Baselines
On average, two patients fell and broke their hip in New Zealand’s hospitals every week in 2012. This typically added an estimated month to their hospital stay, and cost a minimum of $2.6 million. The table below shows the figures.
| 2012 | |
| Falls resulting in a fractured neck of femur | 95 |
| Cost of fractured neck of femur ($m) | $2.6 million |
| Additional bed days associated with a fractured neck of femur | 3385 |
It is important to remember that this is only a fraction of the total harm associated with falls in hospitals, and all harm in hospitals is only a fraction of the total harm associated with falling in and other care settings. Falls that lead to a broken hip occur consistently throughout the year.
There is no single solution to reducing falls that harm patients. The reducing harm from falls programme promotes a range of interventions that can help. An essential first step is to identify the risks of falling for an individual patient, and act accordingly. For this reason, risk assessment is the principle process for improvement that the QSMs measure.
Between December 2012 and February 2013 all district health boards (DHBs) undertook an audit of patients aged 75 and over (the group most at risk of falls that harm) to see how many had received a falls risk assessment. We would expect an assessment to happen nearly all the time, so the threshold for expected performance is set at 90 percent. The results form a baseline of performance (see following table).
This table shows the result of the baseline audit for the risk assessments provided for patients aged 75 and over (or 55 and over if Māori or Pacific peoples). Green indicates 90 percent or more of those audited received a falls risk assessment; grey indicates 75–90 percent received an assessment; progressively deeper red identifies lower proportions.
As the table shows, DHBs are starting from different places. For example, at Counties Manukau, Canterbury and Waikato DHBs, nearly all patients audited had received a falls risk assessment, while at Bay of Plenty DHB less than half of audited patients had. Only 5 out of 20 DHBs reached the threshold of expected performance at the start of the Falls programme.
Between October 2011 and April 2013, the Commission worked with Ko Awatea on a national project to reduce rates of central line associated bacteraemia (CLAB) in intensive care units (ICUs).
The project sought to reduce CLABs in ICUs from a baseline of 3.32 per 1000 line days to less than one per 1000 line days, and measured compliance with central venous line (CVL) insertion and maintenance bundles. The former of these is reported below as a process marker.
The table below shows the general pattern of increased compliance with the use of insertion bundles across the four quarters in 2012.
The process of measuring infection and collecting and understanding the information has greatly improved since the start of the project, where few ICUs were measuring CLAB rates, and only two ICUs had implemented insertion and maintenance bundles.
Green indicates the DHB used the insertion bundle to reduce the risk of CLAB 90 percent of the time in the quarter under review. White indicates this occurred 70–90 percent of the time. Deepening shades of red indicate lower percentages. Data are courtesy of the Target CLAB Zero collaborative and have been aggregated from monthly to quarterly for ease of use. Data apply to insertion of lines in ICUs/high dependency units (HDUs). Data are unavailable for Wairarapa and South Canterbury DHBs. In both cases, insertion of central lines rarely take place in the HDU, hence the lack of data for this measure.
This pattern of progress is shown clearly in the graph below. The blue lines show insertion compliance rates are in general trending up and have in most cases passed through the orange lines, which show a compliance threshold of 90 percent.
*Although adopted as a QSM, this measure originated from the national CLAB project, which predated the development of the QSMs. The project did not use a process target or threshold as part of its work. The ‘threshold’ was first set in December 2012 as part of the finalised QSMs, following agreement with the clinical lead for the CLAB project. Hence prior to January 2013, the threshold should be taken as notional.
Auckland ICU department names:
- CVICU – Cardiovascular Intensive Care Unit
- DCCM - Department of Critical Care Medicine at Auckland City Hospital
- PICU - Paediatric Intensive Care Unit at Starship
This is calculated based on the actual recorded CLABs per 1000 line days (which averaged under 1.0 per 1000 line days) compared with the project’s estimated baseline of 3.3 per 1000 line days, and an estimated cost of $20,000 per CLAB.
