Sociodemographic and clinical factors affecting advance care planning: results from a large community cohort in New South Wales, Australia
E. Yang A , A. Kabir B , J. Rhee C , C. O’Callaghan
D * and M. Barr D
A
B
C
D
Abstract
The ageing population and increasing chronic illness pose significant healthcare challenges, including care late in life. Advance care planning (ACP) is an ongoing process of making decisions regarding future health care for patients. This process can include formal completion of an advance care directive (ACD), which is a legally binding document. ACP can improve patient outcomes and satisfaction, but rates are low across Australia. This study assessed the sociodemographic and clinical predictors of individuals engaging with ACP and ACDs using data from the 45 and Up Study cohort in New South Wales.
A cross-sectional cohort study of 28,626 people responded to ACP-related questions in the wave 2 questionnaire of the Sax Institute’s 45 and Up Study. ACP completion was recorded if people responded ‘yes’ to any of the three ACP questions, and ACD was recorded if they responded ‘yes’ to the ACD question. Poisson regression modelling was used to estimate the prevalence ratio and the 95%CI.
A total of 28,626 people completed the ACP- and/or ACD-related questions, of whom 17,458 (61%) completed ACP and 3744 (13.1%) completed ACD. The predictors associated with an increased likelihood of ACP and ACD completion included having a will, advancing age, being female, having private health insurance, not currently working, and having one or more self-reported medical conditions. Predictors unique to increased overall ACP completion included having a university degree or higher, being married and having a health care concession card (Health Care Card). Being a carer increased ACD rates, whereas being married or in a de facto relationship decreased ACD completion.
These findings could inform interventions aimed at improving ACP uptake by identifying groups that engage less in ACP and provide a basis for future research.
Keywords: advance care planning, advance care directive, associations, cohort, chronic care, survey.
Introduction
Australia’s ageing population and increasing incidence of chronic illness pose significant healthcare challenges, particularly regarding end-of-life care. Patients and carers are placing increasing value on the provision of holistic care in line with patient wishes.1 Advance care planning (ACP) aids in this process, involving decisions about future health care for patients in conjunction with clinicians and family members to safeguard choices when decision-making capacity may be reduced.2,3 The process of ACP can include: (i) the appointment of a substitute decision-maker (SDM), known as an enduring guardian or enduring power of attorney in New South Wales (NSW) and/or (ii) the completion of an advance care directive (ACD), a legally binding document of future care preferences.4,5 Informal documentation by individuals or on behalf of someone else is also included in ACP.6 In some Australian jurisdictions, the use of ACDs is regulated by state-based legislation, but in NSW they can be written without a specific form, unwitnessed, and still be enforceable under the common law.7
The benefits of ACP are widely reported in the literature, improving outcomes and satisfaction for patients, relatives and health professionals.2,8–13 Engaging in ACP may improve compliance with patients’ end-of-life wishes, decrease caregiver burden, and reduce feelings of depression and anxiety among relatives.8,12 ACP completion rates are still low across Australia, varying by location and healthcare setting.6,14,15 Population-based surveys estimate ACP prevalence at 30%, with higher completion rates in the residential aged care facility (RACF) context (48%) but much lower rates in general practice (3%).6 Similarly low completion rates exist internationally.16,17
There are communication challenges to ACP for both patients and healthcare professionals. Both groups generally have a poor understanding of the role of ACP, with the process appearing inaccessible and complex.4,18 Healthcare practitioners often find it challenging to discuss end-of-life care with patients due to a lack of knowledge and concerns about the timing and sensitivity of the topic.19 To improve the implementation of ACP, state policies, community awareness, education of health professionals, and funding for ACP services need to be addressed.20
Some studies have explored the predictive factors for ACP completion in Australia such as age, gender, marriage status and health status, having a will and residential setting.6,21–24 There is a consensus on the need to better understand the factors that deter or promote ACP completion, suggesting that interventions should be based on these findings.25 Thus, our study aimed to assess the sociodemographic and clinical predictors of individuals engaging with ACP using the 45 Up Study cohort dataset.26 The findings may identify population groups that engage less in ACP discussions and provide a basis for future targeted interventions and research.
