 3.4667
3.4667
1.000
15 3.5556 3.4000
3.4000
0.146
15 3.5556 3.4000
3.4000
0.678
15 4.0000 3.8889
3.9333
0.414
15 3.8333 3.7778
3.8000
0.799
15 3.6667 3.5556
3.6000
0.595
15 3.1667 3.8889
3.6000
0.028*
15 3.8333 3.3333
3.5333
0.066*
15 3.5000 3.7778
3.6667
0.280
15 3.5000 3.4444
3.4667
0.837
15 3.5000 3.3333 3.4000 0.533 infection control member. Results are
achieved consensus, six were retained in the second round. Conversely, regarding the remaining three performance measures, high-level consensus was achieved by IC members alone. Overall, in the second Delphi exercise, HM managers achieved consensus on 47 performance measures (39 were retained) while IC members achieved consensus on 36 (all were retained). In total 42 performance measures were retained in the second
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
E. Liaison and communication with ICT 25. Early consultation and authorization from the Infection Control Team before commencement of any maintenance work posing the risk of HAIs 26. Seek the advice of the Infection Control Team (ICT) on such matters concerning infections 27. Liaise with person in charge of area where maintenance is to be carried 28. A system for maintenance staff to liaise with domestic staff regarding cleaning during and on completion of work 29. Set communication channel between maintenance staff and contracted staff F. Service level agreement Contract requirements 30. Contractor should have safe record keeping, and adhere to mandatory code of conduct in infection control 31. Contractor should have arrangement to response to emergency calls 32. Contractor should have procedure to supervise maintenance work and variables, i.e. spares, etc. 33. Select contractors on their strong technical, resource, managerial, and communication capabilities Contracted staff requirements 34. Contractors have to take responsibility for any unsafe equipment, or practice posing risk of infection Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
Mean N HMM
MannMean Combined Whitney U-test ( p) ICM mean
15 4.000
4.0000
4.0000
1.000
15 3.8333 4.0000
3.9333
0.221
15 3.8333 3.6667
3.7333
0.490
15 3.3333 3.6667
3.5333
0.221
15 3.1667 3.4444
3.3333
0.465
15 3.8333 3.6667
3.7333
0.490
15 3.6667 3.4444
3.5333
0.586
15 3.5000 3.2222
3.3333
0.280
15 3.5000 3.1250
3.2857
0.139
15 3.8333 3.7143 3.7692 0.626 infection control member. Results are
Delphi exercise. All performance measures in round 2 with low-level consensus were resubmitted to the Delphi participants for re-rating. Delphi round 3 In the third round, there were 15 participants the same as round 2. Among 25 performance measures contained in the third round Delphi instrument, 20 were re-introduced from the second round. The remaining five performance measures were re-introduced from the first round. Some round 1 Delphi instruments were submitted late, after the second round. Among 25 performance measures contained in the third round, consensus was achieved on 11. As shown in Tables VII-XI, round 3 results are presented in four sections according to the CSFs. Of three performance measures contained under “maintenance resource availability”, two were newly introduced from round 1. As shown in Table VII, the only performance measure in which participants achieved high-level consensus was newly introduced from round 1; i.e., risk assessment to direct resources to maintenance activities posing an HAI risk. The Delphi participants did not agree about involving the HM unit and IC staff in purchasing maintenance materials and products. They also failed to achieve high-level consensus on matching monthly expenditure against maintenance budget in IC.
