Table 2 Peer-reviewed journals

1. Prevalence or incidence, either population-based or within clinical groups or clinical service settings

Coory et al. (2005) [23] Original article

Queensland patients hospitalised with MI in public sector N=14683, Indigenous=558

Design: Retrospective cohort study (I year follow-up)

HF more commonly a concurrent co-morbidity among Indigenous than non-Indigenous identified patients (age-adjusted RR 1.64; CI 1.35-2.00)

• HF one of several co-morbidities assessed; not an endpoint of study.

Data source: linked hospital data

• No adjustment for administrative under-identification of Indigenous status

Period: Admitted for MI 1998-2002

Outcome: Revascularisation

Katzenellenbogen et al. (2012) [24] Original article

All WA patients hospitalised with non-fatal first-ever MI N=7480, Indigenous=532

Design: retrospective cohort study (2 year follow-up)

Data source: Linked administrative data

HF more commonly a concurrent or past co-morbidity among Indigenous than non-Indigenous identified patients (males: 17% vs. 13%; p=0.018; females: 31% vs. 22%; p=0.003)

• HF one of several co-morbidities assessed; not an endpoint of study.

• Indigenous status based on ever-identification in hospital or death records

Period: Admitted Jan 2000-Dec 2004

Outcome: recurrent MI/CVD death

• Crude HF prevalence, no age adjustment thus underestimate of disparity

McGrady et al. (2012) [25] Original Article

Consenting Aboriginal adults (>18 years) residing in one of six Central Australian communities, (Alice Springs, Town Camp or remote) N=436 (mean age 44 years)

Design: cross-sectional clinical survey

HF detected in 5.3% (CI 3.2-7.5%); 65% of these no pre-existing HF diagnosis.

ALVD present in a further 13% (CI 9.4-15.7%)

• Population-based study designed specifically to assess epidemiology of HF and risk factors among Central Australian Aboriginal adults.

Data source: Clinical history & examination, anthropometry, echocardiogram, biochemistry

• Volunteer participants – representativeness uncertain

Period: Assessment May 2008 - Nov 2009

Outcome: presence of HF or ALVD

• No non-Indigenous comparison group

2. Aetiology, risk factors, clinical presentation and pathophysiology

Einsiedel et al. (2012) [26] Original article

Indigenous adults (n=89) admitted to a single general hospital with bronchiectasis and known HTLV-1 serological status

Design: Retrospective cohort study

Data source: Hospital records

HF (35% versus 11%; p=0.013) more common in human HTLV-1 seropositive than HTLV-1 seronegative subjects.

• HF one of several complications assessed

• Indigenous status identified from medical records

• Comparison not age-adjusted,

• however mean age essentially identical in both groups.

• Population restricted

Period: Jan 2000-Dec 2006

Outcome: bronchiectasis outcomes

• to Central Australia, majority ‘remote’ residence (61%)

Greaney (2010) [27] Original article

Indigenous patients with symptomatic HF referred to a heart rehabilitation program in Far North Queensland (n=101)

Design: Descriptive clinical survey

57% had normal systolic function

• Indigenous status identification not explicit

Data source: Hospital and echocardiograph records

• No non-Indigenous comparison group

Period: April 2005-Jan 2008

Outcome: Proportion with normal systolic function

McGrady et al. (2012) [25] Original article

Aboriginal adult volunteers (>18 years) residing in six Central Australian communities, N=436 (mean age 44 years)

Design: cross-sectional clinical survey

Age & sex-adjusted odds ratio for HF: CAD (9.6, p<0.001)

• Population-based study designed specifically to assess epidemiology of HF and risk factors among Central Australian Aboriginal adults.

Data source: Clinical history & examination, anthropometry, echocardiogram, biochemistry

DM (5.4, p=0.002)

HT (4.8, p=0.006)

Period: Assessment May 2008 - Nov 2009

Outcome: presence of HF or ALVD

Obesity (2.9, p=0.022)

• Volunteer participants – representativeness uncertain

RHD history (5.6, p=0.001) 39% of HF cases had preserved ejection fraction

• No non-Indigenous comparison group

3. Co-morbidities

McGrady et al. (2012) [25] Original Article

Aboriginal adult volunteers (>18 years) from six Central Australian communities, N=436 (mean age 44 years)

Design: cross-sectional clinical survey

Crude prevalence in HF cases: Diabetes 78% Hypertension 78% CAD 39% ARF/RHD 26%

• Population-based study designed specifically to assess epidemiology of HF and risk factors among Central Australian Aboriginal adults.

