[Previous] [Contents] [Next]

NEEDLESTICK INJURIES: SHARPEN YOUR AWARENESS

Annex 3
SAFER DEVICES COST BENEFIT ASSESSMENT TEMPLATE

INTRODUCTION

This annex provides an illustrative template for assessing the costs and benefits associated with introducing safer devices. The template is designed to allow individual units, by utilising information available locally, now or in the future, to undertake a cost benefit analysis. It has not been possible to perform a cost benefit analysis at a national level because of a lack of information in several key areas, namely the number of needles used on patients, the cost of safer devices and the effectiveness of safer devices. However this cost benefit assessment template has been created utilising the expertise and information of the members of the working group -expertise which may not be available at a local level. It is therefore an attempt to bridge the gap between information available nationally and locally.

Advice is also given on the method of estimating the local costs and benefits of introducing safer devices.

The aim has been to include all effects (financial and human) and the results should therefore be viewed as indicative of the total benefit of introducing safer devices. However given the degree of uncertainty associated with many of the parameters any results obtained from using the template should not be seen as definitive. The rest of the annex is laid out in five sections:

• Section One describes the resource implications of needlestick injuries.
• Section Two lists all the assumptions that have been made.
• Section Three describes the methodology behind the model.
• Section Four gives the parameter values and provides notes on how to vary parameters to reflect local conditions. An example benefit calculation is provided at the end.
• Section Five briefly discusses the major conceptual problems addressed in the model.

1. THE RESOURCES IMPLICATIONS OF NEEDLESTICK INJURIES

The damage or cost caused by needlestick injuries can be segregated into 5 broad cost bands. These bands can be represented by a triangle with the number of injuries decreasing but the costs per injury increasing as one rises up the triangle. Each band is explained below:

Band A represents the needlestick injuries that result in the transfer of a blood- borne disease. These are rare, less than one every two years. However, for such cases the financial and human costs are high and are estimated to range from approximately £10,000 to £620,000.

Band B represents those needlestick injuries that fall into the SCIEH reportable class (i.e. where the source patients are known to be HIV or HCV positive or who are high risk) but which do not result in a seroconversion. Figures from SCIEH indicate that the expected number of such needlesticks will be 24 a year. These injuries involve a large amount of staff time and cause considerable distress. Costs (human and financial) are estimated to range from £3,000 to £5,000 per case.

Band C is for downstream injuries where the source cannot be traced. It is estimated that there are around 240 cases a year. Such cases, while usually being low risk, can cause considerable distress for the healthcare worker concerned -and have resource consequences for occupational health and laboratory services. The costs of such cases are dominated by the distress caused and are estimated to be between £1,000-2,000 per case.

Band D represents the needlestick injuries that are low risk but are reported to A & E or occupational health. It is assumed that the source blood can be tracked but cases such as these still involve occupational health and laboratory services, and possibly counselling for the individual. From Trust returns it is estimated that there are approximately 2,150 of these incidents a year. The cost of such work is thought to be in the range of £50-£100 per case.

Band E contains those needlestick injuries that are rarely reported or go unreported. Clearly it is difficult to estimate the numbers of such injuries although studies have estimated the proportion of needlestick injuries that are reported ranges from 30-75%. The resource implications of these injuries are due to distraction of the injured healthcare worker and the time required to treat the injury. The cost of such injuries are thought to be up to £10 per case. Due to the small size of these costs (even when multiplied by the number of injuries) they have not been included in the template calculation.

Compensation payments are not explicitly included in the template. This is because the template looks at the costs and benefits to society as a whole. If a financial compensation payment is made to a health worker for an injury the compensation payment is a cost to the NHS and a benefit to the healthcare worker. In economic terms these cancel each other out. The real cost to society is of course the damage to the worker's health, which is valued and is included in the framework.

