Spend and Value
Mortality
Patient experience
Adverse events
Primary health care access
Equity is treated specially in the indicator set
Deaths potentially avoidable through health care (amenable mortality)
This indicator is well-tested and accepted as a whole-of-system health outcome indicator. It shows the extent to which available treatments are applied to diagnosed conditions and shows the potential for gain in health outcomes. As an internationally calculated indicator, it should, in theory, allow international comparisons, although time spent collating consistent data sets slows down calculation (the most recent internationally available data relate to 2006–07, but we have replicated the methodology for New Zealand alone to give figures for 2008 and 2009).
- The Commission's newly designed in hospital patient experience survey went live for the first time in August 2014 and has now run for three quarters in all DHBs. This covers four key domains of experience: communication, partnership, co-ordination and physical and emotional needs.
- More details are available fromhttp://www.hqsc.govt.nz/our-programmes/health-quality-evaluation/projects/health-quality-and-safety-indicators/patient-experience/adult-inpatient-experience/
- We are working on increasing response rates, but we have, so far had very consistent results between quarters
- New Zealand does not as yet routinely collect data which provides a robust measure of adverse events.
- Determining what such a measure should be is not straightforward, and poorly designed measures of adverse events run the risk of creating perverse incentives (for example, minimising reporting of adverse events which reduces the opportunity to learn and improve when things go wrong).
- One proxy measure that we can use is an estimation of the health loss associated with adverse health care events. These are measured by disability adjusted life years (DALYs) which combines early death (Life Years Lost) with non fatal loss of health and disability (Years lived with disability).
- This analysis is part of a larger study which looks at overall DALYs within a population, but this particular measure looks specifically at DALYs associated with adverse healthcare events
Data are unavailable for 2007
Health care cost/expenditure
- We have combined two indicators to give a more nuanced position on New Zealand health expenditure relative to the rest of the developed world: health care cost per capita (US$ purchasing power parity per capita) and health care expenditure as a proportion of gross domestic product (GDP).
- Data include both public and private health expenditure.
Value
- Expenditure in and of itself does not tell us about the value for money of a service. A commonly used definition of value in health care is outcomes of care divided by the cost of health care. There is no internationally agreed approach to measure the outcomes of a whole health system, but we present here DALYs as an outcome against expenditure as something of an experimental measure.
- We have considered a range of potential measures in this area, for example: ability to enrol with PHO or babies enrolled with PHO in first three months of life.
- While this is still a placeholder, there is an OECD/Commonwealth Fund measure relevant to this area which we present under 'Related information'.
Related information
Commentary
Results
Expenditure
Value
Related information
International comparison
Commentary
As we use questions from the Picker library, we are able to compare with other jurisdictions who also use the same set and published these results. Most notable is England where the NHS survey which uses 10 or so identical questions has run for over 10 years. Responses in New Zealand appear slightly more positive than in the NHS. Where these questions are used in other jurisdictions NZ is broadly in line with their responses.
Attributable burden (percentage of Disability Adjusted Life Years lost) for selected risk factors, 2006
Ethnicity
One area of expenditure where New Zealand stands out internationally is in its reduced pharmaceutical expenditure per capita. The role of PHARMAC in helping to achieve this has been internationally recognised as a major achievement.
http://www.commonwealthfund.org/~/media/Files/Publications/Fund%20Report/2012/Nov/1645_Squires_intl_profiles_hlt_care_systems_2012.pdf
Related information
Average annual number of physician visits per capita, 2010
Health care cost per capita (US$)2011
Results
Commentary
Rather than try and generate some specific equity indicators we stratify the other measures to look at variations in quality of care between different ethnic and socioeconomic groups.
Our view, widely supported within the New Zealand health and disability sector, is that this approach gives both a truer and broader view of equity and more useful data to the sector than would be the case if we sought one or two specific measures that sought to encompass the whole of equity in the New Zealand health care system.
The Commission is developing an Atlas looking specifically at Equity in 2015/16 and measures from this will be included in future compendia
To date the overall scores given to each domain of patient experience by respondents are broadly positive and have consistently rated between 8 and 9 out of 10. This is consistent between DHBs.
The individual questions, however, show variation. Those concerned with what might be called "humane" care - compassion, respect and so forth - score very highly with 80-90 per cent of responders giving the most positive possible responses. However, the more technical aspects show less positive results. Fewer than fifty per cent responders felt that they had had potential side effects fully explained; and only just over half of all respondents thought that they and their family had definitely been as involved in their care as they wished.
