Skip to content
Tier 3 - Health system performance

3.07 Selected potentially preventable hospital admissions

Key facts

From July 2015 to June 2017:

Why is it important?

Hospitalisations for conditions that can be effectively treated in a non-hospital setting are referred to as Potentially Preventable Hospitalisations (PPH). This is a key measure of the performance of the health system. In particular, it serves as a proxy measure of access to timely, effective and appropriate primary and community-based care (AIHW 2020). An analysis of the conditions for which people are admitted to hospital reveals that, in many cases, the hospital admission could have been prevented through timely and effective care outside hospital (Li et al. 2009).

Measures of PPH include conditions for which hospitalisation is potentially avoidable through effective preventive measures and early diagnosis or disease management, usually delivered through primary health care (Page et al. 2007). The term ‘PPH’ does not mean that the patient did not require hospitalisation at the time of the admission, but that hospitalisation could potentially have been prevented with effective management in community health care settings (AIHW 2019).

PPH fall into three broad categories:

  • vaccine-preventable conditions—for example, influenza, tetanus, whooping cough, chicken pox, measles and so forth.
  • acute conditions—including cellulitis (skin infections), urinary tract infections, convulsions/epilepsy, dental conditions, ear nose and throat infections
  • chronic conditions—including many forms of cardiovascular disease, chronic obstructive pulmonary disease (COPD), diabetes complications, asthma, iron deficiency and hypertension.

Systematic differences in hospitalisation rates for Aboriginal and Torres Strait Islander people and non-Indigenous Australians can indicate gaps in the provision of population health interventions (such as immunisation), primary care services (such as early interventions to detect and treat chronic disease), and continuing care support (such as care planning for people with chronic illnesses, e.g., congestive heart failure). Higher hospitalisation rates can also reflect appropriate referral mechanisms and access to hospital care. Among Indigenous Australians, there is also a higher prevalence for the underlying diseases, indicating deficiencies in access to primary health care and prevention/health promotion services. Indigenous Australians are also more likely to live in remote areas where non-hospital alternatives are limited (Gibson & Segal 2009; Li et al. 2009).

Findings

What does the data tell us?

In the period from July 2015 to June 2017, there were 81,134 hospitalisations, excluding those for dialysis, for Indigenous Australians that were potentially preventable. This accounted for 15% of all hospitalisations among Indigenous Australians. For potentially preventable conditions, the hospitalisation rate for Indigenous Australians was 2.7 times the rate of non-Indigenous Australians (69 per 1,000 compared with 26 per 1,000) (Table D3.07.3).

From 2013–14 to 2016–17, hospital admissions for conditions that were potentially preventable increased by 22% for Indigenous Australians (from 34,387 to 41,936) and 19% for non-Indigenous (from 556,135 to 662,467). After adjusting for differences in the age structure between the two populations, this equated to an increase of 7.2 per 1,000 for Indigenous Australians (from 63 to 70 per 1,000) compared with 2.7 per 1,000 for non-Indigenous Australians (from 23 to 26 per 1,000) (Table D3.07.11, Figure 3.07.1).

Figure 3.07.1: Age-standardised rates for potentially preventable hospitalisations, Australia, 2013–14 to 2016–17

This line chart shows that the rate of potentially preventable hospitalisations increased for Indigenous Australians from 63 to 70 per 1,000. Rates for non-Indigenous Australians were lower but also increased, from 23 to 26 per 1,000. The gap between Indigenous and non-Indigenous increased from 40 to 44 per 1,000.

Source: Table D3.07.11. AIHW analysis of National Hospital Morbidity Database.

Potentially preventable hospitalisations by age

Over the period July 2015 to June 2017, PPH were highest for those aged 65 and over. The rate for Indigenous Australians aged 65 and over was 163 per 1,000, while for non-Indigenous Australians the rate was 85 per 1,000. This age group also had the largest gap between Indigenous and non-Indigenous Australians, at 78 per 1,000, followed by those aged 55–64, with a gap of 77 per 1,000 (Table D3.07.1, Figure 3.07.2).