The reduction in CLAB rates from 3.32 to under 1.0 per 1000 line days produces substantial effects. Between April 2012 and March 2013, there were in excess of 31,000 line days in ICUs in New Zealand, which means that just over 100 CLABs would be expected if the baseline rate of 3.32 had not changed. In fact, there were 15. Calculated at a cost of $20,000 per CLAB, this represents avoided cost of nearly $2 million.
Hand hygiene is one of the most important measures in the fight against healthcare associated infections, making it a key patient safety issue within the health sector. International evidence is clear that improved hand hygiene practices help reduce healthcare associated infections, including antibiotic-resistant infections within hospitals.
The Commission has entered into partnership with Auckland DHB to implement a national hand hygiene work programme. For information on the programme see the Hand Hygiene New Zealand website.
The WHO approach (‘Five Moments for Hand Hygiene’) being implemented as part of the Hand Hygiene New Zealand programme identifies the following moments as critical to the prevention and control of infections:
- before patient contact
- before a procedure
- after a procedure or body fluid exposure risk
- after patient contact
- after contact with patient surroundings.
Since 2012 all DHBs have undertaken observational audits of compliance with the five moments. These have taken place on a select number of wards and departments inside DHBs. The location of these will vary between DHBs, so some care must be taken in making comparisons. Data for the first two time periods (which end in March 2013) are shown in the following table. The percentage of observed hand hygiene moments that complied with the WHO approach varied from 54 to 75 percent between DHBs. There was also a general increase in compliance in the second period. The number of DHBs where more than 70 percent of observed moments were compliant increased from five to eight between the two time periods.
Green indicates DHB staff were observed to comply with WHO hand hygiene practice at least 70 percent of the time. Grey indicates this occurring 60–70 percent of the time. Deepening shades of red indicate lower percentages. Data are courtesy of Hand Hygiene New Zealand. Further details of the audit can be found on the Hand Hygiene New Zealand website. Note, quarter dates refer to the quarter in which the audit period finished. Thus Q4 2012 ran from July to October 2012 and Q1 2013 from November 2012 to March 2013.
There is clear geographic variation in these data, as the map below shows. The Northern region stands out as having relatively high rates of compliance, while rates for South Island DHBs are considerably lower. It is not immediately clear why this difference should be so marked. The high rates in the north may reflect Auckland DHB’s leadership on this issue.
Green indicates DHB staff were observed to comply with WHO hand hygiene practice at least 70 percent of the time. Grey indicates this occurred 60–70 percent of the time. Deepening shades of red indicate lower percentages. Data are courtesy of Hand Hygiene New Zealand. Further details of the audit can be found on the Hand Hygiene New Zealand website.
Our outcome measure is the number of patients with healthcare associated Staphylococcus aureus bacteraemia per 1000 patient days. This is the most common healthcare associated infection in New Zealand hospitals. In 2012, reported infections reduced.
For the majority of people who undergo surgery, it is safe, effective and beneficial. However, surgery is not without risks. The burden of harm that follows complications of surgery is large. Numerous complications of surgery could be chosen for the QSMs, but on the advice of experts and the sector, deep vein thrombosis (DVT)/pulmonary embolism (PE) and sepsis have been chosen. DVT/PE and sepsis are believed to be reasonably sizeable areas of complication, more readily identifiable from routine data sources, amenable to improvement and related to the process measure (the use of World Health Organization (WHO) surgical safety checklist). Using the National Minimum Dataset of hospital admissions we can identify when these two complications occur within a surgical admission, and where patients are readmitted into hospital with either condition within 28 days of a surgical admission.
The graph below shows case numbers in 2012, when there were over 750 events for DVT/PE and over 500 for sepsis. We estimate that these patients stayed in hospital for an additional 3500 days, at a cost of nearly $3 million.
Events and readmissions show the total number of cases where DVT/PE or sepsis was recorded during a hospital stay following surgery, or where a patient was readmitted with DVT/PE or sepsis within 28 days of surgery.