International comparison suggests something of a mixed picture. While NZ has the joint lowest per capita use of physician visits (in this context GP appointments) according to OECD data, the Commonwealth Fund survey finds that New Zealand has high rates of prompt access to physicians and a relatively low proportion of patients reporting difficulties in accessing out of hours care (although, as we show below, the position on this measure for below average income respondents is less positive).
NZ and Australian age-standardised amenable mortality rates by year 1997-2006
Age-standardised amenable mortality rate 1997/98 and 2006/07 OECD countries
Source: OECD Health Data 2012.
Co-ordination
The 2011 Survey of sicker adults suggested that a high proportion of NZ GPs were well versed in their patient's medical history in comparison with the international average
Commonwealth Fund comparisons
When we compare, by country, years of life lost or disabled with expenditure on health care, New Zealand sits somewhere in the middle of the pack on expenditure and yet has fewer years of life lost or disabled than many similar countries and many that spend more. Each dot in the graph represents a country (a group of similar countries is shown in green and labelled)
This suggests that New Zealand’s health services are getting good results for a reasonable cost, fulfilling the definition of good value at a high level.
New Zealand’s position on these indicators is interesting. On the one hand, expenditure per capita is relatively low, with only accession countries from the former Soviet Bloc and a number of developing economies in the OECD list spending less per head on health care. However, as a proportion of GDP, expenditure in New Zealand is relatively high. What this implies is that while per capita spend on health care in New Zealand is less than many other countries (and its quality generally comparable with the rest of the developed world), New Zealand is less able than many to easily increase what it spends on health care.
There is no 'right' level of expenditure for health care. It is certainly not the case that more expenditure will necessarily drive better outcomes, but equally, low spend does not necessarily equal greater efficiency.
Communication
The 2013 Survey of Patients suggests that all 11 countries had similar, high proportions of patients who found it easy to understand the explanations doctors gave them
Commonwealth Fund surveys of sicker adults (2011) and patients (2013) have included questions in our domain areas of communication, partnership and coordination. These results are reported here
- The goal of patient safety is to avoid harm to patients associated with the provision of healthcare. Adverse healthcare events are those where harm occurred to patients in the course of receiving healthcare which results in loss of life or loss of healthy living. This can be measured through the idea of a disability adjusted life year (DALY) which estimates both years of life lost and permanent and long-term disability resulting from the adverse event.
- Adverse health care events accounted for 30,300 Disability adjusted life years (DALYs) in 2006, or 3.2% of total health loss from all causes. This estimate includes approximately 18,000 years of life lost (YLL) from an estimated 867 deaths and almost 12,000 years of life disabled (YLD)
- This estimate was derived from a systematic sampling of inpatients admitted to 20 general hospitals in 1998 (Davis et al 2002, 2003), followed by a standardised review of their medical records by an expert panel to identify adverse events and their outcome (fatal, permanent disability, fully recovered).
- It is also important to note that the complex nature of health care (often delivered near the end of life) means that not all of the burden of adverse health care events is potentially avoidable: international evidence suggests that only about one-third may be realistically modifiable.
Courtesy of Ministry of Health, 2013, Health Loss in New Zealand: A report from the New Zealand Burden of Disease, Injuries and Risk factors Study 2006-16
The continued difference between New Zealand and Australia is noteworthy. However, care should be taken in the interpretation of these data. Tobias et al http://www.health.govt.nz/publication/saving-lives-amenable-mortality-new-zealand-1996-2006 note that assuming that the higher amenable mortality rate in New Zealand points to a less effective health system is flawed, "Once corrected for differences in non-amenable mortality (as a proxy for these underlying ‘structural’ factors), no difference in amenable mortality remains (or a slight New Zealand advantage is seen in recent years), suggesting that the two health systems are in fact performing at a similar level of effectiveness."
Partnership
The area of partnership, or involvement of patients in their care is well covered in the 2011 Commonwealth Fund survey of sicker adults. This demonstrates that New Zealand was similar to the international average or better across these measures
% GDP expenditure on health 2011
New Zealand's rate of amenable mortality has fallen notably since 1997. In 2012 we noted that this fall had mirrored the pattern seen in most high-income countries. During this period New Zealand has had one of the higher mortality rates internationally, although it is not a particular outlier. For example, the amenable mortality rate here remained around 30-40 percent higher than in Australia between 1997 and 2007, even as the rate fell.