Key performance measures 697
Table IV. Liaison and communication and service level agreement
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
698
Table V. Staff education and customer satisfaction
MannMean Mean Combined Whitney U-test ( p) N HMM ICM mean G. Staff education Staff training 35. Provide all maintenance staff with information on statutory and technical guidance on infection control 36. Employ skilled and competent staff to ensure safe and efficient maintenance operations 37. Conduct site induction on infection control within few weeks of employment Staff development 38. The maintenance department should be represented in infection prevention and control, risk/governance committees 39. Educate maintenance staff on the assessment and management of risk in maintenance-associated hospital-acquired infections (HAIs) H. Customer satisfaction 40. Measure the number of completed maintenance jobs that failed to meet the required standard in infection control 41. System to review, analyse complaints against maintenance services, and recommend improvement 42. Measure the speed to response to maintenance request Notes: HMM, healthcare maintenance manager; ICM, statistically significant at 0.05 level
15 3.6667 3.5556
3.6000
0.678
14 3.8333 3.2500
3.5000
0.91*
15 3.5000 3.2222
3.3333
0.426
15 3.6667 3.6667
3.6667
0.761
14 3.6000 3.3333
3.4286
0.352
14 3.3333 3.6250
3.5000
0.298
15 3.5000 3.4444
3.4667
0.838
15 3.6667 3.2222 3.4000 0.188 infection control member. Results are
Of five performance measures in the maintenance strategies category, four were re-introduced from the second round. High-level consensus was achieved on three. An important performance measure under this category was the water safety plan to identify, manage, and control waterborne infections risk in maintenance. As shown in Table VII, the grouped mean score for HM managers and IC members was 3.92. Two other important performance measures were agreed under maintenance strategies. The first is about the HM staff accounting for critical maintenance equipment/assets that may cause HAIs. In round 3, the combined mean score for HM managers and IC staff increased from 3.27 to 3.4. The Delphi participants also achieved high-level consensus on applying computer-based maintenance system (i.e. reliability-centred maintenance) to coordinate IC maintenance work. Although the IC staff mean score went up by 0.11, they only attained medium-level consensus. In contrast, the HM managers’ mean score went up by 0.72 and they were able to achieve high-level consensus. The Mann-Whitney U test shows a significant difference ( p ¼ 0.007) between HM managers and IC staff. Despite the difference, participants achieved a combined mean of 3.4. The performance measure on daily checking all critical maintenance systems posing HAI risk did not achieve high-level consensus. The IC practice CSF was divided into three categories containing seven performance measures (Table VIII). Two performance measures presented under cleaning requirements were from the second round and no performance measures achieved high-level consensus.
1. Introduce computer system that promotes mobility and allows maintenance staff to carry all the information they require, and communicate back to coordinators when job cannot be completed first time 3.6667 2. Keep account of the effectiveness of all critical maintenance equipment/assets that may cause HAI 3.5000 3. Use a computer-based maintenance system (i.e. reliability-centred maintenance) to coordinate all maintenance work 3.6667 4. Conduct daily check of all critical maintenance systems posing the risk of HAIs 3.6667 5. Maintenance staff must not work in clinical areas if any symptoms of infection exist, i.e., diarrhoea or vomiting 3.2000 6. Wash and sanitize drainage equipment after use 3.5000 7. Before commencement of maintenance work, obtain infection control permit, and assess patients for risk of maintenance-associated HAIs 3.1667 8. Pre-employment health check and immunization programme for all in-house and contracted maintenance staff 3.3333 9. Set communication channel between maintenance staff and contracted staff 3.1667 10. Contractor should have procedure to supervise maintenance work and variables, i.e., spares, etc. 3.5000 11. Select contractors on their strong technical, resource, managerial, and communication capabilities 3.5000 12. Educate staff and set clear lines of individual responsibility in managing the risk of maintenance-related infections 3.5000 13. Employ skilled and competent staff to ensure safe and efficient maintenance operations 3.8333 14. Conduct site induction on infection control within few weeks of employment 3.5000 15. Conduct annual review of staff training 3.3333 16. Maintenance staff team briefings and appraisal schemes in infection control 3.3333 17. Measure the speed to response to maintenance request 3.6667 Consensus 14 Retained 6 Notes: HMM, healthcare maintenance manager; ICM, infection control member. Results are statistically
