English proficiency and time spent in Australia since first arrival
This section reports the results of logistic regression modelling for English proficiency in combination with time since arrival in Australia and each long-term health condition. The aim is to assess whether the increased proportion of people reporting the health conditions related to time since arriving in Australia varies by level of proficiency in spoken English, and whether the social determinants of health influence the findings.
Results are presented as time since arrival stratified by each of the English proficiency groups including:
- people who spoke ‘English only’,
- people with high English proficiency (those who spoke English very well or well) and,
- people with low English proficiency (those who did not speak English well or at all).
The overall results observed indicate that:
- people who have arrived in Australia more recently who spoke ‘English only’ were less likely to report long-term health conditions in the 2021 Census, but the odds moved towards the levels observed among Australian-born people with increasing time spent in Australia.
- people who arrived in Australia more recently with high English proficiency were less likely to report long-term health conditions (with males with diabetes being the exception) than people who had been in Australia longer. The odds ratios moved towards the Australian-born levels with increasing time in Australia.
- people who arrived in Australia more recently with low English proficiency were less likely to report long-term health conditions in the 2021 Census, but the odds moved towards the Australian-born levels the longer they stayed in Australia.
In all regression models, people who first arrived in Australia 0 to 10 years ago (before the 2021 Census) were selected as the reference category, when estimating the odds ratios. Full output from all regression models is provided in the supplementary tables (Tables S6.1–S6.9).
Detailed findings from the modelling are presented below.
This section reports the results of logistic regression modelling the odds of selected long-term health conditions from a person’s time spent in Australia since first arrival for people who spoke ‘English only’. Full output from all regression models is provided in the supplementary tables (Tables S6.1–S6.9).
The overall results observed from all regression models indicate people who have arrived in Australia more recently who spoke ‘English only’ were less likely to report long-term health conditions in the 2021 Census, but the odds moved towards the levels observed among Australian-born people with increasing time spent in Australia. Detailed findings from the modelling are presented below.
1. Unadjusted models
The results from the unadjusted models showed that, compared with people who spoke ‘English only’ and had arrived in Australia in the last 10 years, the odds of reporting the reported long-term health conditions were higher for those who arrived more than 10 years ago. For people who spoke ‘English only’ at home, the odds ratios increased as time since arrival increased and was highest in those who arrived in Australia more than 20 years ago, for all reported long-term health conditions, other than asthma and mental health conditions. For reported asthma and mental health conditions, the odds ratios were highest among people born in Australia and spoke ‘English only’ at home.
2. Adjusting separately for age and social determinants of health
When the results were separately adjusted for age and the social determinants of health, controlling for age substantially changed the odds ratios for reporting all long-term health conditions other than asthma and mental health conditions, particularly for people who arrived in Australia more than 20 years and spoke English only’ at home. After adjusting for age alone, the odds ratios for people who arrived more than 20 years ago and spoke ‘English only’ at home reporting asthma and mental health conditions were slightly higher than the corresponding unadjusted odds ratios.
Adjusting separately for other social determinants of health had less effect on the odds ratios than adjusting for age alone. This impact was generally most notable when adjusting for labour force status or occupation, followed by marital status, income or housing tenure, for people who arrived in Australia more than 20 years ago. For example, the unadjusted odds ratios for reporting multimorbidity were 9.0 for males and 6.6 for females who spoke ‘English only’ and first arrived in Australia more than 20 years ago compared with those arrived in Australia up to ten years ago. These odds ratios dropped to:
- 2.3 for males and 2.5 for females after adjusting only for age
- 5.3 for males and 4.6 for females after adjusting only for labour force status
- 6.4 for males and 4.6 for females after adjusting only for marital status
- 6.5 for males and 4.9 for females after adjusting only for income.
3. Fully adjusted models
In the fully adjusted models, the odds ratios for the reported long-term health conditions in both males and females who spoke ‘English only’ were similar to those observed when adjusting only for age across all time since arrival groups, including those born in Australia. For example, when unadjusted, the odds of reporting heart disease for males who spoke ‘English only’ and first arrived in Australia more than 20 years ago was 9.9 times higher than that for those who arrived in Australia in the last 10 years and spoke ‘English only’ at home. The odds ratio was 1.5 when the odds of reporting heart disease were modelled using age and time since arrival for males who arrived in Australia more than 20 years ago. This odds ratio of 1.5 was maintained in the fully adjusted model (Table 6 below). In the fully adjusted models, for males and females who spoke ‘English only’, the odds ratios for reporting long-term health conditions increased with time since arrival and were generally higher in the Australian-born group.