While amenable mortality is probably the best measure that we have to consider the effect of healthcare on mortality (other measures such as life expectancy are influenced by much broader causes such as poverty, inequality, and social infrastructure as well as quality of healthcare) it does have some weaknesses. It is dependent upon similar recording of details about patients in different countries which cannot necessarily be guaranteed (although similar recording systems are used, local practice in their interpretation can vary). The precision and complexity of calculating the measure together with the need to get nationally consistent data sets makes this quite an 'out of date' indicator, the most recent internationally comparable data available to us relates to 2006-07.
There have been two main approaches to calculation of amenable mortality with different treatments of trauma related deaths. We have replicated the method used by Tobias et al to estimate NZ rates for the years 2008 and 2009. These appear to show a continued reduction in the rate.
We include a complementary measure, potential years of life lost alongside the amenable mortality measure. This looks at deaths under the age of 70 and calculates the total years of life lost through premature death. These data are available from the OECD up to 2010. Again New Zealand has a relative high number of years of life lost, consistent with its relatively high amenable mortality rate.
These data are not available for New Zealand for 2010 so have not been updated in this set.
Potential years of life lost, women, OECD countries 2010 (or nearest available year)
Potential years of life lost, men, OECD countries 2010 (or nearest available year)
Courtesy of Ministry of Health, 2013, Health Loss in New Zealand: A report from the New Zealand Burden of Disease, Injuries and Risk factors Study 2006-16
Functional outcomes
- Functional health outcomes scores are being considered for this area.
- These are very complex measures, and hard to collect.
- We are, however, able to show data relating to Years of Life Disabled, drawing from the Global Burden of Disease survey. Further information about this major international study can be found at http://www.thelancet.com/themed/global-burden-of-disease
- The full burden of disease is measured by a combination of deaths (Years of Life Lost) and loss of function and quality of life (Years of Life Disabled). These can be combined into Disability adjusted life years (DALYs).
- To some extent these measures describe the state of public health the determinants of which are far broader than the functioning of a health system. However these comparisons tell us something about the relative successes and challenges of the system.
Related information
Results
Commentary
The mortality-based approach addresses the issue of living longer, but ignores the related issue of people living better. The measure of Disability Adjusted Life Years (DALYs) lost to disease and injury seeks to address this.
The DALY measures the gap between the population’s current state of health and that of an ideal population in which everyone experiences long lives free from illness or disability. This definition comprises years of life lost (YLL) – based on expected years of life at each age compared with the lowest observed death rates for each age across all countries – and years lived with disability (YLD) – time spent in less than full health.
A recent update of the measure has been developed by the WHO as part of the global burden of disease study. This shows that in 2012 New Zealand had one of the lower estimated DALY burdens globally, of 234 per 1,000 population. In contrast to the amenable mortality measure, New Zealand is in the lowest quartile for DALY burden of even high income countries
Canceled surgery
Older people bed-days
Falls
Ethnicity
End of life
Screening
Under development - hospital days during last six months of life
- The logic of this measure is the belief that in most cases managing death outside of the hospital environment is to be preferred and, therefore, improvements in the management of death should be reflected in shorter average in-hospital care.
- However, this measurement has considerable technical challenges in its construction, and ethical considerations in its interpretation.
Here we show all available analyses by ethnicity.
Cancellations of elective surgery by hospital after admission
- This indicator measures the percentage of elective surgery (excluding maternity surgery) cancelled by the hospital after the patient had been admitted.
- The results provide insights into how close the system is running to capacity and a measure of patient experience.
- This indicator includes patients who were rebooked and admitted at a later date.
Eligible population up to date with cervical screening
- This indicator provides insight into cervical cancer prevention and access to primary health care services.
- Effective screening programmes allow early detection and treatment of cervical precancer, lowering the rate of premature mortality among women.
Occupied bed-days for older people admitted two or more times as an acute admission per year
- This indicator is a useful proxy for the effectiveness of the integration of primary, acute and long-stay care. It illustrates effectiveness at avoiding unnecessary admissions and ‘stepping down’ to less intensive forms of care.
- For ease of international comparison, 'older people' is defined as all those aged 75 and over. We received very helpful feedback that a more useful indicator for New Zealand would also include Maori and Pacific peoples aged 55 and over. This is included in the 'Related information' section.
In hospital falls resulting in fractured neck of femur
- Falls in health care have been identified as the most commonly reported type of harm in the annual serious and sentinel events report. Each year around half of all events with serious harm are falls and around half of these lead to a fractured neck of femur
- 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.