Methods
Study design and population
A cross-sectional study was conducted among Australian residents aged 45 years and over in NSW, Australia. In the 45 and Up Study, 267,357 individuals were randomly sampled from the Services Australia Medicare enrolment database between 2005 and 2009 with oversampling of people aged 80 years and over, and those living in rural and remote areas. Approximately 19% of those invited (accounting for ≈11% of the NSW population) joined the study. Participants consented to self-completion of the questionnaires and to the linking of the survey data to administrative health records.27,28 Participants in this study were a subset of those enrolled in the Sax Institute’s 45 and Up Study who completed both the baseline survey and the 2014/2015 follow-up questionnaires, which included three specific questions regarding ACP.27 This study was approved by NSW Population and Health Services Research Ethics Committee (reference number 2016/06/642). The 45 and Up Study was approved by the University of New South Wales Human Research Ethics Committee. Participants gave informed consent to participate in the study before taking part.
Ascertainment of advance care planning
Study participants were categorised as having completed ACP if they answered ‘yes’ to any of the following questions from the 45 and Up Study follow-up questionnaire: (1) ‘Have you discussed your wishes for your future health care with someone close to you?’, (2) ‘Have you legally nominated a person to make health decisions for you if you lose capacity to do this for yourself (e.g. nominating an enduring power of attorney)?’ and (3) ‘Have you written down your wishes for your future health care in a document such as an ACD?’. ACD was defined if they answered ‘yes’ only to question (3).
Data analysis
The covariates associated with ACP used in this analysis included demographics, chronic conditions, psychological distress, physical functioning, quality of life, carer status and if they had made a will. Variables such as highest education, born in Australia, and language spoken at home, were sourced from the baseline survey questionnaire, and all other variables were from the first follow-up questionnaire. Age was calculated at the first follow-up interview date. The psychological distress index was calculated based on the Kessler Psychological Distress Scale (K10) using 10 indicators ranging between 10 and 50, and then classified into low (<22) and high (≥22) scores.29 The physical functioning score was calculated using 10 items of the 36-item short-form (SF-36) and categorised into no limitation (score of 100), minor limitation (score 90–<100), moderate limitation (score 60–<90) and severe limitation (score 0–<60).30
The percentage of people completing ACP or ACD was calculated for each category of the variables (Supplementary Table S1). Crude and adjusted prevalence ratios (PRs) with 95% confidence intervals (CIs) were estimated using simple and multiple Poisson regression modelling to measure the association between ACP and variables. All variables were included in the multiple Poisson regression model to estimate the adjusted PRs. This method was also used to separately analyse the relationship between variables and formal ACD completion rates.
We conducted a sensitivity analysis removing the 45 and Up Study question, ‘Have you legally nominated a person to make health decisions if you lose capacity?’ from the ACP definition to examine its impact on ACP completion rates and factors influencing the ACP completion. This was undertaken because the question is quite broad and does not necessarily indicate that enduring power of attorney documentation has been completed.
Results
Assessing advance care planning
The first set of follow-up survey questionnaires were completed by 139,487 people between 2012 and 2015. The ACP questions were included only in the 2014/2015 versions of the survey. A total of 28,626 participants completed the 2014/2015 surveys and responded to all of the ACP questions. Of these, 13,715 (47.9%) discussed their future healthcare wishes with someone else, 11,418 (39.9%) had legally nominated a person to make healthcare decisions for them if they lost their capacity (e.g. enduring power of attorney), 3744 (13.1%) documented an ACD, and 17,458 (61.0%) completed any of these (Table 1).