| Health outcome | Odds ratio (95% CI) Australia-born | Odds ratio (95% CI) 11 to 20 years | Odds ratio (95% CI) More than 20 years |
|---|---|---|---|
1 or more of any chronic condition | Males: 1.85 (1.82–1.87) Females: 2.09 (2.06–2.11) | Males: 1.31 (1.29–1.33) Females: 1.38 (1.36–1.40) | Males: 1.58 (1.56–1.60) Females: 1.72 (1.69–1.74) |
Multimorbidity | Males: 2.65 (2.56–2.74) Females: 2.98 (2.89–3.06) | Males: 1.47 (1.41–1.53) Females: 1.59 (1.53–1.64) | Males: 2.13 (2.06–2.21) Females: 2.36 (2.29–2.42) |
Arthritis | Males: 2.00 (1.93–2.07) Females: 2.05 (1.99–2.11) | Males: 1.35 (1.29–1.40) Females: 1.39 (1.35–1.44) | Males: 1.63 (1.57–1.69) Females: 1.76 (1.71–1.82) |
Asthma | Males: 1.84 (1.80–1.89) Females: 2.16 (2.12–2.21) | Males: 1.21 (1.18–1.25) Females: 1.29 (1.26–1.32) | Males: 1.46 (1.42–1.49) Females: 1.69 (1.65–1.72) |
Diabetes | Males: 1.22 (1.18–1.27) Females: 1.34 (1.29–1.39) | Males: 1.11 (1.06–1.16) Females: 1.18 (1.12–1.23) | Males: 1.29 (1.25–1.34) Females: 1.40 (1.35–1.46) |
Heart disease | Males: 1.59 (1.52–1.66) Females: 1.93 (1.82–2.04) | Males: 1.20 (1.14–1.26) Females: 1.24 (1.16–1.33) | Males: 1.48 (1.41–1.54) Females: 1.64 (1.55–1.74) |
Mental health condition | Males: 2.35 (2.29–2.41) Females: 2.42 (2.38–2.47) | Males: 1.51 (1.46–1.55) Females: 1.50 (1.47–1.54) | Males: 1.92 (1.88–1.98) Females: 1.93 (1.89–1.97) |
Notes
- Results are from the fully-adjusted model which included the covariates, time spent in Australia, age, education, income, employment, tenure, housing suitability, remoteness, citizenship, marital status for people who spoke ‘English only’ at home. Analysis excluded overseas visitors, people who live in non-private dwellings or Migratory, offshore and shipping SA1s, non-classifiable households or Visitor only households.
- Analyses included 7,672,185 males and 7,986,177 females aged 15 and over living in Australia in occupied private dwellings on Census Night, who were not overseas visitors and provided a valid response to the 2021 Census questions on year of arrival in Australia and proficiency in spoken English (combined), age, the selected social determinants of health, and the long-term health conditions.
- The '0 to 10 years' category was selected as the reference category, when calculating the odds ratios for the levels of time spent in Australia since first arrival.
- OR (95% CI) refers to odds ratio and the 95% confidence interval.
- All odds ratios are rounded to two decimal places.
Source: AIHW analysis of PLIDA, 2021.
The overall patterns observed from all regression models for people with high English proficiency (who spoke English very well or well) were similar to those observed from the models for people who spoke ‘English only’ at home. Namely, people who arrived in Australia more recently with high English proficiency were less likely to report long-term health conditions (with males with diabetes being the exception) than people who had been in Australia longer. The odds ratios moved towards the Australian-born levels with increasing time in Australia. Detailed findings from the modelling are presented below.
1. Unadjusted models
The results from the unadjusted models showed that, compared with people with high English proficiency who arrived in Australia up to ten years ago, the odds of reporting long-term health conditions were higher than those who arrived in earlier periods, that is more than ten years ago. For people with high English proficiency, the odds ratios increased as time since arrival increased and were higher in those who arrived in Australia more than 20 years ago, for all reported long-term health conditions, other than asthma and mental health conditions. The odds of reporting asthma or mental health conditions were higher among the Australian-born population with high English proficiency.
2. Adjusting separately for age and social determinants of health
When the results were separately adjusted for age and the social determinants of health, controlling for the effects of age had the largest effect on the odds ratios for reporting long-term health conditions other than asthma and mental health conditions for the Australian-born population with high English proficiency. The effect of adjusting for age was greatest on the odds of those with high English proficiency who arrived in Australia more than 20 years ago reporting long-term health conditions. The effect was small for those who spent 11 to 20 years in Australia. As was seen for people who spoke ‘English only’, the impact of adjusting for other social determinants of health on the odds ratios was small. However, changes in the odds ratios for people who arrived in Australia more than 20 years ago and were highly proficient in spoken English were observed when adjusting for the effects of labour force status, occupation, marital status, income, or housing tenure.