2. ASSUMPTIONS

The lack of data in this field meant that several untested assumptions had to be made. These assumptions are:

• There is no under-reporting of needlesticks (other than 'Band E' injuries).
• There is no interaction between the introduction of safer needles and the needlestick injury rate - i.e. staff will not take more care because they have to take more precautions or less care because they think it is safer.
• There is no interaction between the likelihood of incurring a needlestick injury and the probable prevalence of HIV & HCV in the patient.
• The recommendations of the Short Life Working Group and any action taken as a result will have no impact on the number of needlestick injuries. While it is recognised that as awareness is raised, reporting of needlestick injuries will probably increase, this cannot be quantified and has not been factored into the template.
• Safer devices will not take more time to use and have no impact on patient care.
• The impact of a needlestick injury is only on the person injured -i.e. other staff members are not affected.
• Lost hours multiplied by the average wage rate is a good proxy for lost output.

3. METHODOLOGY

The benefits from introducing safer devices arise from avoiding the costs associated with needlestick injuries. The template therefore seeks to calculate the cost of needlestick injuries and then derives a benefit as based on the percentage of needlestick injuries (and hence the cost) that would be avoided if safer devices were used.

The total cost of needlestick injuries is made up of four components; the estimated cost from HCV infection, the estimated cost from HIV infection, the lost output and treatment cost of needlestick injuries and the pain and suffering incurred when infection is unknown.

• The estimated cost from HCV infection per needlestick injury is calculated as below and then multiplied by the number of needlestick injuries per year to give the cost from HCV infection per year:
  • The known prevalence of HCV multiplied by the sero-conversion rate from a single percutaneous exposure gives the risk of infection of HCV from a needlestick injury. The seroconversion rate used is 3%.
  • The 'human costs' of HCV are calculated by multiplying the value of losing a year of one's life by the loss of QALYs associated with HCV infection. The value of losing a life year has been calculated as £30,500¹ and a recent SNAP report estimated the average loss of QALYs across all who contract HCV (i.e. including those who do not develop a serious form of the disease).
  • The human cost is then added to the costs of treating HCV (£2,019 for the average 30-year-old infected in the SNAP report).
  • Multiplying the total cost of one HCV infection by the risk of infection of HCV from a needlestick injury gives a theoretical cost per needlestick injury.
  • Should an effective treatment become available for HCV the costs will have to be factored into the calculation as below for HIV.
• The theoretical cost from HIV infection is calculated in the same way with the addition of Post-Exposure Prophylaxis (PEP) as a complicating factor:
  • It is assumed that PEP is not given unless the case is known to be high risk. Where PEP is given it is assumed to be 80% effective (i.e. it reduces the sero-conversion rate by 80%) and costs £830 per patient.
  • The seroconversion rate used is 0.31%. The value of losing a life year is, of course, the same at £30,500. The loss of QALYs has been estimated at 17 for those for whom PEP is not effective
  • The cost of treatment has been calculated at £100,608 based on treatment costs of £12,000 a year over a life expectancy of 12 years, discounted at 6%^2.
• The lost output is calculated by multiplying the average staff cost per hour (£10.84) by the average working hours lost (2.4) per needlestick injury (supplied by one trust) to give a cost of lost time (as a proxy for lost output). To this is added the average treatment costs of £55 (supplied by one trust). For those cases involving a needlestick injury from a known HIV or HCV positive source the lost output and treatment costs are estimated to be £1,000.
• The pain and suffering incurred when infection is unknown is calculated by multiplying the number of downstream injuries by £1,000 and the number from known sources that are either HIV or HCV positive by £2,000. Downstream needlesticks were reported by trusts as being approximately 10% of all needlesticks.

It should be noted that three factors have not been quantified. These are:

• The value of the lost output that occurs when a healthcare worker contracts HCV or HIV over and above the lost output that normally occurs due to declining health. This additional lost output is due to the working restrictions placed on healthcare workers who are HIV positive or HCV positive.
• The cost of replacing staff (i.e. recruitment and training costs) that are unable to work due to contracting a blood borne virus from a needlestick injury.
• The worry caused to other healthcare workers when some-one else incurs a needlestick -i.e. the anxiety of healthcare workers that they may receive a needlestick injury.