Performance measures
3.0000 3.0000 3.8889 3.0000 3.8889 3.2222 3.4444 3.2222 3.1250
Yes Yes No Yes No Yes No Yes Yes
No
No
No Yes
Yes
Yes No
No No
No
No
3.0000 No 3.2500 No 3.2222 No 3.2222 No 3.2143 No 3.2222 No 03 03 significant at 0.05 level
3.1111
Yes
Yes Yes Yes Yes Yes Yes
3.1111
Yes
Yes No No No Yes No Yes No Yes Yes Yes No Yes Yes No No Yes
0.084* 0.107 0.006* 0.029* 0.193 0.400 0.028* 0.699 0.465 0.280 0.139 0.328 0.091* 0.426 0.785 0.524 0.188
MannWhitney HMM ICM Mean Consensus Mean Consensus U-test Retention
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Key performance measures 699
Table VI. Round two performance measures with consensus in only one Delphi group
Table VII. Maintenance resource availability and strategies Medium/ NRT Medium/ NRT
0.926 0.678
High/R
1.000
2.6000
–
0.678
0
–
–
MannWhitney Difference U-test (P) in mean (R3 –R2) R2
B. Maintenance strategies 15 4.0000 3.8571 3.9167 0.398 High/RT – – 4. The development of a water safety plan (reviewed annually) by maintenance and infection control teams, to identify, manage and control risks of waterborne infections associated with maintenance activities 5. Keep account of the effectiveness of all critical maintenance 15 3.5000 3.3333 3.4000 0.533 High/RT 3.2667 0.107 0.1333 equipment/assets that may cause HAI 6. Use a computer-based maintenance system (i.e. reliability-centred 15 3.8333 3.1111 3.4000 0.007* High/RT 3.2667 0.006 0.1333 maintenance) to coordinate all maintenance work 7. Daily check of all critical maintenance systems posing the risk 15 3.5000 3.1111 3.2667 0.221 Medium/ 3.2667 0.029 0 of HAIs NRT 8. Categorize hospital assets, and maintenance equipment into 15 2.8333 3.0000 2.9333 0.500 Medium/ 2.9333 0.697 0 significant and non-significant items in infection control NRT Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
A. Maintenance res. availability 1. Use risk assessment to direct maintenance resources to highest risk 11 3.4000 3.5000 3.4545 activities 2. Involve the HMU and IC department in the purchase of maintenance 12 3.2000 3.2857 3.2500 materials and products 3. Conduct monthly review of expenditure against budget in IC 15 2.6000 2.5556 2.6000
n
MannComb. Whitney U-test (P) Consensus/ mean R2 R3 retention
700
CSFs and performance measures
Comb. Mean Mean mean HMM ICM (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
Medium/ NRT
1.000
–
3.2667
2.8667
3.0000
3.2143
0.3333
0.673
0.699
0.2
0.486
–
0.1619
–
−0.0143 0.400
D. Risk assessment 15 3.0000 3.3333 3.2000 0.286 Medium/ 3.2000 0.943 0 15. Use a recognized risk assessment tool (i.e. infection control risk NRT assessment – ICRA) to match the level of risk associated with maintenance work Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
High/RT
0.652
Medium/ NRT High/RT
0.572
15 3.1667 3.0667 3.2000
Medium/ NRT Medium/ NRT
0.558
0.569
0.255
15 3.3333 3.1111 3.2000
15 3.5000 3.0000 3.2000
Administrative requirements 12. Develop a work culture that supports prioritization of maintenance 12 3.4000 3.7143 3.5833 work in infection control 13. Pre-employment health check and immunization programme for all 14 3.5000 3.3750 3.4286 in-house and contracted maintenance staff 13 3.0000 3.0000 3.0000 14. Have an agreed HAI plan to control all contract works on site. Review plan annually to see level of compliance and provide annual improvement action plan based on previous year’s findings
10. Provide temporal hand washing facilities for maintenance staff working in high risk patient areas Transport requirements 11. Redirect pedestrian traffic from work area
C. Infection control practices Cleaning requirement 9. Wash and sanitize drainage equipment after use
Mean Mean n HMM ICM
MannMannComb. Whitney Difference Comb. Whitney mean U-test (P) Consensus/ mean U-test (P) in mean (R3 –R2) R2 R2 R3 retention (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Key performance measures 701
Table VIII. Infection control practices and risk assessment
Table IX. Liaison and communication and service level agreement 0.894
High/RT
3.2000
0.601
0.1333
MannWhitney Difference U-test (P) in mean (R3 – R2) R2