Figure 3.07.2: Potentially preventable hospital admissions, by Indigenous status and age group, July 2013–June 2015

This bar chart shows that the rate of of potentially preventable hospitalisations for Indigenous Australians decreased between ages 0-4 and 5-14 (from 49 to 23 per 1,000), then increased with age to 163 per 1,000 for those aged 55 and over. Rates for non-Indigenous Australians were lower than the rates for Indigenous Australians in all age groups, declining between ages 0-4 to 25-34 (from 28 to 11 per 1,000), then increased steadily before peaking at 55 and over (85 per 1,000).

Source: Table D3.07.1. AIHW analysis of National Hospital Morbidity Database.

Potentially preventable hospitalisations by sex

In 2016–17, PPH for chronic conditions were higher for Indigenous females than for Indigenous males (36 and 31 per 1,000, respectively). The rate was similar for non-Indigenous females and males (both 12 per 1,000) (Table D3.07.9).

For PPH for acute conditions, the rate was also higher for Indigenous females than for Indigenous males (31 and 26 per 1,000, respectively). However, the rate was similar for non-Indigenous females and males (12 and 13 per 1,000, respectively) (Table D3.07.10).

Leading causes of potentially preventable hospitalisations

In the period from July 2015 to June 2017, acute conditions accounted for 51% (41,058) of all hospitalisations that were potentially preventable for Indigenous Australians, chronic conditions for 39% (31,851) and vaccine-preventable conditions for 12% (10,000). After adjusting for differences in the age structure between the two populations, Indigenous Australians were hospitalised for chronic conditions at a rate of 33 per 1,000 (compared with 12 per 1,000 for non-Indigenous Australians), acute conditions at a rate of 28 per 1000 (compared with 12 per 1,000) and vaccine‑preventable conditions at a rate of 9 per 1,000 (compared with 2 per 1,000) (Table D3.07.12).

For Indigenous Australians, the leading causes of hospitalisation due to potentially preventable chronic conditions were:

  • COPD (28% of all potentially preventable chronic conditions) 11 per 1,000 for Indigenous Australians, compared with 2.5 per 1,000 for non-Indigenous Australians after adjusting for difference in age structure between the two populations
  • diabetes complications (20%) 6.1 per 1,000 compared with 1.7 per 1,000
  • asthma (12%) 2.6 per 1,000 compared with 1.3 per 1,000 (Table D3.07.12).

The leading causes of hospitalisation due to potentially preventable acute conditions for Indigenous Australians were:

  • cellulitis, a skin infection usually caused by bacteria (28%) 7.1 per 1,000
  • convulsions and epilepsy (19%), 5.4 per 1,000
  • dental conditions (17%) 3.5 per 1,000 (Table D3.07.12).

The leading causes of hospitalisation due to potentially preventable acute conditions for non-Indigenous Australians were:

  • urinary tract infections (24%), 2.8 per 1,000
  • dental conditions and ear nose and throat infections (21%), 2.7 and 1.9 per 1,000 respectively
  • cellulitis (14%) 2.4 per 1,000 (Table D3.07.12).

The largest gap in hospitalisation due to potentially preventable acute conditions between Indigenous and non-Indigenous Australians was 4.7 per 1,000 for cellulitis.

The leading causes of hospitalisation due to vaccine-preventable conditions for Indigenous Australians were chronic hepatitis (5.9 per 1,000), influenza (1.5 per 1,000) and pneumonia (1.3 per 1,000). These conditions were also the leading causes for non-Indigenous Australians, although they each accounted for fewer than 1 per 1,000 of hospital admissions of potentially preventable conditions. Of these conditions, the largest gap between Indigenous and non-Indigenous Australians was for chronic hepatitis where Indigenous Australians were almost 8 times as likely to be admitted to hospital as non-Indigenous Australians (Table D3.07.12, Figure 3.07.3).

Although the hospitalisation rates for potentially preventable conditions were higher for Indigenous Australians, the average time spent in hospital was higher for non-Indigenous Australians (3.6 bed days compared with 4 days, respectively) (Table D3.07.5).