4. UTILISING THE TEMPLATE

The table below contains all the parameters within the model, the values of those parameters and whether those parameters are dependent on local conditions. Notes on how to estimate those parameters that are dependent on local conditions are provided below.

Parameters were also broadly categorised as to whether or not they were sensitive to local conditions. These are those marked 'Yes' in the table.

Other parameters are marked 'No' because they were based on the best information available to the group at the time of analysis and were unlikely to be readily available at a local level. If it were felt locally that better information on these parameters exists either now or in the future then it would, of course, be appropriate for Trusts/Health Boards to adjust these parameters.

The baseline year for all figures is 1999/00. The only exception is the average cost of lost time which is based on 1998/99 data as no HCHS inflation figure was, at the time of writing, available for 1999/00. The average cost of lost time should therefore be modified by this figure to reflect pay increases in 1999/00.

Notes on Estimation of Parameters to be set locally

The number of needlestick injuries per annum should be estimated from local trust data. Ideally the number of needlestick injuries per annum should relate only to those needles that are to be replaced by safer devices. If this information is not available then the number could be derived from pro-rating from the total number of needlesticks a year and the number of needles used a year (an adjustment must be made for the proportion of needles that are not used on patients)

Expected Reduction in Cases: The effectiveness of the safer needle as established by testing- i.e. the proportion of needlestick injuries that would be avoided by introducing safer devices. A figure of 60% has been used as this represents an approximate mid-point of evaluations performed in the United States. These evaluations have found effectiveness levels ranging from 29% to 89%. It should be noted that this is a critical figure in the template and must be updated before any realistic estimates can be obtained.

Calculating Cost

To complete the cost benefit analysis the cost of using safer devices must be calculated. This cost is the additional cost of safer devices plus any other additional costs such as stocking and/or training cost. To calculate the additional cost of using safer devices one multiplies the additional cost per device by the number of devices per year that are used on patients.

A stocking cost may arise where the number of safer devices used per year is small due to devices being held for some time without being used. Hence the cost occurs before the benefit is realised and this should be recognised. Any training costs should be spread over the period for which the training will be beneficial.

5. CONCEPTUAL POINTS

Costs relating to Hepatitis B have not been modelled, despite it having a higher seroconversion rate than Hepatitis C and HIV. This is because the vast majority of clinical health workers are immunised against Hepatitis B. Those who have not been immunised or who have not responded to immunisation can be offered immunoglobulin as protection after the needlestick. Costs relating to possible Hepatitis B infection would therefore be small and due to their insignificance have not been included. There are approximately eighteen other pathogens which can be transmitted from infected patients via needles. Costs relating to these pathogens are estimated to be small and hence these costs have also not been included.

There is currently some debate over the HCV seroconversion rate from a percutaneous needlestick injury. Early research estimated the sero-conversion rate to be approximately 3%. More recent research has provided evidence that the rate may be substantially lower. It was felt prudent to use the higher seroconversion rate until the debate is resolved

Example-Benefit assessment Using safer devices just on 'at risk' groups

The at risk category considered to which the following parameters relate is current injecting drug users (IDUs).

1. The number of needlestick injuries involving current IDUs was estimated using the prevalence of HIV and HCV in current IDUs (approximately 40% and 50% respectively - figures from the Drugs Misuse Statistics Scotland 1999 and Scottish Health Statistics 1999).
2. Prevalence of HCV among current IDUs estimated at 62% from figures supplied by SCIEH.
3. Prevalence of HIV among current IDUs estimated at 5% from figures supplied by SCIEH.
4. It is assumed that no-one catches both HIV and HCV from the same needlestick injury.
5. It is assumed that 'at risk' groups are classed as high risk and PEP is given.

Example Benefit Calculation - Introducing Safer Devices Just on Known IDUs (Scotland)

[Previous] [Contents] [Next]