- Since June 2013 the Commission has worked with the sector to address falls within hospital. This had led to improved practice to avoid falls and their subsequent harm
Older people occupied bed-days
Mortality
Occupied bed-days associated with people aged 75+ admitted twice or more as an emergency, per 1,000 population, by ethnic group
Absolute comparisons between different ethnic groups are complicated for this measure as the age distributions are so different (a much greater proportion of the total 'other' - primarily NZ European - population is aged 75+) so we present 75+ and 55+ for each ethnic group. These show however higher occupied bed days associated with Maori and especially Pacific peoples populations.
Results
Commentary
Results
Commentary
International comparison
Related information
Equity
Occupied bed-days associated with people aged 55+ admitted twice or more as an emergency, per 1,000 population, by ethnic group 2008/09 - 2010/11
Results
International comparison
Commentary
Equity
Progress
Between 2008 and 2012 around 1 per cent of operations were cancelled after admission relatively consistent. However, since September 2012, this has reduced to around 0.9 per cent, a small drop but one which is significant and equates to 800 cancellations avoided across the country, we estimate that this reduction has avoided costs of at least $600k,
This reduction appears to have been sustained as the run chart opposite demonstrates.
Occupied bed days associated with people aged 75+ and Maori and Pacific Islanders aged 55-74 admitted twice or more as an emergency per 1000 population
Occupied bed-days associated with 75s and over admitted twice or more as an emergency in New Zealand and England
Related information
Results
Occupied bed-days associated with people aged 75+ admitted twice or more as an emergency, per 1,000 population, by ethnic group
This variant of the measure includes Maori and Pacific peoples aged 55-74. When tested during our consultation process, respondents considered this measure more appropriate for New Zealand.
This Commonwealth Fund chart based on OECD data shows that New Zealand's rate of cervical screening in 2011 was towards the top of the international range, being very similar to that of Norway. The screening rate has, of course, increased significantly since then. Being based upon a registry rather than survey data, the New Zealand figure is also more reliable than many. In this instance the numbers on the graph show the percentage of women screened.
Comparison with England demonstrates that New Zealand has notably low levels of bed occupancy associated with older people returning to hospital as an acute admission. This is suggestive of relatively successful integration of primary, hospital and aged care.
Absolute comparisons between different ethnic groups are complicated for this measure as the age distributions are so different (a much greater proportion of the total 'other' - primarily NZ European - population is aged 75+) so we present 75+ and 55+ for each ethnic group. These show, however, higher occupied bed days associated with Maori and especially Pacific peoples populations, regardless of which age group is considered.
Related information
Commentary
The issue of a "good death" and how to achieve it is one which is emerging within healthcare circles. Dr Atul Gawande's bestselling book Being Mortal has opened up this conversation with a broader audience.
wh
The Commission brought Dr Gawande to New Zealand in May 2015 (http://www.hqsc.govt.nz/publications-and-resources/publication/2189/).
We continue to develop an approach to measuring this sensitive topic. Our early work has been concerned with what the situation in New Zealand actually is. Much early work in the area has focused on the US system which has many internal incentives for a highly medicalised death in hospital. As New Zealand does not have the same incentives we were interested in whether the same situation held. Our early results suggest not, and we will publish more on this as we firm up the analysis.
Given the fact that cervical cancer is a greater risk for women over 40, and Maori, Pacific and Asian women, screening rates across different groups are of particular interest.
While it was the case that women over the age of 60 were considerably less likely to be screened for cervical cancer, this difference is reducing. Screening rates between the ages of 35 and 60have stayed relative constant since 2012, but for those aged 60-64 there has been a five percent increas and for those over 65 a 10 percent increase.
In terms of inter-ethnic variation it is notable that screening rate for women in the 'other ethnic groups' (primarily these are NZ European) remains higher, but that increases have been greatest in the Pacific group.
Occupied bed-days associated with people aged 55+ admitted twice or more as an emergency, per 1,000 population, by ethnic group 2008/09 - 2010/11
Good integration of care services is an increasing priority for health systems in the developed world, and an area of particular concern for ageing populations. Poorly integrated care results in older people ‘falling down the gaps’ until the most urgent, intensive and expensive care – an acute admission to hospital – is required. A low number of occupied bed-days per capita and low regional variation are desirable.
After a reduction in both absolute rate and variation between 2008 and 2011, both measures have stayed reasonably consistent for the last 3 years. The variation between different parts of the country is nearly 3 fold which suggests that further improvement is undoubtedly possible.
Compared with the UK (the other country where there is a consistent time series for this indicator), New Zealand has around a 40 percent lower level of bed occupancy and considerably less regional variation. The variation that exists prompts the question, could this rate be improved further through widespread adoption of the integration practices seen in areas with the lowest rates?