Additional bed days is an estimate of the additional days associated with patients over what would have happened if DVT/PE or sepsis had not occurred. Cost is based on an estimate of $770 per additional day. This is likely to be a conservative estimate.
The total of observed deaths in hospital is shown among this group of patients. This does not imply that all died directly from DVT/PE or sepsis, nor that none may have died had these complications not occurred. It is not a calculation of “excess” or “unexpected” deaths.
Our process measures concentrate on the use of the WHO surgical safety checklist, a commonsense approach to ensuring the correct surgical procedures are carried out on the correct patient. It involves checking the right people are present and that they all are in agreement about why they are operating.
The Commission urges hospitals to use the checklist consistently. When implemented properly, the checklist requires hospital staff to stop and think what they are doing and why. Thinking about what could go wrong is also important; for example, checking for allergies to medicines is part of the checklist. More information on the checklist, including a copy of the Australian and New Zealand checklist, is available here.
The checklist comprises three parts: check-in, time out and sign out. Our marker looks at the proportion of operations where all three parts of the checklist are used.
The table (above) show the result of the baseline audit for use of the surgical safety checklist. Green indicates that all three parts of the checklist were used in 90 percent or more of the cases audited; grey indicates use 75–90 percent of the time; progressively deeper red identifies lower proportions.
*Southern DHB is excluded from this list because it is able to show the position for Southland Hospital (where use is recorded as 100 percent) but not for Dunedin Hospital for use of all three parts of the checklist.
At least some variation may be down to differences in recording. The data for this measure was collected in nearly all cases through case note review which, while easier to undertake than observational audit, does not allow any judgment of the effective use of the checklist. It is possible that a checklist is being used effectively, but not recorded as such, or that the checklist is used in an ineffective way, despite being well recorded.
To address this in future, we plan to include data collected through observation of the checklist in practice. This may have the effect of considerably changing the results for this measure.
Falls
On average, two patients fell and broke their hip in New Zealand’s hospitals every week in 2012. This typically added an estimated month to their hospital stay, and cost a minimum of $2.6 million. The table below shows the figures.
| 2012 | |
| Falls resulting in a fractured neck of femur | 95 |
| Cost of fractured neck of femur ($m) | $2.6 million |
| Additional bed days associated with a fractured neck of femur | 3385 |
It is important to remember that this is only a fraction of the total harm associated with falls in hospitals, and all harm in hospitals is only a fraction of the total harm associated with falling in and other care settings. Falls that lead to a broken hip occur consistently throughout the year.
There is no single solution to reducing falls that harm patients. The reducing harm from falls programme promotes a range of interventions that can help. An essential first step is to identify the risks of falling for an individual patient, and act accordingly. For this reason, risk assessment is the principle process for improvement that the QSMs measure.
Between December 2012 and February 2013 all district health boards (DHBs) undertook an audit of patients aged 75 and over (the group most at risk of falls that harm) to see how many had received a falls risk assessment. We would expect an assessment to happen nearly all the time, so the threshold for expected performance is set at 90 percent. The results form a baseline of performance (see following table).
This table shows the result of the baseline audit for the risk assessments provided for patients aged 75 and over (or 55 and over if Māori or Pacific peoples). Green indicates 90 percent or more of those audited received a falls risk assessment; grey indicates 75–90 percent received an assessment; progressively deeper red identifies lower proportions.
As the table shows, DHBs are starting from different places. For example, at Counties Manukau, Canterbury and Waikato DHBs, nearly all patients audited had received a falls risk assessment, while at Bay of Plenty DHB less than half of audited patients had. Only 5 out of 20 DHBs reached the threshold of expected performance at the start of the Falls programme.
Central line associated bacteraemia
Between October 2011 and April 2013, the Commission worked with Ko Awatea on a national project to reduce rates of central line associated bacteraemia (CLAB) in intensive care units (ICUs).