3. Fully adjusted models
After adjusting for the effects of age and the social determinants of health, for males and females with high English proficiency, the odds ratios for all categories of time since arrival in Australia were consistent with the corresponding age-adjusted odds ratios. In the fully adjusted models, the odds ratios for the most reported long-term health conditions for the Australian-born population with high English proficiency were generally similar to or higher than that for those with high English proficiency and who first arrived in Australia more than 20 years ago. For example, among males with high English proficiency, the odds ratio for reported diabetes when compared with those who arrived in the last ten years was highest in those who first arrived in Australia 11 to 20 years ago (1.3) and lowest in those born in Australia (0.9) (Table 7).
Health outcome | Odds ratio (95% CI) Australia-born | Odds ratio (95% CI) 11 to 20 years | Odds ratio (95% CI) More than 20 years |
|---|---|---|---|
1 or more of any chronic condition | Males: 3.04 (3.00–3.09) Females: 3.50 (3.46–3.55) | Males: 1.69 (1.66–1.71) Females: 1.73 (1.71–1.75) | Males: 2.73 (2.70–2.77) Females: 3.35 (3.30–3.39) |
Multimorbidity | Males: 5.29 (5.08–5.50) Females: 7.29 (7.04–7.56) | Males: 2.08 (1.99–2.17) Females: 2.31 (2.22–2.40) | Males: 4.46 (4.30–4.63) Females: 6.71 (6.49–6.93) |
Arthritis | Males: 3.57 (3.42–3.73) Females: 3.44 (3.32–3.55) | Males: 1.56 (1.49–1.63) Females: 1.67 (1.61–1.73) | Males: 2.63 (2.53–2.74) Females: 3.15 (3.05–3.25) |
Asthma | Males: 5.74 (5.58–5.89) Females: 4.92 (4.80–5.03) | Males: 1.71 (1.67–1.77) Females: 1.71 (1.66–1.75) | Males: 3.27 (3.19–3.36) Females: 3.09 (3.02–3.16) |
Diabetes | Males: 0.86 (0.83–0.88) Females: 1.26 (1.22–1.30) | Males: 1.34 (1.31–1.38) Females: 1.43 (1.39–1.48) | Males: 1.23 (1.20–1.26) Females: 1.75 (1.70–1.80) |
Heart disease | Males: 1.80 (1.73–1.88) Females: 2.52 (2.38–2.67) | Males: 1.36 (1.30–1.42) Females: 1.40 (1.32–1.49) | Males: 1.41 (1.36–1.46) Females: 1.95 (1.85–2.05) |
Mental health condition | Males: 6.13 (5.93–6.33) Females: 5.65 (5.51–5.78) | Males: 1.88 (1.81–1.95) Females 1.91 (1.86–1.96) | Males: 4.73 (4.57–4.88) Females: 5.02 (4.91–5.15) |
Notes
- Results are from the fully-adjusted model which included the covariates, time spent in Australia, age, education, income, employment, tenure, housing suitability, remoteness, citizenship, marital status in people who spoke English very well or well (high English proficiency). Analysis excluded overseas visitors, people who live in non-private dwellings or Migratory, offshore and shipping SA1s, non-classifiable households or Visitor only households.
- Analyses included 7,672,185 males and 7,986,177 females aged 15 and over living in Australia in occupied private dwellings on Census Night, who were not overseas visitors and provided a valid response to the 2021 Census questions on year of arrival in Australia and proficiency in spoken English (combined), age, the selected social determinants of health, and the long-term health conditions.
- The '0 to 10 years' category was selected as the reference category, when calculating the odds ratios for the levels of time spent in Australia since first arrival.
- OR (95% CI) refers to odds ratio and the 95% confidence interval.
- All odds ratios are rounded to two decimal places.
Source: AIHW analysis of PLIDA, 2021.
The overall patterns observed from all regression models for people with low English proficiency (who did not speak English well or at all) was similar to those observed from the models for people who spoke ‘English only’ at home or those with high English proficiency. Namely, people with low English proficiency who arrived in Australia more recently were less likely to report long-term health conditions in the 2021 Census, but the odds moved towards the Australian-born levels the longer they stayed in Australia. People with low English proficiency who had been in Australia more than 20 years were more likely to report having diabetes than Australian-born people with low English proficiency, after adjusting for the effects of age, sex and social determinants of health. Detailed findings from the modelling are presented below.
1. Unadjusted models
The results from the unadjusted models showed that people with low English proficiency who arrived in Australia in the last ten years were less likely to report long-term health conditions than those with low English proficiency who arrived more than ten years ago. This result was observed for all reported long-term health conditions other than asthma and mental health conditions. For reported asthma and mental health conditions, the odds ratios were highest among the Australian-born population with low English proficiency.