F. Service level agreement Contract requirements 17. Take into account changes in assets and legislation when 15 3.8333 3.2222 3.4667 0.025* High/RT 3.1333 0.840 0.3334 renewing contracts 18. Customer satisfaction surveys should be part of service level 15 3.0000 2.8889 2.9333 0.673 Medium/ 2.8667 0.840 0.0444 agreement with contractors NRT Contracted staff requirements 19. Contracted workers must attend all mandatory induction and 15 3.8333 3.6250 3.7143 0.653 High/RT 3.2143 0.328 0.5 training on infection control Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
3.3333
MannComb. Whitney U-test (P) Consensus/ mean R2 R3 retention
702
E. Liaison communication with ICT 16. Regularly meet with infection control and clinical representatives 15 3.3333 3.3333 to ensure maintenance processes complement clinical care
n
Comb. Mean Mean mean HMM ICM (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Participants did not consider washing and sanitising drainage equipment and hand washing facilities important. Under transport requirements, the performance measure on re-directing pedestrian traffic from maintenance work areas did not achieve consensus. Although consensus increased slightly in round 3, the two Delphi groups only arrived at a medium consensus. The combined mean score increased from 2.87 in round 2 to 3.2 in round 3. There were three performance measures under administrative requirements. Two were newly introduced in round 3. One newly introduced performance measures on developing work culture that supports prioritizing IC maintenance achieved high-level consensus. The second performance measure on developing a construction HAI plan to manage IC contracted staff did not achieve high-level consensus. The pre-employment health check and immunization programme for maintenance staff, re-introduced from round 2, achieved high-level consensus. Consensus for both groups increased from 3.27 to 3.43. There was only one performance measure under the risk assessment CSF – applying a recognized tool to minimize maintenance-associated HAI risk. In round 3, the mean score for HM managers fell from 3.33 to 3. The IC staff mean score increased from 3.22 to 3.33 (+11). However, with a combined mean score of 3.2 (medium-level consensus), the performance measure was not added to the key performance measures. In Table IX, under the liaison and communication CSF, there was only one performance measure: “holding regular meetings between HM managers, IC and clinical representatives to ensure maintenance work complements clinical care”. As the combined mean score for both Delphi groups went up from 3.2 in round 2 to 3.3, the performance measure was included in the HM IC key performances. Under the service level agreement (SLA) CSF, there were three performance measures. Participants agreed that taking into account changes in assets and legislation when renewing contracts with external providers is important. The HM managers’ mean score increased from 3.17 to 3.87 in round 3. The IC staff mean score also increased, from 3.11 to 3.22 (medium-level consensus) in round 3. In Table IX, there was a significant difference ( p ¼ 0.025) in the agreement between HM managers and IC staff on this performance measure. Despite this difference, the combined mean score for two Delphi groups increased from 3.13 to 3.47 in round 3. Mandatory induction and training for contracted staff also achieved high-level consensus. In Tables VII-XI, between rounds 2 and 3, the combined mean score for HM managers and IC staff increased from 3.21 to 3.71. Therefore, the performance measure is considered important and is included in the key performance measures. On the other performance measure - requiring HM unit staff to have customer satisfaction surveys in the SLA – did not achieve high-level consensus. The staff education CSF contained three performance measures. Participants achieved high-level consensus on annually reviewing staff training, HM staff team briefings and IC appraisal schemes. However, they did not achieve consensus about equal access and improving working lives for HM staff. Delphi participants did not achieve high-level consensus in all three performance measures categorized under customer satisfaction. The first had to do with measurement and total maintenance products that fail to conform to request. The IC staff mean score stayed the same for the two Delphi rounds. In Table X, the combined mean score for both Delphi groups increased slightly (+0.07) between the Delphi rounds. Nevertheless, in round 3, the combined mean for the Delphi participants was only 3.13. Therefore, this performance measure was not included. The other two performance measures not achieving consensus included “visually displaying response to complaints” and “making available complaints boxes and leaflets for people to raise issues about maintenance quality”.