The project sought to reduce CLABs in ICUs from a baseline of 3.32 per 1000 line days to less than one per 1000 line days, and measured compliance with central venous line (CVL) insertion and maintenance bundles. The former of these is reported below as a process marker.
The table below shows the general pattern of increased compliance with the use of insertion bundles across the four quarters in 2012.
The process of measuring infection and collecting and understanding the information has greatly improved since the start of the project, where few ICUs were measuring CLAB rates, and only two ICUs had implemented insertion and maintenance bundles.
Green indicates the DHB used the insertion bundle to reduce the risk of CLAB 90 percent of the time in the quarter under review. White indicates this occurred 70–90 percent of the time. Deepening shades of red indicate lower percentages. Data are courtesy of the Target CLAB Zero collaborative and have been aggregated from monthly to quarterly for ease of use. Data apply to insertion of lines in ICUs/high dependency units (HDUs). Data are unavailable for Wairarapa and South Canterbury DHBs. In both cases, insertion of central lines rarely take place in the HDU, hence the lack of data for this measure.
This pattern of progress is shown clearly in the graph below. The blue lines show insertion compliance rates are in general trending up and have in most cases passed through the orange lines, which show a compliance threshold of 90 percent.
*Although adopted as a QSM, this measure originated from the national CLAB project, which predated the development of the QSMs. The project did not use a process target or threshold as part of its work. The ‘threshold’ was first set in December 2012 as part of the finalised QSMs, following agreement with the clinical lead for the CLAB project. Hence prior to January 2013, the threshold should be taken as notional.
Auckland ICU department names:
- CVICU – Cardiovascular Intensive Care Unit
- DCCM - Department of Critical Care Medicine at Auckland City Hospital
- PICU - Paediatric Intensive Care Unit at Starship
This is calculated based on the actual recorded CLABs per 1000 line days (which averaged under 1.0 per 1000 line days) compared with the project’s estimated baseline of 3.3 per 1000 line days, and an estimated cost of $20,000 per CLAB.
The reduction in CLAB rates from 3.32 to under 1.0 per 1000 line days produces substantial effects. Between April 2012 and March 2013, there were in excess of 31,000 line days in ICUs in New Zealand, which means that just over 100 CLABs would be expected if the baseline rate of 3.32 had not changed. In fact, there were 15. Calculated at a cost of $20,000 per CLAB, this represents avoided cost of nearly $2 million.
Hand hygiene
Hand hygiene is one of the most important measures in the fight against healthcare associated infections, making it a key patient safety issue within the health sector. International evidence is clear that improved hand hygiene practices help reduce healthcare associated infections, including antibiotic-resistant infections within hospitals.
The Commission has entered into partnership with Auckland DHB to implement a national hand hygiene work programme. For information on the programme see the Hand Hygiene New Zealand website.
The WHO approach (‘Five Moments for Hand Hygiene’) being implemented as part of the Hand Hygiene New Zealand programme identifies the following moments as critical to the prevention and control of infections:
- before patient contact
- before a procedure
- after a procedure or body fluid exposure risk
- after patient contact
- after contact with patient surroundings.
Since 2012 all DHBs have undertaken observational audits of compliance with the five moments. These have taken place on a select number of wards and departments inside DHBs. The location of these will vary between DHBs, so some care must be taken in making comparisons. Data for the first two time periods (which end in March 2013) are shown in the following table. The percentage of observed hand hygiene moments that complied with the WHO approach varied from 54 to 75 percent between DHBs. There was also a general increase in compliance in the second period. The number of DHBs where more than 70 percent of observed moments were compliant increased from five to eight between the two time periods.
Green indicates DHB staff were observed to comply with WHO hand hygiene practice at least 70 percent of the time. Grey indicates this occurring 60–70 percent of the time. Deepening shades of red indicate lower percentages. Data are courtesy of Hand Hygiene New Zealand. Further details of the audit can be found on the Hand Hygiene New Zealand website. Note, quarter dates refer to the quarter in which the audit period finished. Thus Q4 2012 ran from July to October 2012 and Q1 2013 from November 2012 to March 2013.