2. Adjusting separately for age and social determinants of health
Adjusting for age increased the odds ratios for the reported long-term health conditions other than asthma and mental health conditions for the Australian-born population with low English proficiency, compared with those who arrived in Australia in the last ten years. The effect of adjusting for age on the odds ratios was strongest for those with low English proficiency and who arrived in Australia more than 20 years ago, and was generally smaller for those who arrived 11 to 20 years ago, relative to the unadjusted odds ratios. As was seen for people who spoke ‘English only’ or for those with high English proficiency, the impact of adjusting for other social determinants of health on the odds ratios were much less consistent, and generally most notable when adjusting for employment or occupation, marital status, income or housing tenure, particularly for people who arrived in Australia more than 20 years ago.
3. Fully adjusted models
After adjusting for the effects of age and the social determinants of health, for both males and females with low English proficiency, the odds ratios for all categories of time since arrival in Australia remained similar to the age adjusted odds ratios. In the fully adjusted models, the odds ratios for reporting the long-term health conditions for the Australian-born population with low English proficiency was mostly higher than for those who first arrived in Australia more than 20 years ago and lowest for those who arrived 11 to 20 years ago (Table 8 below). The exceptions were:
- females who arrived more than 20 years ago were as likely to report arthritis as Australian-born females
- females who arrived more than 20 years ago were more likely to report a mental health condition than females born in Australia
- males and females who arrived in Australia more than 20 years ago were likely to report diabetes than people born in Australia.
Health outcome | Odds ratio (95% CI) Australia-born | Odds ratio (95% CI) 11 to 20 years | Odds ratio (95% CI) More than 20 years ago |
|---|---|---|---|
1 or more of any chronic condition | Males: 4.53 (4.32–4.75) Females: 3.56 (3.38–3.76) | Males: 1.45 (1.41–1.50) Females: 1.49 (1.46–1.53) | Males: 2.25 (2.20–2.30) Females: 2.66 (2.61–2.71) |
Multimorbidity | Males: 3.77 (3.48–4.09) Females: 3.51 (3.22–3.82) | Males: 1.47 (1.39–1.55) Females: 1.76 (1.69–1.83) | Males: 2.56 (2.47–2.67) Females: 3.46 (3.35–3.57) |
Arthritis | Males: 4.44 (4.01–4.92) Females: 2.79 (2.53–3.08) | Males: 1.41 (1.32–1.52) Females: 1.68 (1.62–1.75) | Males: 2.57 (2.45–2.71) Females: 2.88 (2.79–2.98) |
Asthma | Males: 6.90 (6.31–7.56) Females: 5.32 (4.85–5.83) | Males: 1.93 (1.78–2.09) Females: 1.97 (1.86–2.09) | Males: 4.61 (4.33–4.90) Females: 3.79 (3.62–3.97) |
Diabetes | Males: 1.23 (1.13–1.35) Females: 1.25 (1.13–1.37) | Males: 1.17 (1.12–1.22) Females: 1.22 (1.18–1.26) | Males: 1.40 (1.35–1.44) Females: 1.63 (1.58–1.67) |
Heart disease | Males: 2.17 (1.95–2.42) Females: 1.75 (1.53–2.01) | Males: 1.15 (1.09–1.22) Females: 1.18 (1.12–1.25) | Males: 1.30 (1.25–1.35) Females: 1.30 (1.25–1.35) |
Mental health condition | Males: 6.02 (5.59–6.49) Females: 4.27 (3.95–4.62) | Males: 1.76 (1.65–1.89) Females: 1.91 (1.82–2.00) | Males: 4.01 (3.80–4.22) Females: 5.02 (4.84–5.22) |
Notes
- Results are from the fully-adjusted model which included the covariates, time spent in Australia, age, education, income, employment, tenure, housing suitability, remoteness, citizenship, marital status in people who do not speak English well or at all (low English proficiency). Analysis excluded overseas visitors, people who live in non-private dwellings or Migratory, offshore and shipping SA1s, non-classifiable households or Visitor only households.
- Analyses included 7,672,185 males and 7,986,177 females aged 15 and over living in Australia in occupied private dwellings on Census Night, who were not overseas visitors and provided a valid response to the 2021 Census questions on year of arrival in Australia and proficiency in spoken English (combined), age, the selected social determinants of health, and the long-term health conditions.
- The '0 to 10 years' category was selected as the reference category, when calculating the odds ratios for the levels of time spent in Australia since first arrival.
- OR (95% CI) refers to odds ratio and the 95% confidence interval.
- All odds ratios are rounded to two decimal places.
Source: AIHW analysis of PLIDA, 2021.