Key performance measures 703
Table X. Staff education and customer satisfaction 3.1333
3.0000
15 3.3333 2.7778
15 3.3333 3.0000
3.2857
15 3.5000 3.1250
0.224
0.107
0.270
0.533
Medium/ NRT
High/RT
High/RT
3.0000
3.2143
3.2667
0.486
0.524
0.785
0
0.0714
0.1333
Medium/ 3.0667 0.724 0.0666 NRT 15 3.1667 2.8889 3.0000 0.324 Medium/ 3.0000 1.000 0 NRT 25. Make available complaint boxes/ leaflets to enable people to 15 2.8333 2.8667 2.8667 0.673 Medium/ 2.8000 0.422 0.0667 raise issues related to quality of maintenance services NRT Notes: HMM, healthcare maintenance managers; ICM, infection control member; R, delphi rounds; RT, retained; NRT, not retained. Results are statistically significant at 0.05 level
H. Customer satisfaction 23. Measure the number of maintenance products that do not conform to request 24. Ensure visual display of response to complaints
3.4000
15 3.5000 3.3333
MannMannWhitney Difference Whitney U-test (P) Consensus/ Comb. U-test (P) in mean (R3 – R2) R2 R3 retention mean R2
704
G. Staff education Staff training 20. Conduct annual review of staff training Staff development 21. Maintenance staff team briefings and appraisal schemes in infection control 22. Equal access, and improve working lives for staff
Mean Mean n HMM ICM
Comb. mean (R3)
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
From 25 performance measures in the third round, high-level consensus was achieved on 11. In the third round, there were also performance measures with high-level consensus in only one Delphi group. As shown in Table XI, there were eight performance measures on which only HM managers achieved high-level consensus. However, only two performance measures were retained as key performance measures. Neither performance measures on which only IC members achieved high-level consensus were considered as key performance measures in the third round. Discussion and conclusion Between round 2 and 3, Delphi participants identified 53 important HM IC performance measures. One important performance measure relates to preventing dust from spreading in the healthcare environment. Dust contamination in hospital wards (especially in wards with high risk patients) is an important factor in transmitting maintenance-associated HAIs. Healthcare workers’ hands remain a main route for transmitting HAIs (NAO, 2004). According to the Department of Health, poor hand hygiene is linked to hospital infection rates (National Audit Office (NAO), 2009). Where advised, maintenance staff should protect themselves by using personal protective equipment (i.e. overalls and facemasks). They should also report injuries, especially those related to sharps and cover wounds or sores. If there are symptoms; i.e., diarrhoea or vomiting, maintenance staff should report and seek IC advice. New recruits in the HM unit working in close proximity to patients should undergo pre-employment health checks and be immunized according to the same standards applied to clinical staff. Establishing close collaboration between HM and IC team is among the most important CSF. The HM staff need to consult IC staff on all maintenance activities (refurbishment, alteration, maintaining premises/equipment, etc.) with implications for IC. The consultation process must start early to give the IC staff time to assess and respond to IC issues. Basing their judgment on sound evidence, IC staff may either recommend that certain measures be put in place before starting a maintenance project, decide to set up a special committee to assess and monitor any maintenance project impact from start to completion. In the worst-case scenario, IC staff should be allowed to delay or not approve a maintenance project on IC grounds. Despite the benefits of HM