Data source: Clinical history & examination, anthropometry, echocardiogram, biochemistry

Period: Assessment May 2008 - Nov 2009

Outcome: presence of HF or ALVD

• Volunteer participants – representativeness uncertain

• No non-Indigenous comparison group

4. Mortality & survival

Brown (2010) [28] Original article

Patients admitted to two NT hospitals with ACS (n=214 Indigenous, 278 non-Indigenous)

Design: Retrospective audit

Frequency of death attributed to HF was similar in both Indigenous (approx. 2.2%) and non-Indigenous (approx. 2.0%).

• Indigenous identification is relatively good in NT administrative records; no additional effort to improve identification

• Sample size relatively small; Only 2 hospitals in sample

Data source: Hospital records

Period: Admissions Jan 2001-Dec 2002; 2 year follow-up

Outcome: All-cause and CVD deaths

Carapetis et al. (1999) [29] Original article

80 consecutive patients (70 Indigenous) with surgical valve replacement for RHD

Design: Cohort study

29 late deaths, 27 attributed to RHD, 12 of these were HF deaths (plus 1 due to ‘HF and pneumonia’)

• No comparison group.

Data source: Hospital records

• Long calendar period of case acquisition (1964–1996) limits contemporary interpretability of prognosis.

Period: Surgery 1964–1996; all patients followed up until Mar 1997

Outcome: Deaths

Katzenellenbogen et al. (2011) [24] Original article

See Part 1. above

See Part 1. above

HF as a co-morbidity independently associated with about double the risk of composite outcome (recurrent AMI or death) in both Indigenous and non-Indigenous subjects

• HF was one of a number of demographic and co-morbidity variables in the model

5. Quality of life

Nil

6. Therapeutic interventions

Nil

7. Health service utilisation (including medication adherence, outpatient attendances, hospitalisations, cardiac rehabilitation)

Bolton et al. (2011) [30] Conference abstract

Patients referred to inner-suburban AMS-controlled cardiology clinic

Design: Cross sectional survey

2 of 68 patients (3%) referred to an inner-suburban AMS-controlled cardiology clinic had HF.

• Encounter proportions without comparison group difficult to interpret

Data source: Clinic records

Period: July 2009-2011

Outcome: Attendance, encounter proportions

• Uncertain generalisability to Australian Indigenous population.

Thomas et al. (1998) [31] Original article

Primary care (AHW and/or doctor) encounters in AMS clinic in Darwin, NT.

Design: Cross sectional survey

Proportion of encounters involving HF: 3.4% (95% CI 1.9-4.9) compared with 1.6% in national comparison data (AMTS)

• Comparison of encounter proportions difficult to interpret

Data source: Clinic records

• Uncertain generalisability to Australian Indigenous population.

Period: 2 separate study weeks 6 months apart: in Darwin’s wet season (Feb 1994) & in dry season (Aug 1994)

Outcome: Encounter proportions

8. Health service delivery issues (including needs, access and barriers)

Clark et al. (2007) [32] Original article

(sample n/a) Period: Jan 2004-Dec 2005

Design: Cross sectional geomapping survey

Highest prevalence of HF in areas with people aged >65 years and higher proportions of Indigenous people.

• No direct measure of HF prevalence (based on international prevalence data).

Data source: Census data, international prevalence estimates

• Indigenous population distribution derived from Census data.

AIHW-derived Indigenous HF prevalence estimates.

• Indigenous:non-Indigenous HF prevalence ratio estimated from AIHW HF mortality data.

Outcome: Indirect measure of access to CR services

Geographical inequity in provision of HF specialist management programs, with limited access in rural areas.

Aspin et al. (2012) [32] Original article

19 Indigenous subjects (age range 34–70) from Western Sydney or Aust Capital Territory: 11 had HF, with/without co-existing diabetes and/or COPD.

Design: Interviews with Aboriginal people with chronic disease, recruited via Aboriginal health services

Negative influences were poor access to culturally appropriate health services, dislocation from cultural support systems, racism, poor communication with health professionals and economic hardship Positive influences were strength drawn from being part of the Aboriginal community, regular ongoing access to primary care and a supportive family network

• Findings not specific to HF but reflect issues related to chronic disease care of HF prevalence (based on international prevalence data).

Data source: n.a.

• Indigenous population distribution derived from Census data.

Period: Jan 2004-Dec 2005

Outcome: Description of barriers and facilitators of access to care and support

• Indigenous: non-Indigenous HF prevalence ratio estimated from AIHW HF mortality data.

9. Costs related to HF diagnosis and care

Nil