A systematic review of published interventions for primary and secondary prevention of ischaemic heart disease (IHD) in rural populations of Australia

We sought to identify studies within the published peer-review literature that were focussed on rural populations and that aimed to prevent or reduce IHD burden or risk factors. This systematic review was registered with Prospero, (number CRD42016033431). The term ‘rural’ used throughout this paper, refers to all areas classified as being outside of major cities of Australia, by the Australian Bureau of Statistics, Accessibility Remoteness Index of Australia (ARIA) [16]. ARIA has five categories of remoteness, which are, in increasing order of remoteness major cities, inner regional, outer regional, remote and very remote. These definitions are based on remoteness scores derived from relative road distance to population localities and services [16].

Search terms used were related to four major topics including, ‘ischaemic heart disease’, ‘rural’, ‘intervention or prevention,’ and ‘Australia’. The search was conducted in November and December of 2015. The six databases included in the search were CINAHL, Medline, Academic Search complete, Rural and Remote Health Database, Health and Society Database and Embase. An additional hand search was undertaken of reference lists from included studies.

The following were the Inclusion criteria:

The lead researcher (LA) reviewed all results from the six databases, removed duplicates and screened all results based on titles and, abstracts against the review criteria (see Fig. 1). Two additional researchers (KP & JJ) each screened a 50 % sample of titles and abstracts as a second reviewer. Any discrepancies were identified and resolved by consensus among the three researchers producing a list of papers for full text assessment for eligibility against the review criteria. Reference lists of all full texts were then searched for additional potentially eligible studies.

Data were extracted into a spreadsheet from the full texts by the lead researcher. The details collected included the publication details of the study, years of intervention, intervention type, follow up period, outcome measures (such as changes in clinical and modifiable risk factors), results and authors’ conclusions. Each intervention was then then categorised as either primary or secondary prevention, or both, and by the broad type of intervention (e.g. delivery through initial screening/education/exercise or whole community programs). The studies were then synthesized into a narrative analysis, with a focus on changes in outcome measures. We applied a narrative analysis because quantitative meta-analysis was deemed inappropriate due to the small sample size and heterogeneity of the interventions returned by the search strategy.

Two researchers independently assessed each study using the Cochrane Collaboration’s tool for assessing risk of bias [16]. The tool is used to assess the risk of bias within each individual study based on five different types of bias including: selection bias (randomisation of participants), performance bias (blinding of participants), detection bias (blinding of outcome measures), attrition bias (incomplete outcome data) and complete reporting. Studies were rated as either high, low or unclear risk against each of the criteria. The Cochrane tool does not use a total score to assess overall risk of bias, so each type of bias is assessed individually.

Six primary prevention studies were identified; five in inner and outer regional areas [18, 20‐23], and one in a remote area [24]. Two reported on interventions in ATSI populations [22, 24]. Most studies [17, 20, 21, 24] were published more than 8 years ago. The intervention activities included an exercise program for a high risk population [23], cardiac rehabilitation [17], a full community intervention [19], and five risk factor screening and/or subsequent education or treatment programs within small communities [18, 20‐22, 24]. Generally, the studies reviewed showed that IHD prevention efforts in rural communities are feasible and were effective in either reducing one or more risk factors, or IHD mortality, however the studies were limited by short follow up periods, small population numbers and a lack of inclusion of control groups in study designs.

Kerr et al. [23] measured the effect of a 12-month exercise program on IHD risk factors in a population of war veterans living in regional Queensland (n = 164), without a control group. The main outcome measures included measurement of heart rate (HR), blood pressure (BP), skinfold and girth measurements, exercise heart rate response and estimated aerobic capacity. The results were used to determine if the program could be effective in this rural, high risk population. The study showed that an organised exercise group could be feasible in a rural setting for high risk clients, with positive effects shown for resting HR (−4 bpm), diastolic (−6.4 mmHg) and systolic BP (−8.4 mmHg) by the end of the program (p = <0.05). Weight was unchanged at 12 months, however there were some improvements in body composition. The generalisability of results from this study is limited because only 54 % of participants completed the final 12 month follow up assessment.

Three studies [18, 20, 21] used primary risk factor screening as the start point of the intervention, screening patients’ BP, cholesterol and BMI, and when compared to baseline, showed that these types of programs are potentially feasible in rural areas. Krass and colleagues [20], assessed the impact of a pharmacy screening and health promotion program (n = 389) on the risk of IHD and stroke in two towns in regional New South Wales. The health promotion program included individual education on lifestyle improvements including diet, exercise and smoking cessation advice. After three months, significant changes were observed in mean total cholesterol for both towns (−0.26 mmol/l, 95 % CI 0.10-0.42, p = <0.003), while BP was reduced in participants from one town (−10.5 mmHg, 95 % CI 4.0-16.9, p = 0.012), changes in physical activity and smoking prevalence were reported, with increases in activity reported by participants at 3 months. The authors noted that there was little change in smoking prevalence, and attributed this to the relatively short period of intervention and follow up of this study design, which with no control group comparison, would also make it difficult to draw concrete conclusions on the overall effectiveness of this intervention. Carrington and Stewart [18], describe a similar, yet nurse-led intervention in regional Victoria in which 530 self-selecting patients were screened for cardiovascular risk, then provided with counselling and advice tailored to their risk level [18]. Just over 60 % of the patients (n = 326) had clinically significant improvements in risk factor levels at 6 months post-intervention, with BP, total cholesterol and weight all decreasing from baseline levels, yet these results were not compared to a control group. No further follow up was undertaken after 6 months post-intervention, making it difficult to determine either the sustainability of this design or the long term impact on risk factors in the rural community studied. A third screening study published in 2001 [21], without inclusion of controls, assessed the effect of a one-time screening session delivered by the mobile ‘Heart Bus’ on self-reported behaviour changes in inner regional Queensland. The study followed up 135 participants and found that 76 % of participants self-reported they had made positive changes to their diet and exercise behaviours 6 months later. The major limitations of this study include the small sample design, lack of control and the self-reported nature of all of the outcome measures.

