PET-CT scanners were well utilised

Of the hospitals included in the audit:

• Liverpool, Royal Prince Alfred and Westmead Hospitals utilised their scanners at over 85 per cent of potential capacity
• Nepean Hospital had a moderate utilisation of 61 per cent, due to older equipment and insufficient ‘uptake’ rooms for patients to receive radioactive injections
• Lismore Base Hospital’s utilisation was only 32 per cent, due to lower demand than the metropolitan hospitals, which in turn meant that it only provided PET-CT services for three days per week.

Exhibit 2 shows our calculation of PET-CT utilisation based on data provided by each hospital. 

There was limited data on patient wait times for PET-CT scans

Clinicians we interviewed during the audit advised that there was generally no delay in patients accessing PET-CT scanners. However, most hospitals did not collect patient wait time data to confirm this view.

Only Royal Prince Alfred Hospital collected data on the time patients waited for PET-CT scans, which is the difference between the date of scan that referring doctors requested and the actual date scanning takes place. This allowed the hospital to monitor its performance. The data show that in 2015–16, 54 per cent of scans at Royal Prince Alfred Hospital were undertaken on the date requested by the referring doctor, while 99 per cent of scans were undertaken within five days.  

The consistent use of patient wait time data across hospitals would offer greater assurance that services are being provided in a timely way, as well as allow hospitals to benchmark their performance against like services.

Most scans were reported promptly and in line with clinical standards

Clinical standards (issued by the Royal Australian and New Zealand College of Radiologists) require that completed scans should generally be reported to referring doctors within 24 hours of the scan being completed. Exhibit 3 shows that across the hospitals we examined, 90 per cent of results were reported within 24 hours of the scan being taken. Only Royal Prince Alfred Hospital achieved 100 per cent completion of scan reports within 24 hours of the scan being taken.

There was good oversight of PET-CT scanner maintenance and repair

Hospitals we examined had service contracts with equipment manufacturers or distributors for routine servicing and repair of scanners. Clinical staff responsible for administering PET-CT scans were also directly responsible for overseeing service contractors’ work.  

Based on service reports we inspected, there was good oversight of equipment maintenance and repair. There were about five days each year during which each PET-CT scanner was not used due to routine maintenance.  

Hospitals monitored radiation exposure consistent with safety standards

All hospitals operating PET-CT scanners must meet Australian Department of Health accreditation standards, including those relating to radiation safety.

On our hospital site inspections, we observed that hospitals placed great emphasis on limiting radiation exposure for staff and patients around PET-CT services. Practices we observed included:

• personal radiation monitoring equipment, and mechanisms to review and limit exposure, such as rostering of staff
• services having dedicated radiation limiting dosage rooms. Some services also had capacity to remotely administer radioactive injections, special purpose waiting bays, and CCTV monitoring of patients during examination
• post-scan arrangements to limit public exposure to radiation, including extended waiting times post-scan for patients who intended to catch public transport.

The funding of health services in NSW public hospitals involves a complex arrangement between the Australian and NSW Governments. In practice, this means:

• Services delivered to patients who present at Emergency Departments and / or are admitted as public patients are funded by NSW Health. For some patients who are admitted but consented to being admitted as private patients, service costs are partly recoverable from their Private Health Insurance.
• Services delivered to patients who are privately referred to a public hospital’s outpatient clinic (where patients attend for diagnosis or treatment, but are not admitted to the hospital) are funded by Medicare, and if applicable, other sources such as Workers Compensation or patient out-of-pocket charges.

Exhibit 4 shows the number of PET-CT scans provided by hospitals we examined over three financial years for different patient types. PET-CT scans that were provided in outpatient clinics and were not eligible for Medicare rebate accounted for 25 per cent of total scans in 2013–14, and had increased to 30 per cent of total scans in 2015–16. The costs for a small number of these scans were recoverable from other sources such as Workers Compensation, while for the majority the costs were met by a range of sources, including hospitals’ general purpose funds and patient out-of-pocket charges. Across the five hospitals, out-of-pocket charges varied and ranged from $250 to $950 per scan.

Local Health Districts are responsible for planning local PET-CT services

In New South Wales, responsibility for delivering health services has been devolved from the Ministry of Health to Local Health Districts. This approach is intended to ensure that local decision-making is responsive to local community needs.  

However, Local Health Districts do not have unfettered discretion in how they plan their local delivery of PET-CT services. To be eligible for funding under Medicare, the Australian Government Department of Health requires that PET-CT services must comply with a range of requirements, including that the service must be performed:

• by an appropriately qualified person
• in a comprehensive facility that can provide a full range of diagnostic imaging services and cancer treatment services
• using equipment that meets certain requirements
• only following a referral from a recognised physician or specialist consultant.