| Question | n | % | ||
|---|---|---|---|---|
| 1 | Discussed wishes with someone else | 13,715 | 47.9 | |
| 2 | Nominated a person to make health decisions if lose capacity | 11,418 | 39.9 | |
| 3 | Advance care directive (ACD) | 3744 | 13.1 | |
| 1,2,3 | Any of these | 17,458 | 61.0 | |
| 1,2,3 | Advance care planning (ACP) | 17,458 | 61.0 | |
| 3 | Advance care directive (ACD) | 3744 | 13.1 |
Factors associated with advance care planning completion (any of these)
Sociodemographic factors such as age, gender, education and working status associated with ACP and ACD are listed in Tables 2 and 3. Rates of ACP completion increased with increasing age. Compared with the 49–59 year age group, the likelihood of completing ACP increased by 16% for those aged 60–74 years (adj. PR: 1.16; 95%CI: 1.10–1.21), by 30% for those aged 75–84 years (adj. PR: 1.30; 95%CI: 1.21–1.38), and by 45% for people aged 85+ (adj. PR: 1.45; 95%CI: 1.30–1.60). Similarly, ACD completion rates also increased with age, with 6.4% in the 49–59 year age group and 29.9% in the 85+ age group, following a similar trend [adj. PR: 1.53 (95%CI: 1.35– 1.71) for 60–74 years, 2.46 (95%CI: 2.11–2.82) for 75–84 years, and 3.15 (95%CI: 2.56–3.79) for 85+ years]. Women had higher ACP rates (63.1% vs 58.8%; adj. PR: 1.07; 95%CI: 1.03–1.11) and higher rates of ACD completion (14.6% vs 11.5%; adj. PR: 1:19; 95%CI: 1.09–1.28) than men. Individuals with university degrees or higher had higher ACP rates (62.4%) compared with those without a school certificate or other qualifications (58.4%; adj. PR: 1.11; 95%CI: 1.02–1.20). ACD completion rates were lower, with 14.0% and 15.1%, respectively, for those with versus those without higher education (adj. PR: 1.27; 95%CI: 1.07–1.51). Those not working had higher ACP and ACD completion rates compared with part-time and full-time workers. Completion rates for ACP were 67.9% for those not working, 55.6% for part-time, and 48.9% for full-time workers [adj. PR: 0:90 (95%CI: 0.86–0.95) for part-time and 0.86 (95%CI: 0.82–0.91) for full-time]. Similarly, ACD completion rates were 16.7% for those not working, 9.6% for part-time, and 7.1% for full-time workers [adj. PR: 0.82 (95%CI: 0.73–0.91) for part-time and 0.72 (95%CI: 0.63–0.82) for full-time workers].
| Factors | n | Did not have ACP, n (%) | Had ACP, n (%) | Crude PR (95%CI) | Adj. PR (95%CI) | |
|---|---|---|---|---|---|---|
| Age at follow-up (years) | ||||||
| 49–59 | 7965 | 4126 (51.8) | 3839 (48.2) | 1 | 1 | |
| 60–74 | 14,935 | 5529 (37.0) | 9406 (63.0) | 1.31 (1.26–1.36) | 1.16 (1.10–1.21) | |
| 75–84 | 4616 | 1269 (27.5) | 3347 (72.5) | 1.50 (1.44–1.58) | 1.29 (1.21–1.38) | |
| 85+ | 1110 | 244 (22.0) | 866 (78.0) | 1.62 (1.50–1.74) | 1.44 (1.30–1.60) | |
| Gender | ||||||
| Male | 13,974 | 5758 (41.2) | 8216 (58.8) | 1 | 1 | |
| Female | 14,652 | 5410 (36.9) | 9242 (63.1) | 1.07 (1.04–1.11) | 1.07 (1.03–1.11) | |
| Household income (A$) | ||||||
| <$20,000 | 3151 | 1206 (38.3) | 1945 (61.7) | 1 | 1 | |
| $20,000–39,999 | 5643 | 2033 (36.0) | 3610 (64.0) | 1.04 (0.98–1.10) | 1.01 (0.95–1.08) | |
| $40,000–69,999 | 5862 | 2260 (38.6) | 3602 (61.4) | 1.00 (0.94–1.05) | 1.02 (0.95–1.09) | |
| $70,000 or more | 9063 | 3854 (42.5) | 5209 (57.5) | 0.93 (0.88–0.98) | 1.05 (0.98–1.12) | |
| Won’t disclose | 4868 | 1801 (37.0) | 3067 (63.0) | 1.02 (0.96–1.08) | 1.01 (0.94–1.08) | |
| Highest education | ||||||