Two studies assessed IHD interventions in ATSI populations only [22, 24] and showed that IHD primary prevention efforts in these populations have the potential to be effective. One study was a clinical audit of cardiac prevention and screening services [22], and the other was an assessment of a diet and exercise program with comparison made to a self-selected control group [24]. Burgess et al. [22] analysed the effectiveness of cardiac prevention services through clinical audits every three months of cardiovascular risk assessments, and level of pharmaceutical prescription delivered through primary health care services over 2 years. The study focussed on results from 2586 participants, who were identified to have a five year CVD risk of 16 % or greater. Blood pressure medication was prescribed for 67 % of participants and lipid lowering medications for 55 %, with clinical follow-up every three months to assess if target levels were achieved. By the end of the two year evaluation, the number of participants who achieved clinical targets for BP was 1366 (56 %) while 989 (40 %) reached targets for cholesterol, however changes in the proportion of participants reaching targets did not change significantly over time for either outcome. Rowley and colleagues [24] assessed the effectiveness of a primary health care service providing diet and/or exercise education and support in a small rural ATSI population. The study included an intervention group (n = 32) and a self-selected control group (n = 17). There were significant differences (p = 0.03) observed between the intervention and control arms, for mean change in 2 h plasma glucose, and triglyceride levels at two years post-intervention. There were also changes in dietary behaviour and physical activity in the intervention group, however these did not appear to be significant when compared to controls. Response rates ast the two year follow up were low for younger participants aged 15–34, with 43 % responding, compared to 80 % for those aged 35 and over, however results shown here are clearly limited by the small study sample.

Only one secondary prevention intervention [17] was identified. The study evaluated a 7-week bi-weekly education and exercise cardiac rehabilitation program in a rural area of Western Australia (Heart Smart), and included 203 participants with a current CVD diagnosis. The evaluation compared quality of life and cardiac knowledge scores of these participants with to 159 non-participants (who were eligible, but did not wish to participate in the intervention). All follow up data, including clinical measures were self-reported. After 6 months, the intervention group had reportedly increased their physical activity (p < =0.001), and reduced their weight (−0.5 kg, p < 0.05) from baseline. Higher quality of life and cardiac knowledge scores were observed for the intervention group, however the non-intervention group had a low response rate to the follow up survey (42 %). Self-reported cholesterol levels were 3.6 mmol/L at baseline, and these reduced to 2.8 mmol/L by 6 months post follow up, however no changes were observed for self-reported BP. The authors did not specify if these measures were taken by the same medical clinic, or if any clinical documentation was collected with self-report results.

Higginbotham et al. [19] was the only study to include both a primary and secondary prevention program. The Coalfields Healthy Heartbeat was a 10-year community intervention, which employed multiple strategies and included health promotion and awareness advertising, mobilisation of community resources, school health-promotion programs, exercise, cooking and education groups, and a cardiac rehabilitation program. The remainder of the Hunter Valley region served as a comparison population for the program evaluation. The authors found a larger reduction in fatal AMI cases in the intervention area relative to the comparison population over a 9-year monitoring period [19], however there was no reduction in non-fatal AMI rates. There were no significant differences in risk factors changes between the intervention and non-intervention areas.

Table 1: Characteristics of prevention programs aimed at reducing IHD burden in rural Australia.

Due to the lack of randomization used in the design of all studies and minimal use of control groups, all of the included studies were assessed to have a high risk of bias for the selection, performance and detection criteria of the Cochrane tool, and low or unclear risk for attrition and reporting bias criteria. Only one study reported using a behavioural science theory in the design of their study, identifying the health promotion and behavioural change models in the design of their follow up questionnaire [16]. There was also a lack of detail around the length of the intervention, and follow up periods for 4 of the 8 studies, and generally follow up periods were small.

A major finding from this review is the lack of high quality studies assessing the effectiveness of interventions to reduce IHD burden in rural areas of Australia. Priority must be placed on action-orientated prevention research in rural communities, with a focus on methodologies such as community-based participatory research to empower communities, create tailored strategies and address inequalities [35] Future research must have specific focus on rigorous methodology, including comparison to control groups, more comprehensive measurements of changes in modifiable risk factors, and longer follow-up time frames in order to assess sustainability of such programs in the rural context. . There is significant potential for countries with comparable rural health inequalities (for example, geographically large, high income countries such as Australia, Canada and the US) to learn from international examples [7, 9, 26, 27] and create a stronger body of evidence for the prevention of IHD among rural populations globally international examples and create a stronger body of evidence for the prevention of IHD among rural populations globally.