Local Health Districts are also required to comply with NSW Health’s Guide to the Role Delineation of Clinical Services (‘the Guide’). The Guide describes the minimum support services, workforce and other requirements that must be provided to ensure the safe delivery of clinical services. The Guide describes seven types of clinical services that require access to PET-CT scanners.

These requirements of the Australian Government and NSW Health set the broad parameters within which Local Health District’s plan and provide PET-CT services.  

Demand for PET-CT scans grew from 2013–14 to 2015–16

Ensuring adequate supply of PET-CT services requires accurate measurement of current and future demand.

From 2013–14 to 2015–16, the demand for PET-CT scans increased in each of the hospitals we visited. As shown in Exhibit 5, the increase in the number of scans performed was:

• largest at Nepean Hospital (increasing 47 per over three years) 
• followed by Lismore Base Hospital (29 per cent), Westmead Hospital (25 per cent) and Royal Prince Alfred Hospital (16 per cent)
• smallest at Liverpool Hospital (10 per cent).

Demand for PET-CT scans is likely to continue to increase

Across the five hospitals we examined, 73 per cent of PET-CT scans were provided to patients aged 55 years and older. Almost one in three scans were provided to patients aged between 65 and 74 years. 

The main recipients of PET-CT scans were cancer patients, though the technology is also increasingly used for Alzheimer’s disease and in cardiology. The prevalence of cancer, Alzheimer’s disease and heart disease all increase with age. 

As the population of New South Wales ages, demand for PET-CT services is likely to increase. Local Health Districts will need to plan to meet this increased demand.

Performance reporting could inform planning and benchmark service delivery

In addition to information about current and future demand, planning needs to take into account how well existing performance meets service expectations.  

Performance reporting is frequently used in the health system to inform system planning, as well as to improve service delivery. It provides a mechanism to evaluate the quality and efficiency of services. This can be done through aggregate state-level data on system performance, or at the level of individual Local Health Districts or hospitals.

Performance measurement also allows benchmarking between comparable hospitals. Different outcomes in performance can highlight good practice that can then be shared between services.

Our audit found no standardised performance reporting on PET-CT services. There was some sharing of good practice and experiences between Local Health Districts, though this was ad hoc, informal, and highly dependent on personal relationships between staff.

Some examples of performance measures that might be useful include rates of equipment utilisation, cost per scan, staffing per machine, and equipment age. Clinical outcomes data and measures of patient experience could also contribute to this reporting regime. NSW Health could play an enabling role by facilitating more routine and standardised performance reporting across PET-CT services. This would be consistent with the role of the Secretary of NSW Health under the Health Administration Act 1982 to plan the provision of comprehensive, balanced and co-ordinated health services throughout New South Wales.   

NSW Health was effective at combining its purchasing power by mandating that Local Health Districts purchase new PET-CT scanners through NSW Health’s state contracts with equipment manufacturers. Before a purchase were made, Local Health Districts were required to submit project documentation to the Ministry of Health for approval. This documentation included value-for-money assessments, such as:

• the anticipated benefits and service need from purchasing equipment
• a short list of different options for operating the equipment 
• analysis of different options and the preferred options

Generally, there is a community expectation that once a type of health service is established, it will be maintained. Equipment replacement plans that set out when PET-CT equipment will be replaced, whether the replacement scanner should be leased, purchased or shared, and possible funding sources, would assist in promoting value for money in the continuity of service.  

We found planning for the future replacement of PET-CT scanners was inconsistent and uncertain. For example, we found that Local Health Districts did limited planning to identify funding for the replacement of their PET-CT scanners.  

Some Local Health Districts could clearly articulate how equipment replacement planning fitted within their local governance and budget processes. However, other Local Health Districts were far less clear, conveying a sense that equipment replacement was a matter that can be deferred to the future, at which time a solution might be found through political will, community support, and other ad hoc mechanisms.  

Some clinicians were concerned by this lack of certainty around how crucial clinical equipment would be replaced. Clear planning for equipment replacement would offer greater assurance of future continuity of service in a value-for-money way.

We examined the maintenance histories of 50 items of biomedical equipment sampled from NSW Health’s fixed asset register and hospitals’ information systems. Ten were selected from each hospital. This sample was intended to be illustrative, rather than statistically significant.

While this section focuses on the scheduling and timeliness of testing and maintenance, it is important to acknowledge that there were a range of other factors that contributed to the management of biomedical equipment. These included incident management, acceptance testing, user training, professional training of biomedical engineers, systems and processes for alerts and notification, as well as biomedical governance.  