This measures the percentage of eligible women (aged 25-69 years) who have received a cervical smear in the past three and five years. This measure has been part of the PHO Performance Programme since 1 January 2011. However, we have used data collected and reported by the National Screening Unit which is responsible for organising the Ministry of Health's National Cervical Screening Programme, which includes health promotion, smear taking, laboratory analysis of cervical smears, colposcopy and management of women with abnormal smear results.
In New Zealand, approximately 160 women develop cancer of the cervix each year, and about 60 women die from it. Some groups of women have higher rates of cervical cancer, including women over 40, Māori, Pacific and Asian women, unscreened women and under-screened women.
The National Screening Unit estimate that without screening 1 in 90 women will develop cervical cancer and 1 in 200 women will die of cervical cancer. In contrast with screening 1 in 570 women will develop cervical cancer and 1 in 1,280 women will die of cervical cancer.
Since 2012 the proportion of women who have been screened in the last three years has increased by two per cent, and in the last five years by three per cent
More information about the National Cervical Screening Programme and cervical screening itself is available from http://www.nsu.govt.nz/current-nsu-programmes/908.aspx
Gout
Diabetes
Vaccinations
Screening
Mental Health
We have settled on fractured neck of femur following a fall in hospital as a reasonable compromise between an event of unequivocal harm and cause and one with reasonable numbers. These represent around a quarter of all serious and sentinel events reported in the annual SSE report, so are a substantial proportion of recorded harm.
The number of falls has remained fairly consistent over the last two and a half years.
We estimate that the average increase in length of stay associated with falling in hospital and fracturing neck of femur is over month.
The first focus of the Commission’s Open campaign was in Falls from May 2013. This has led to improved practice in terms of falls prevention actions such as risk assessment and care planning. However the hip fracture has, until, very recently, remained consistent. This is not altogether unsurprising as these sorts of programmes do take time to show improved outcomes. Nevertheless, very recent data have shown a reduction in falls with a fractured neck of femur which could be the start of a significant shift down.
Post discharge community care (DHB only)
In terms of inter-ethnic variation it is notable that screening rate for women in the 'other ethnic groups' (primarily these are NZ European) remains higher, but that increases have been greatest in the Pacific group.
There is comparatively little variation between different ethnic groups (ranging from 92 percent for Maori to 97 percent for Asian at 24 months). As overall rates of immunisation have increased, inequalities between different ethnic group have reduced.
There is no statistically significant difference in access to post discharge community care between the Maori and "Other" (primarily NZ European) groups. Both groups have seen an increase in access to community care over the last three years, but this has been more rapid for the Maori group.
Percentage of discharges from acute inpatient units where a community mental health contact with client participation was recorded in the seven days immediately following that discharge by ethnic group
Diabetes is an interesting disease to consider from the point of view of ethnic equity as prevalence is disproportionately high among Pacific Island populations. Consideration of two aspects of management, regular testing of HbA1c levels, and the use of metformin or insulin, suggests that there is little disparity in treatment between different ethnic groups
These data along with many others is available from the Commission's Atlas of Variation http://www.hqsc.govt.nz/our-programmes/health-quality-evaluation/projects/atlas-of-healthcare-variation/diabetes/
Gout is the most common form of inflammatory arthritis, estimated to affect approximately 3.75 percent of adult New Zealanders. This is high internationally. Both Māori and Pacific peoples are more likely to have gout than other populations So access to the correct therapy is a matter of equity and justice.
Long-term urate-lowering therapy (of which allopurinol is the most commonly prescribed) is effective to prevent acute gout flares and can reduce the risks of permanent disability in people with gout, but only 40 per cent of gout patients receive this, and Māori and Pacific populations, despite being more likely to have the disease, are less likely to receive this therapy
Colchicine is frequently prescribed as treatment for acute gout flares. The proportion of people prescribed colchicine but not allopurinol, , gives an indication about how well gout is being managed. This indicator suggest gout is notably less well managed in Māori and Pacific populations – the very people most likely to suffer from gout.
Commentary
Day case overstay
Socioeconomic
Infections
Vaccinations
Infection prevention and control
- Preventing healthcare associated infections is a part of the Commission's patient safety campaign, including ensuring good hand hygiene and preventing central line associated bacteraemia (CLAB) in intensive care units.
- Central line associated bacteremias in ICUs have been a long standing issue in healthcare and have often been considered an inevitability. Recent evidence shows however that the introduction of a small and low cost bundle of interventions can virtually eliminate these. New Zealand established a collaborative (Target CLAB Zero) working between DHBs to attempt to do this.