Figure 3.07.3: Age-standardised top 10 potentially preventable hospitalisations for Indigenous Australians, Australia, July 2015 to June 2017

This bar chart shows that the most common type of potentially preventable hospitalisation for Indigenous Australians was chronic obstructive pulmonary disease (11 per 1,000), followed by cellulitis (7 per 1,000), diabetes complications (6 per 1,000), urinary tract infections (6 per 1,000) and convulsions and epilepsy (5 per 1,000). The causes were different and lower for non-Indigenous Australians, with the most common types being urinary tract infections (2.8 per 1,000), dental conditions (2.7 per 1,000) and chronic obstructive pulmonary disease (2.5 per 1,000).

Source: Table D3.07.5. AIHW analysis of National Hospital Morbidity Database.

Causes of potentially preventable hospitalisations by age group

The most common causes of hospital admissions that might have been preventable varied by age group. For both Indigenous and non-Indigenous infants aged under 1 year, the leading cause of hospitalisation was ear, nose and throat infections (38 per 1,000 and 16 per 1,000, respectively). For children aged 1–14, dental conditions were the leading cause of PPH for both Indigenous and non-Indigenous children (8.2 per 1,000 and 5.8 per 1,000, respectively) (Table D3.07.6).

Cellulitis was the leading cause of potentially preventable hospital admissions for Indigenous Australians aged 15–24 (3.8 per 1,000) and those aged 25–44 (7.9 per 1,000). For Indigenous Australians aged 45 and over, COPD was the leading cause of PPH at a rate of 18 per 1,000 for those aged 45–64 and 46 per 1,000 for those aged 65 and over (Table D3.07.6).

Trends in potentially preventable hospitalisations

The potentially preventable hospitalisation rates for both chronic and acute conditions among Indigenous Australians increased between 2010–11 and 2016–17. The rate for chronic conditions was 30 per 1,000 in 2010–11 and 34 per 1,000 in 2016–17 (Table D3.07.9). The rate of hospitalisation for Indigenous Australians with acute conditions was 24 per 1,000 in 2010–11 and 29 per 1,000 in 2016–17 (Table D3.07.10). Time-series data are not presented separately for rates of vaccine preventable hospitalisations because there was a change in data coding practices in 2013–14 (Table D3.07.11).

Potentially preventable hospitalisations by jurisdiction

From July 2015 to June 2017, hospital admissions that were potentially preventable for Indigenous Australians were lowest in Tasmania (24 per 1,000) and highest in the Northern Territory (127 per 1,000) (Table D3.07.2, Figure 3.07.4).

Figure 3.07.4: Age-standardised potentially preventable hospitalisations, by Indigenous status and jurisdiction, July 2015 to June 2017

This bar chart shows that jurisdiction with the highest rate of potentially preventable hospitalisations for Indigenous Australians was the NT (127 per 1,000). Rates for Indigenous Australians in the other jurisdictions were much lower, ranging from 91 per 1,000 in WA to 24 per 1,000 in Tasmania. Rates for non-Indigenous Australians were lower than Indigenous Australians in all jurisdictions, ranging from 31 per 1,000 in NT to 21 per 1,000 in ACT.

Source: Table D3.07.2. AIHW analysis of National Hospital Morbidity Database.

Potentially preventable hospitalisations by remoteness

From July 2015 to June 2017, after adjusting for differences in the age structure between the two populations, PPH in Non-remote areas for Indigenous Australians was twice as high as for non-Indigenous Australians (56 per 1,000 compared with 25 per 1,000). In Remote and Very remote areas combined, the rate for Indigenous Australians was much higher than the rate in Non-remote areas, at 119 per 1,000. This was 3.9 times the rate of non-Indigenous Australians in Remote and Very remote areas combined (30 per 1,000) (Table D3.07.13). When this is disaggregated into the five remoteness categories, the rate was the lowest in Major cities and Inner regional areas (both 49 per 1,000) and highest in and Very remote areas (120 per 1,000) (Table D3.07.3, Figure 3.07.3).