Hospitals adopted the Australian/New Zealand Standard on biomedical equipment management

All hospitals had committed to comply with the Australian/New Zealand Standard 3551 ‘Management programs for medical equipment’7 (the Standard). The Standard states:

…throughout the lifetime of medical equipment in clinical use, there is a need for regular assessment and testing of the medical equipment to ensure it is safe, and continues to be safe, for its intended clinical application.  

In addition to contributing to the safety and quality of healthcare, regular testing and maintenance also assists in minimising the whole-of-life costs associated with equipment, by prolonging the life of equipment and reducing repair costs.

Intervals for testing and maintenance were determined under the Standard

The Standard requires that, at the time of accepting biomedical medical equipment, hospitals should establish how regularly equipment testing and maintenance will be performed. This process is referred to as determining ‘intervals’ for equipment testing and maintenance.  

The Standard requires hospitals to follow the intervals recommended by manufacturers. Where hospitals decide to deviate from a manufacturer’s recommendation, the determined intervals must be supported by a rigorous and documented risk management process that is regularly reviewed.

The hospitals in this audit followed manufacturers' recommendations for equipment testing and maintenance. In most cases, this meant that hospitals committed to perform this work at 12-month intervals. For certain higher risk equipment, hospitals applied a six-month interval. For some lower risk equipment, a two-year interval was adopted.

Equipment testing and maintenance did not always meet intervals determined under the Standard

Our review found that only about half of the equipment items in our sample had their testing and maintenance completed in accordance with scheduled intervals or within 30 days of it being due. This is detailed in Exhibit 6 below.  

Only two hospitals used risk rating to oversee equipment maintenance

Only two of the five hospitals implemented risk ratings in the management of biomedical equipment, under which equipment was classified according to clinical risk. A third hospital had documentation about risk rating, but there was no evidence that it had been implemented. The two remaining hospitals advised that their existing equipment information systems did not support equipment risk rating.  

In the context of biomedical equipment testing and maintenance, there are two primary uses of equipment risk rating:

• The first was to guide a biomedical engineering department in prioritising testing and maintenance of high risk equipment on a day to day basis. An effective risk rating process assists biomedical engineering departments to determine and track their priorities in a way that minimises risks to patients associated with delayed testing and maintenance.
• The second was to allow a biomedical engineering department to determine testing and maintenance intervals that deviate from equipment manufacturer’s recommendations. The Standard states that hospitals may extend the intervals they apply beyond a manufacturer’s recommendation if the decision is supported by an assessment of risk. Hospitals may decide to extend testing and maintenance intervals for some equipment, to lower costs or allow their biomedical engineering team to focus on higher priority tasks.

The Ministry of Health advised that equipment risk rating would be incorporated into the Asset and Facilities Management Online (AFM Online), a state-wide asset management system that NSW Health had developed and was implementing. The risk rating to be used state-wide will be based on the risk rating used by the Children’s Hospital at Westmead.  

AFM Online is further discussed in section 2.4.

Some hospitals had inadequate oversight of work performed by external contractors

Some biomedical equipment in our sample was managed using service contracts with external contractors (often the manufacturer or distributor) who tested, maintained and repaired the equipment. The Standard recognises that service contracts may be used by hospitals to manage high-risk biomedical equipment and as a way of dealing with limited internal biomedical engineering resources. There is an expectation that service contracts are likely to become more common as biomedical equipment becomes more sophisticated and complex.  

The use of external contractors can be appropriate, provided that the work is subject to oversight by hospitals and detailed records are kept. As one senior executive explained ‘…a service contract can be great, however, you need someone with expertise to oversee it.’

Some hospitals relied on external contractors to keep service histories of their equipment. The Standard requires health services to maintain their own complete service histories regardless of whether equipment was maintained in-house or by external contractors. The failure of hospitals to maintain these records means they may not be aware of the condition of equipment that was serviced by contractors. This may expose patients to clinical risks and poor health outcomes, particularly as service contracts are frequently used to maintain high risk equipment.

Some biomedical engineering department staff reported poor practices by external contractors, including instances in which providers had incorrectly recorded items they had tested, or had refused to provide details of testing and maintenance performed. This reinforces the need for hospitals to actively manage and monitor work performed under service contracts.  

The Standard requires hospitals to keep records of testing and maintenance activities. Without accurate and accessible records, it is difficult for biomedical engineers to

• efficiently schedule and prioritise their work
• self-audit and quality assure their work
• offer assurance that equipment is being tested and maintained to ensure it is safe and suitable for its intended clinical use.