- S. aureus bacteremia is the most common healthcare associated infection in New Zealand hospitals, and can be associated with increased time in hospital, disablement and even death. Good hand hygiene is one way of reducing the risk of this infection.
- Surgical site infections (SSIs) can cause emotional and financial stress, serious illness, longer hospital stays, long-term disabilities, and can result in loss of life; but a significant number of them are preventable. To address this, in 2012 the Health Quality & Safety Commission (the Commission) launched the Surgical Site Infection Improvement (SSII) Programme – New Zealand’s first national quality improvement programme to reduce SSIs.
Planned day case turns into unplanned overnight stay
- This indicator captures inconvenience to patients and disruption to planned hospital flow.
- The data may reflect an adverse incident in a procedure, unrealistic expectations about which patients are suitable for day-case surgery or some local factor.
- The indicator operates as a prompt for further enquiry and a measurement of quality and efficiency.
- This particular indicator has data only the end of July 2013
International comparison
Rating care poor of fair
Vaccinations
Age-appropriate vaccinations for two-year-olds
- This indicator on the effectiveness of immunisation programmes provides a perspective on public health programmes as well reflecting level of access to primary health care services.
- Children who receive the complete set of age-appropriate vaccinations are less likely to become ill from the associated diseases.
The Commonwealth fund 2013 survey of patients looked at issues of experience and access for patients looking at both experience of those on below average incomes and the comparison between this and those with above average incomes. This gives a mixed picture.
Poorer New Zealanders rate their healthcare highly compared with the international average and the difference between different income groups is not marked. Similarly poorer New Zealanders are slightly less likely than average not to get a test or treatment because of the cost. However, they are more likely than average not to visit a doctor because of the cost (second only to the US) and are much more likely to have difficulty in accessing out of hours care compared with wealthier groups.
A similar pattern emerges for deprivation. There is effectively no difference recorded between immunisation rates for different deprivation quintiles at 24 months. Again this represents an increase in equity over the last five years.
Did not visit doctor because of cost
Commentary
Difficult to get out of hours care
Did not get test or treatment because of cost
Equity
Results
Commentary
Related information
Commentary
Equity
Day case overstay rate by ethnic group 2010-11
% day cases become overnight stays and % day cases
While day cases as a proportion of no-acute hospital events have increased slightly since 2008, the level of day case overstay has remained stable.
There is no significant difference in the proportion of day cases that overstay between different ethnic groups. The figures on the graph show the actual number of overstays in the year, rather than the percentage of overstays, allowing a comparison of the size of the overstay issue.
% day cases become overnight stays
Commentary
There is comparatively little variation between different ethnic groups (ranging from 92 percent for Maori to 97 percent for Pacific populations at 24 months). This represents a substantial reduction in unequal access to immunisation over the 5 years. As overall rates of immunisation have increased, inequalities between different ethnic group have reduced.
The vaccinations that fall within the two-year-old group are for measles, mumps, rubella, diphtheria, tetanus, whooping cough, polio, hepatitis B, pneumococcus and Haemophilus.
High coverage is important to protect the health of both individuals and whole communities. It reduces the spread of disease to those who have not been vaccinated either by choice or because of medical reasons, such as children who are immune compromised.
Overall coverage levels are relatively high. The most recent data suggest that around 93 percent of children have received the complete set of age-appropriate vaccinations at age two, and that regional variation is comparatively low.
Day cases that overstay (numbers)
A number of stakeholders expressed an interest in seeing this measure as a total volume of activity rather than a percentage. In total there are typically 600-700 day cases that end up as overnight stays per month in New Zealand. The number has increased slightly in recent years, but this reflects an increase in the number of day cases.
Nationally, the proportion of day cases that turn into unplanned overnight stays has remained consistent over the last five years. Nevertheless, on the face of it, this figure equates to up to 10,000 people a year who expected to be in and out of hospital in a day who had to make an overnight stay.
This measure does not identify the reasons for an overstay and there may be a very legitimate clinical reasons for keeping patients overnight. Hence, the results need to be interpreted with caution. To help with this we show the change in national intended day case rate in the next pane. Whilst there are some caveats to this measure (we have excluded two DHBs from this calculation as their recording of day cases is inconsistent compared to the rest of the country), there is no obvious relationship between day case rate and day case overstay rate. In other words the places with the highest day case rates are not those with the most overstays.
We would anticipate variation in results between DHBs in relation to demographic or geographic factors (for example, in rural settings a potentially longer distance to hospital may affect ability to travel within the same day).