working closely with IC staff, the two groups appear to function separately, with HM only requesting ad hoc help from IC staff. A survey conducted by the NAO (2004) found that 17 per cent of NHS Trust managers did not always consult IC staff on issues regarding theatre ventilation or air conditioning/air pressure control systems. A further 22 per cent did not consult IC staff when reviewing plans for alterations and additions to clinical buildings. In our study, HM managers disagreed significantly with IC staff on obtaining IC permits before maintenance work. HM staff that fail to liaise and establish clear communication with IC staff are more likely to perform poorly in IC. Communication between IC and HM staff (in-house and contracted) is central to good IC practices. One underdeveloped HM CSF is customer satisfaction. Customer here refers to anyone (patients, doctors, nurses, etc.) in a healthcare establishment. The only healthcare facilities management (HFM) services that is well known to healthcare users is cleaning (May and Clark, 2009). In the past, because of poor IC, cleaning services faced strong public scrutiny and criticisms. As a result, not enough attention was directed at other HFM services; i.e., maintenance performance in IC. Therefore, to raise standards and give HM staff credence, NHS managers must take maintenance-associated HAIs seriously. It is also important that they raise HM staff profile among healthcare users, who should
Key performance measures 705
Table XI. Round three performance measures with consensus in only one Delphi group
1. Involve the HMU and IC department in the purchase of maintenance materials and products 3.2000 No 3.2857 Yes 2. Use a computer-based maintenance system (i.e. reliability-centred maintenance) to coordinate all maintenance work 3.8333 Yes 3.1111 No 3. Conduct daily check of all critical maintenance systems posing the risk of HAIs 3.5000 Yes 3.1111 No 4. Wash and sanitize drainage equipment after use 3.5000 Yes 3.0000 No 5. Provide temporal hand washing facilities for maintenance staff working in high risk patient areas 3.3333 Yes 3.1111 No 6. Use a recognized risk assessment tool (i.e. infection control risk assessment – ICRA) to match the level of risk associated with maintenance work 3.0000 No 3.3333 Yes 7. Take into account changes in assets and legislation when renewing contracts 3.8333 Yes 3.2222 No 8. Maintenance staff team briefings and appraisal schemes in infection control 3.5000 Yes 3.1250 No 9. Equal access, and improve working lives for staff 3.3333 Yes 2.7778 No 10. Measure the number of maintenance product that do not conform to the request 3.3333 Yes 3.0000 No Consensus 8 2 Retained 2 0 Notes: HMM, healthcare maintenance manager; ICM, infection control member. Results are statistically significant at
Performance measures
706 0.006* 0.029* 0.400 0.486 0.943 0.840 0.524 0.486 0.724
0.007* 0.221 0.255 0.569 0.286 0.025* 0.270 0.107 0.224
0.05 level
– 0.926
No No No Yes No No
−0.657 −0.815 −0.254 −0.379 −0.5
Yes No No
No
0.083
0.001 0.192 −0.145
–
Mann-Whitney U-test HMM ICM Round Round Mean Cons. Mean Cons. (R) 3 (R) 2 R3 – R2 Retained
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
IJHCQA 28,7
appreciate HM staff contribution to safe patient care. Continuous improvement could also be ensured by putting measure in place to listen to customer views. By giving an ear to customers, HM staff will be in a better position to identify IC areas for further improvement. References Allen, D. (1993), “What is building maintenance?”, Facilities, Vol. 11 No. 3, pp. 7-12. American Thoracic Society (2012), “Aspergillosis fungal disease”, American Journal of Respiratory and Critical Care Medicine, Vol. 186 No. 4, pp. 1-2.