Figure 3.07.5: Age-standardised potentially preventable hospitalisations by Indigenous status and remoteness, Australia, July 2015 to June 2017

This bar chart shows that the rate of potentially preventable hospitalisation for Indigenous Australians increased with remoteness, ranging from 49 per 1,000 in both Major cities and Inner regional areas to 120 per 1,000 in Very remote areas. Rates for non-Indigenous Australians were lower than for Indigenous Australians in all remoteness categories, ranging from 25 per 1,000 in Major cities to 31 per 1,000 in Very remote areas.

Source: Table D3.07.3. AIHW analysis of National Hospital Morbidity Database.

What do research and evaluations tell us?

Please refer to measure 1.24 Avoidable and preventable deaths for information on relevant evaluations.

Research shows that disparities exist in the PPH rates between Indigenous and non-Indigenous Australians, reflecting differences in utilisation of care as well as the underlying conditions and social determinants.

A study of Indigenous residents living in remote Northern Territory communities found that those who utilised primary health care at medium/high levels were less likely to be admitted to hospital (and to die) than those in the low utilisation group (Zhao et al. 2014). Higher levels of primary care utilisation for renal disease reduced avoidable hospitalisations by 82–85% and for ischaemic heart disease the reduction was 63–78%.

Research by Harrold and others (2014) found that the Indigenous Australian rate of PPH was 2.16 times as high as the rate for non-Indigenous Australians who lived in the same Statistical Local Area in New South Wales, after controlling for age and sex (Harrold et al. 2014). The disparity was greatest in rural and remote areas relative to major cities. The largest differences in PPH were for diabetes complications, COPD and rheumatic heart disease. Geospatial analysis of PPH data can help identify areas where greater effort is needed to target the determinants of disease and to better manage chronic disease through culturally appropriate primary health care (Harrold et al. 2014).

An Australian study found that personal sociodemographic and health characteristics are major drivers of geographic variation in PPH rates in New South Wales, and explain more of the variation than general practitioner (GPs) supply (Falster M et al. 2015). These personal characteristics (which included age, sex, Indigenous status, highest level of education, annual household income, level of psychological distress and more) also explained a greater amount of the variation for chronic conditions than for acute or vaccine-preventable conditions.

A recent Australian study found that there are substantial inequalities in paediatric avoidable hospitalisations between Indigenous and non-Indigenous children, regardless of where they live. The avoidable hospitalisation rates were found to be almost double in Aboriginal compared with non-Aboriginal children aged less than 2. Respiratory and infectious conditions were the most common reason for hospitalisation for children of all ages in the study, with Indigenous children being more likely to be hospitalised for all conditions (Falster K et al. 2016).

Implications

PPH rates have typically been seen as useful indicators of the effectiveness of, or access to community-based health services (Passey et al. 2015).The rates of PPH for Indigenous Australians are rising and are considerably higher than rates for non-Indigenous Australians, particularly in Remote and Very remote areas. A recent AIHW report highlights opportunities that exist to prevent hospitalisations through primary health care interventions including:

  • reducing and managing risk factors for disease
  • vaccination
  • oral health checks
  • lifestyle interventions
  • management of chronic conditions
  • antenatal care (AIHW 2020).

Several studies have found that improving patient-provider communication and collaboration makes it easier for people to navigate, understand and use information and services to take care of their health. This could include matching information to the patient’s needs and abilities, recognising the importance of asking questions, shared decision-making, and providing a range of avenues for communication (Hernandez et al. 2012; Øvretveit 2012).

For Indigenous children, there is scope to reduce PPH though targeted prevention, early intervention through primary health care and better access to treatment for common childhood conditions. Policy measures that aim to reduce disparities in social determinants (such as access to better housing) may also help to reduce the incidence of these conditions in Indigenous children.