Unfortunately this measure is no longer calculable from NMDS data and may need to be retired
CLAB
Following the implementation of the Target CLAB Zero CLABs appear to have been almost eliminated in New Zealand Intensive Care Unit. From what was considered a conservative estimate of 3.3 CLABs per 1000 days in 2011, there are now considerably fewer than 1 per 1,000 line days.
To put another way, in the since April 2012 there have been around 260 fewer CLABs recorded in New Zealand ICUs than would have been had CLABs continued to occur at the pre-existing rate. This has avoided costs of over $5m
Relationship between day case overstays and proportion of activity attempted as a day case by DHB 2009/10 - 2010/11
A similar pattern emerges for deprivation. There is effectively no difference recorded between immunisation rates for different deprivation quintiles at 24 months. Again this represents an increase in equity over the last five years.
A number of stakeholders raised the question of whether high overstay rates were associated with high day case rates. In other words did those hospitals with the highest levels of overstay do so because they try to do more work as a day case. The simple answer is 'no'. There is no evidence for this. In fact the reverse is closer to the truth, although the correlation between the two measures is overall low.
S. aureus bacteremia
Measurement of harm in the field of infection control has, not unreasonably, concentrated on reduction of infections. There is less history of measurement of harm (for example in increased mortality) or cost (although estimations such as those of Graves et al 2003 and Cummings et al 2010 exist).
In particular, original research in the effects of hand hygiene programmes have tended to concentrate on S. aureus infection rates ( Kirkland et al 2012, Roberts et al 2012) or Methicillin-resistant Staph. Aureus infection rates (Grayson et al 2008, ). These papers tend to demonstrate that improvements in hand hygiene are associated with a reduction in infection rates, making clear the intervention logic and appropriateness of linking together these measures as related process and outcome markers.
Hand Hygiene NZ adopted healthcare associated Staphylococcus aureus bacteraemia per 1000 patient days as its outcome measure. Since S. aureus is the most common healthcare associated pathogen in most New Zealand hospitals we believe this to be the best easily available measure.
Data collected by HHNZ shows a relatively consistent rate of infection over the last two years.
Surgical site Infection
Despite month by month variation, surgical site infection rates have remained relatively consistent over the last 18 months
Surgical harm
Age
Readmissions
Screening
While it was the case that women over the age of 60 were considerably less likely to be screened for cervical cancer, this difference is reducing. Screening rates between the ages of 35 and 60have stayed relative constant since 2012, but for those aged 60-64 there has been a five percent increas and for those over 65 a 10 per cent increase.
Perioperative harm
- There are potentially many harms associated with operations that could be included in this analysis. We have chosen to include two which are relatively numerous and unequivocal and potentially serious : Deep Vein Thrombosis/Pulmonary Embolism, and Postoperative Sepsis
- This measure will align with the quality and safety marker for perioperative harm.
- For comparative purposes we have included data from OECD in 'Related information'.
Emergency readmission to hospital within 28 days of discharge
- This indicator is a proxy of both the care received in hospitals and the coordination of care back to and within the outpatient setting.
- While some readmissions are part of planned care and are desirable, others may be an indication of a quality issue related to shortened length of stay and premature discharge, inadequate care, lack of patient adherence to the care regimen following discharge from hospital or poor integration of care.
- The precise definition of readmission indicators is notoriously variable as defining what precisely is an unplanned readmission varies. We present here the measure as defined by the Ministry of Health. The Ministry is currently working on a redefinition of this measure, and the data series therefore stops at 2011.
- Work on a revised measure has recently commenced and we look forward to restoring this measure in future.
International comparison
Commentary
Related information
Related information
Results
The use of control charts as a method to detect meaningful change shows that postoperative DVT/PE has have been largely stable over this period, but post operative sepsis has significantly increased.
Comparison of Average LOS with 28 day acute readmission rate, by DHB, 2010/11
At first sight, and at a whole DHB level, there is little evidence of a clear link between length of stay and readmission rate. Further analysis would be beneficial here including looking at a smaller subset of conditions with higher readmission rates.
% of hospital admissions followed by an acute readmission within 28 days of discharge
Variants of both DVT/PE and sepsis incidence are produced by the OECD, and both show New Zealand towards the lower end of the performance on these measures.
International comparisons are, however, notoriously tricky. The information systems from which these data are drawn in different countries vary in terms of their coverage, the consistency of how complications are recorded, different interpretations of what constitutes a complication, the underlying health of the population, the culture of when surgeons will or will not operate depending upon the underpinning health of the patient, coding systems used and many others.