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
Balaras, C.A., Dascalaki, E. and Gaglia, A. (2007), “HVAC and indoor thermal conditions in hospital operating rooms”, Energy and Buildings, Vol. 39 No. 4, pp. 454-470. Bartley, J.M. (2000), “APIC state-of-the-art report: the role of infection control during construction in health care facilities”, American Journal of Infection Control, Vol. 28 No. 2, pp. 156-169. Boote, J., Barber, R. and Cooper, C. (2006), “Principles and indicators of successful consumer involvement in NHS research: results of a delphi study and subgroup analysis”, Health Policy, Vol. 75 No. 3, pp. 280-297. Burrill, G. (2008), “Healthcare construction under way”, available at: www.touchbriefings.com/ pdf/3202/burrill.pdf (accessed 27 September 2012). Cheng, S.M. and Streifel, A. (2001), “Infection control considerations during construction activities: land excavation and demolition”, American Journal of Infection Control, Vol. 29 No. 5, pp. 321-328. European Centre for Disease Prevention and Control (2013), “Point prevalence survey of healthcare-associated infections and antimicrobial use in European acute care hospitals 2011–2012”, Surveillance Report No. 2011–2012, ECDC, Stockholm. Gupta, U.G. and Clarke, R.E. (1996), “Theory and applications of the Delphi technique: a bibliography (1975-1994)”, Technological Forecasting and Social Change, Vol. 53 No. 2, pp. 185-211. Health Canada (2001), “Construction-related nosocomial infections in patients in health care facilities. Decreasing the risk of Aspergillus legionella and other infections communicable Disease Report volume 27S2, Ottawa. Hoffman, P.N., Bennett, A.M. and Scott, G.M. (1999), “Controlling airborne infections”, Journal of Hospital Infection, Vol. 43 No. 1, pp. 203-210. Joseph, A. (2006), “The impact of environment on infections in healthcare facilities”, available at: www.healthdesign.org/sites/default/files/Impact%20of%20the%20Environment%20on% 20Infections%20in%20HC%20Facilities_0.pdf (accessed 22 October 2012). Kidd, F., Buttner, C. and Kressel, A.B. (2007), “Construction: a model program for infection control compliance”, American Journal of Infection Control, Vol. 35 No. 5, pp. 347-350. McDonald, J. (2009), Handbook of Biological Statistics, 3rd ed., Sparky House Publishing, Baltimore, MD. May, D. and Clark, L. (2009), “Achieving patient focussed maintenance services/systems”, Journal of Facilities Management, Vol. 7 No. 2, pp. 128-141. National Audit Office (2004), Improving Patient Care by Reducing the Risk of Hospital Acquired Infection (No. 876 2003-2004), NAO, London. National Audit Office (2009), Reducing Healthcare Associated Infections in Hospitals in England, NAO, London.
Key performance measures 707
IJHCQA 28,7
Downloaded by Florida Atlantic University At 05:07 31 January 2016 (PT)
708
Riley, D., Freihaut, J., Bahnfleth, W. and Karapatyan, Z. (2004), “Indoor air quality management and infection control in healthcare facilities”, available at: www.engr.psu.edu/iec/ publications/papers/indoor_air_quality.pdf (accessed 1 July 2012). Royal Liverpool Children’s NHS Trust (2004), Environmental Control/Aspergillus Policy (No. C32), Royal Liverpool Children’s NHS Trust, Liverpool. Skulmoski, G.J., Hartman, F.T. and Krahn, J. (2007), “The Delphi method for graduate research”, Journal of Information Technology Education, Vol. 6 No. 1, pp. 1-21. Streifel, A.J. (2003), “Hospital accreditation for airborne infection control: how to prepare for environment-of-care inspection for infection control by health-care accrediting organizations”, Heating/Piping/Air Conditioning Engineering, Vol. 75 No. 3, p. 46. Streifel, A. and Hendrickson, C. (2002), “Minimizing the threat of infection from constructioninduced air pollution in healthcare facilities related to construction”, available at: www. industrialairsolutions.com/contamination-control/hospital-air-purifiers-pdf/HPACConstruction-maintenance-health%20care-facilities.pdf (accessed 22 November 2012). Tabbara, K.F. and Jabarti, A.A. (1998), “Hospital construction-associated outbreak of ocular aspergillosis after cataract surgery”, Ophthalmology, Vol. 105 No. 3, pp. 522-526. Thompson, G.R. and Patterson, M. (2008), “Pulmonary aspergillosis”, Seminars in Respiratory and Critical Care Medicine, Vol. 28 No. 2, pp. 103-110.
Corresponding author Dr Stanley Njuangang can be contacted at: [email protected]
For instructions on how to order reprints of this article, please visit our website: www.emeraldgrouppublishing.com/licensing/reprints.htm Or contact us for further details: [email protected]