As we found during this audit, without good record keeping, it is also difficult for third-parties to review and assure testing and maintenance processes for biomedical equipment. Despite our sample being limited to just ten items from each hospitals, we found that some hospitals were unable to provide service histories in a timely way. In some cases, it took months for complete information about service histories to be provided.

Some hospitals did not keep complete and accurate service histories

The records provided to us by some hospitals did not clearly and consistently identify the activities performed on items of equipment. For example, several biomedical engineers told us they assumed testing and maintenance had been performed in relation to a number of ‘repair’ database entries, even though there was no mention of these activities in the database. Testing and maintenance activities were often recorded inconsistently between equipment items and even within the records for the same equipment item

Biomedical equipment information systems were inefficient and outdated

At Nepean and Westmead Hospitals, records were maintained using a combination of a paper based filing system and an unsupported information system developed in the early 1990s. Lismore Base Hospital used an unsupported version of the same system, also in combination with a paper based filing system.  

At Liverpool Hospital, maintenance records for several items were not tracked on a biomedical engineering information system at all, with service records consisting only of scanned service reports saved to a network drive. Royal Prince Alfred Hospital used the same system as Liverpool Hospital, although it had developed and customised the system to allow for tracking and reporting on testing and maintenance, and electronic entry of testing and maintenance reports.

Exhibit 7 depicts three capability levels of information systems to support biomedical equipment management, and shows the level of capability of information systems in each hospital. None of the hospitals used an information system that provided the full three-levels of capability outlined below.

Centralised procurement of biomedical equipment saved money

All hospitals we examined used HealthShare, NSW Health’s centralised shared services organisation, to purchase in bulk the majority of their biomedical equipment. HealthShare purchases were made based on NSW Health’s ‘state contracts’ with equipment manufacturers.

The use of state contracts resulted in savings from bulk purchases, as all Local Health Districts consolidated their purchasing power to improve their bargaining position with equipment manufacturers. For example, financial data for the purpose of electromedical equipment under state contract 956 showed savings of $1.38 million, or 26 per cent, compared to the maximum price of the equipment before the contract. This contract was one of 88 such state contracts managed by HealthShare.

HealthShare advised that there were further benefits from the use of state contracts beyond monetary savings, such as reduced task duplication, and the consolidation of logistical and storage requirements associated with equipment purchases.

There was good governance of equipment acquisition and disposal

The hospitals we examined had formal processes governing biomedical equipment acquisition and disposal.  

Equipment that needed to be replaced was identified by a hospital’s biomedical engineering department, on the basis of one or more of the following considerations:

• an item’s projected replacement date
• an item’s repair history, which could indicate whether it would be more cost effective to replace than to maintain
• a manufacturer’s end of support notification (stating that the manufacturer will not maintain the item or provide spare parts beyond a certain date)
• a notification from a nursing unit manager, subsequently affirmed by a technician, that the equipment required replacement.

Hospitals’ processes for approving the replacement of equipment were tailored and proportionate to the value of the equipment. Replacement of low value equipment (valued at under $10,000) was generally processed by the biomedical engineering department. A procurement and asset management committee was responsible for approving the replacement of medium-value equipment (valued between $10,000 and $250,000 in most hospitals we examined), while high value equipment (over $250,000 in most hospitals we examined) replacement requests were escalated to a more senior committee with broader planning responsibilities.

If equipment was new to a hospital, a procurement or planning committee must approve the purchase. In deciding whether to give approval, the committee considered whether the equipment would generate sufficient revenue to justify the purchase, or whether there was a clinical, research or other justification. A trial process was followed before a vendor was selected, with equipment being tested on-site and results logged on a state-wide database. At the hospitals we examined, equipment disposal processes were aligned with Australian Standards, NSW Health’s Goods and Services Procurement Policy, and the NSW Environment Protection Authority’s hazardous equipment guidelines

Inefficient data system limited hospitals’ ability to plan for equipment replacement

Due to the large number of biomedical equipment items, planning for such equipment is considered as part of each hospital’s annual asset strategy.

There was a large amount of data available on the service history of biomedical equipment, but this data could not be efficiently queried and used for planning purposes. Given that planning for biomedical equipment replacement was partly based on service history, it was difficult for hospitals to identify equipment that should be replaced rather than repaired.  

Identifying this equipment requires hospitals to have systems and processes in place that identify equipment with a high rate of failure. These systems and processes did not exist in the hospitals we examined. A more efficient data system and automated reporting will enable improved planning.