For this reason change over time is in many ways a more certain figure from which to draw conclusions, we at least can understand the contextual changes and their likely effect in one country. Across multiple countries this endeavour is almost impossible.
This is not a counsel of complacency. These results, and in particular the apparent increase in post operative sepsis raise questions that demand an answer and show that our concentration on perioperative harm one in the Open campaign is warranted.
This measure is essentially the same as that used in the Quality and Safety markers, but to ensure that the measure can be meaningfully compared over time it is turned into a rate against admissions. The use of the “at risk” admissions is to denote that certain cases, where there is no real risk of the complication, are excluded from the calculation of the rate. Fuller details of the construction of these indicators is available at the HQMNZ website.
Postoperative DVT/PE appears to have been largely stable over this period, but post operative sepsis is increasing, the reasons for this are unclear, but may reflect operations being more commonly performed on patients more susceptible to post operative infections .
Readmission rates have consistently increased each year, rising from 8 percent in 2007 to 9.2 percent in 2011. The most recent rates internationally suggest that New Zealand’s readmission rate is fairly typical although precise definitions of the indicator vary between countries. England has shown a substantial increase in readmission rates in the last 10 years, and these stand at around 11 percent. Similarly, recent data from Canada suggest a readmission rate of around 8 percent.
This is a fairly crude indicator that does not take into account the nature of unplanned readmission, or whether appropriate care available in the community may have prevented the need for admission.
It is likely to be influenced by demographic factors, such as the proportion of older people within a district population, and by existing levels of co-morbidity.
Nonetheless our analysis shows that there is not, as is often supposed, a direct link between higher readmission rates and shorter length of stay.
Medication
Mental health
Measure of safe medication management
- Ensuring safe medication management is part of the Commission's patient safety campaign.
- This measure will align with the quality and safety marker for medication management.
- In this edition to provide context we show a Commonwealth Fund survey response on medication error
Mental health post-discharge community care
A responsive community support system for people who have experienced an acute psychiatric episode requiring hospitalisation is essential to maintain clinical and functional stability and to minimise the need for hospital readmission. Service users leaving hospital after an admission with a formal discharge plan involving linkages with community services and supports are less likely to need early readmission. Research indicates that service users have increase vulnerability following discharge including higher risk of suicide.
Results
Commentary
Related information
Commentary
Percentage of discharges from acute inpatient units where a community mental health contact with client participation was recorded in the seven days immediately following that discharge by ethnic group
Based upon this three year old survey of 750 sicker adults, the proportion of patients who recall being given a wrong drug or dose is similar to the average of 11 industrialised nations surveyed
Results
Measuring harm from medication error is recognised as both important, and very difficult to do. Many, indeed most, medication errors result in little or no harm, but they can be catastrophic and even fatal.
Notes on source
The data show a steady increase in the national average proportion of service users who are in contact with community services within 7 days of discharge since 2010. However, there remains considerable variation between different parts of the country for this measure. However, the highest level has fallen for the last two years.
No DHB has achieved the aspirational target of 90 per cent of service users being in contact with community services within 7 days of discharge.
This indicator and several others is available from the HQSC Atlas of Healthcare Variation.
The national KPI Programme for Mental Health and Addiction Services is a Ministry of Health funded provider-led initiative in which district health boards (DHBs) and non-government organisations (NGOs) report nationally comparable indicators of service performance.
The vision of the KPI work is to enable the Sector to bring about demonstrable improvement to service user outcomes through ready access to, and use of comparative service performance information. This approach promotes the sharing of information, ideas and learning to drive service improvement and is supported by provider governance, national coordination and the formation of peer groupings based on issues of mutual interest and benefit.
The KPI Framework was developed as a quality and performance improvement tool, to improve outcomes for people who use mental health and addiction services. The indicators are designed to be used as tools to promote greater understanding about the differences between services and, in particular, to prompt discussion about the activities that lead to improved outcomes.
The complexity of mental health and addiction services means that no one indicator stands alone as a measure of organisational performance.
Pressure injury
CVD
Measure of cardiovascular disease (CVD) management
- CVD is a major killer in New Zealand, as it is in many developed countries. The Commission has recently explored management of CVD using triple therapy in its Atlas of Healthcare Variation and we are considering what measure may be most appropriate to use for this disease.
Measures of pressure injury are being developed through the Office of the Chief Nurse. Once routinely available we intend to include these here.
Related information
Results
Commentary
Related information
Results
Commentary