Research has started to show that geographic variation in PPH may not be simply explained by the supply of GPs. PPH may be more a reflection of social determinants and individual health factors, particularly for chronic conditions, than access to GPs (Falster M et al. 2015). PPH may, therefore, indicate areas where primary health care could be more effective or where it may be underutilised. There is a need for improvements in efforts to address the social determinants of health beyond the Health sector.

The data in this measure has illustrated the increasing rate of PPH for Indigenous Australians, indicating that primary health care is not adequately meeting the need. However, measure 1.24 Avoidable and preventable deaths has shown that deaths from avoidable causes are declining for Indigenous Australians. Further research is needed using data linkage, at smaller levels of geography across Australia. This should include measures of social determinants, and primary care utilisation to explore areas of unmet health need or underutilised health care for Indigenous Australians.

Understanding the reasons for underutilisation of primary health care at the local level is important in order to address barriers to accessing care (see measure 3.14 Access to services compared with need) such as distance, cost, availability and cultural safety.

Aboriginal Community Controlled Health Services (ACCHSs) play an essential role in providing comprehensive, appropriate and culturally safe care for Indigenous Australians. The evaluations in measure 1.24 Avoidable and preventable deaths point to mainstream services being less effective than ACCHSs. Comprehensive, accessible and well-integrated care is needed, particularly in managing chronic conditions, in order to prevent hospitalisations and death from avoidable causes. Ensuring services are accessible and appropriate is therefore important to drive reductions in PPH and avoidable deaths.

The policy context is at Policies and strategies.

References

  • Australian Institute of Health and Welfare (AIHW) 2019. Potentially preventable hospitalisations in Australia by age groups and small geographic areas, 2017–18. Canberra: AIHW.
  • AIHW 2020. Disparities in potentially preventable hospitalisations across Australia: Exploring the data. Canberra: AIHW.
  • Falster K, Banks E, Lujic S, Falster M, Lynch J, Zwi K et al. 2016. Inequalities in pediatric avoidable hospitalizations between Aboriginal and non-Aboriginal children in Australia: a population data linkage study. BMC pediatrics 16:169.
  • Falster MO, Jorm LR, Douglas KA, Blyth FM, Elliott RF & Leyland AH 2015. Sociodemographic and health characteristics, rather than primary care supply, are major drivers of geographic variation in preventable hospitalizations in Australia. Medical care 53:436.
  • Gibson O & Segal L 2009. Avoidable hospitalisation in Aboriginal and non-Aboriginal people in the Northern Territory. The Medical Journal of Australia 191:411.
  • Harrold. TC, Randall. DA, Falster. MO, Sanja L & Louisa J 2014. The Contribution of Geography to Disparities in Preventable Hospitalisations between Indigenous and Non-Indigenous Australians. PloS one.
  • Hernandez SE, Conrad DA, Marcus-Smith MS, Reed P & Watts C 2012. Patient-centered innovation in health care organizations: A conceptual framework and case study application. Health Care Management Review.
  • Li SQ, Gray N, Guthridge S, Pircher S, Wang Z & Zhao Y 2009. Avoidable mortality trends in Aboriginal and non‐Aboriginal populations in the Northern Territory, 1985‐2004. Australian & New Zealand Journal of Public Health 33:544-50.
  • Øvretveit J 2012. Summary of 'Do changes to patient-provider relationships improve quality and save money?'. London: The Health Foundation.
  • Page A, Ambrose S, Glover J & Hetzel D 2007. Atlas of Avoidable Hospitalisations in Australia: ambulatory care-sensitive conditions. Adelaide: PHIDU.
  • Passey ME, Longman JM, Johnston JJ, Jorm L, Ewald D, Morgan GG et al. 2015. Diagnosing Potentially Preventable Hospitalisations (DaPPHne): protocol for a mixed-methods data-linkage study. BMJ Open 5:e009879.
  • Zhao Y, Thomas SL, Guthridge SL & Wakerman J 2014. Better health outcomes at lower costs: the benefits of primary care utilisation for chronic disease management in remote Indigenous communities in Australia's Northern Territory. BMC health services research 14:463.

View measure data

View data visualisations, download data tables and review data sources for this measure.

Data