There is clear geographic variation in these data, as the map below shows. The Northern region stands out as having relatively high rates of compliance, while rates for South Island DHBs are considerably lower. It is not immediately clear why this difference should be so marked. The high rates in the north may reflect Auckland DHB’s leadership on this issue.
Green indicates DHB staff were observed to comply with WHO hand hygiene practice at least 70 percent of the time. Grey indicates this occurred 60–70 percent of the time. Deepening shades of red indicate lower percentages. Data are courtesy of Hand Hygiene New Zealand. Further details of the audit can be found on the Hand Hygiene New Zealand website.
Our outcome measure is the number of patients with healthcare associated Staphylococcus aureus bacteraemia per 1000 patient days. This is the most common healthcare associated infection in New Zealand hospitals. In 2012, reported infections reduced.
Perioperative harm
For the majority of people who undergo surgery, it is safe, effective and beneficial. However, surgery is not without risks. The burden of harm that follows complications of surgery is large. Numerous complications of surgery could be chosen for the QSMs, but on the advice of experts and the sector, deep vein thrombosis (DVT)/pulmonary embolism (PE) and sepsis have been chosen. DVT/PE and sepsis are believed to be reasonably sizeable areas of complication, more readily identifiable from routine data sources, amenable to improvement and related to the process measure (the use of World Health Organization (WHO) surgical safety checklist). Using the National Minimum Dataset of hospital admissions we can identify when these two complications occur within a surgical admission, and where patients are readmitted into hospital with either condition within 28 days of a surgical admission.
The graph below shows case numbers in 2012, when there were over 750 events for DVT/PE and over 500 for sepsis. We estimate that these patients stayed in hospital for an additional 3500 days, at a cost of nearly $3 million.
Events and readmissions show the total number of cases where DVT/PE or sepsis was recorded during a hospital stay following surgery, or where a patient was readmitted with DVT/PE or sepsis within 28 days of surgery.
Additional bed days is an estimate of the additional days associated with patients over what would have happened if DVT/PE or sepsis had not occurred. Cost is based on an estimate of $770 per additional day. This is likely to be a conservative estimate.
The total of observed deaths in hospital is shown among this group of patients. This does not imply that all died directly from DVT/PE or sepsis, nor that none may have died had these complications not occurred. It is not a calculation of “excess” or “unexpected” deaths.
Our process measures concentrate on the use of the WHO surgical safety checklist, a commonsense approach to ensuring the correct surgical procedures are carried out on the correct patient. It involves checking the right people are present and that they all are in agreement about why they are operating.
The Commission urges hospitals to use the checklist consistently. When implemented properly, the checklist requires hospital staff to stop and think what they are doing and why. Thinking about what could go wrong is also important; for example, checking for allergies to medicines is part of the checklist. More information on the checklist, including a copy of the Australian and New Zealand checklist, is available here.
The checklist comprises three parts: check-in, time out and sign out. Our marker looks at the proportion of operations where all three parts of the checklist are used.
The table (above) show the result of the baseline audit for use of the surgical safety checklist. Green indicates that all three parts of the checklist were used in 90 percent or more of the cases audited; grey indicates use 75–90 percent of the time; progressively deeper red identifies lower proportions.
*Southern DHB is excluded from this list because it is able to show the position for Southland Hospital (where use is recorded as 100 percent) but not for Dunedin Hospital for use of all three parts of the checklist.
At least some variation may be down to differences in recording. The data for this measure was collected in nearly all cases through case note review which, while easier to undertake than observational audit, does not allow any judgment of the effective use of the checklist. It is possible that a checklist is being used effectively, but not recorded as such, or that the checklist is used in an ineffective way, despite being well recorded.
To address this in future, we plan to include data collected through observation of the checklist in practice. This may have the effect of considerably changing the results for this measure.
Published: 1 Nov 2021
Modified: 8 Nov 2021