| No school certificate or other qualification | 2046 | 851 (41.6) | 1195 (58.4) | 1 | 1 | |
| School or intermediate certificate | 5417 | 2088 (38.5) | 3329 (61.5) | 1.05 (0.99–1.12) | 1.02 (0.94–1.11) | |
| Higher school or leaving certificate | 2761 | 1126 (40.8) | 1635 (59.2) | 1.01 (0.94–1.09) | 1.04 (0.95–1.14) | |
| Trade or apprenticeship | 2926 | 1248 (42.7) | 1678 (57.3) | 0.98 (0.91–1.06) | 1.02 (0.93–1.11) | |
| Certificate or diploma | 6599 | 2513 (38.1) | 4086 (61.9) | 1.06 (0.99–1.13) | 1.07 (0.99–1.16) | |
| University degree or higher | 8674 | 3259 (37.6) | 5415 (62.4) | 1.07 (1.00–1.14) | 1.11 (1.02–1.20) | |
| Private health insurance | ||||||
| No | 8689 | 3874 (44.6) | 4815 (55.4) | 1 | 1 | |
| Yes | 19,937 | 7294 (36.6) | 12,643 (63.4) | 1.14 (1.11–1.18) | 1.13 (1.08–1.17) | |
| Health care concession card (Health Care Card) | ||||||
| No | 20,860 | 8520 (40.8) | 12,340 (59.2) | 1 | 1 | |
| Yes | 7766 | 2648 (34.1) | 5118 (65.9) | 1.11 (1.08–1.15) | 1.04 (1.00–1.09) | |
| Born in Australia | ||||||
| No | 6115 | 2574 (42.1) | 3541 (57.9) | 1 | 1 | |
| Yes | 22,367 | 8532 (38.1) | 13,835 (61.9) | 1.07 (1.03–1.11) | 1.04 (0.99–1.08) | |
| Speaks language other than English at home | ||||||
| No | 26,882 | 10,353 (38.5) | 16,529 (61.5) | 1 | 1 | |
| Yes | 1744 | 815 (46.7) | 929 (53.3) | 0.87 (0.81–0.92) | 1.01 (0.94–1.09) | |
| Employment status | ||||||
| Not working | 16,199 | 5197 (32.1) | 11,002 (67.9) | 1 | 1 | |
| Part-time | 5364 | 2382 (44.4) | 2982 (55.6) | 0.82 (0.79–0.85) | 0.90 (0.86–0.95) | |
| Full-time | 6753 | 3449 (51.1) | 3304 (48.9) | 0.72 (0.69–0.75) | 0.86 (0.82–0.91) | |
| Current partner (married/de facto) | ||||||
| No | 7391 | 2991 (40.5) | 4400 (59.5) | 1 | 1 | |
| Yes | 21,235 | 8177 (38.5) | 13,058 (61.5) | 1.03 (1.00–1.07) | 1.07 (1.03–1.12) | |
| Number of self-reported conditions | ||||||
| None | 6255 | 2922 (46.7) | 3333 (53.3) | 1 | 1 | |
| One | 9205 | 3730 (40.5) | 5475 (59.5) | 1.12 (1.07–1.17) | 1.06 (1.01–1.11) | |
| Two | 7073 | 2584 (36.5) | 4489 (63.5) | 1.19 (1.14–1.25) | 1.10 (1.04–1.16) | |
| Three | 3809 | 1261 (33.1) | 2548 (66.9) | 1.26 (1.19–1.32) | 1.08 (1.01–1.19) | |
| Four or more | 2284 | 671 (29.4) | 1613 (70.6) | 1.33 (1.25–1.41) | 1.13 (1.05–1.21) | |
| Self-reported good quality of life | ||||||
| No | 25,782 | 10,084 (39.1) | 15,698 (60.9) | 1 | 1 | |
| Yes | 2112 | 788 (37.3) | 1324 (62.7) | 1.03 (0.97–1.09) | 0.94 (0.87–1.02) | |
| Needing help with daily activity | ||||||
| No | 26,432 | 10,577 (40.0) | 15,855 (60.0) | 1 | 1 | |
| Yes | 1861 | 460 (24.7) | 1401 (75.3) | 1.26 (1.19–1.32) | 1.14 (1.06–1.23) | |
| Psychological distress | ||||||
| Low | 25,271 | 9664 (38.2) | 15,607 (61.8) | 1 | 1 | |
| High | 1545 | 638 (41.3) | 907 (58.7) | 0.95 (0.89–1.02) | 0.99 (0.91, 1.07) | |
| Physical functioning | ||||||
| No limitations | 6507 | 2967 (45.6) | 3540 (54.4) | 1 | 1 | |
| Minor limitations | 9073 | 3652 (40.3) | 5421 (59.7) | 1.10 (1.05–1.15) | 1.00 (0.96–1.05) | |
| Moderate limitations | 7547 | 2714 (36.0) | 4833 (64.0) | 1.18 (1.13–1.23) | 1.03 (0.98–1.09) | |
| Severe limitations | 3200 | 954 (29.8) | 2246 (70.2) | 1.29 (1.22–1.36) | 1.07 (0.99–1.14) | |
| Carer status | ||||||
| No | 24,813 | 9779 (39.4) | 15,034 (60.6) | 1 | 1 | |
| Yes | 3362 | 1221 (36.3) | 2141 (63.7) | 1.05 (1.00–1.10) | 1.02 (0.97–1.08) | |
| Fall in the past 12 months | ||||||
| No | 22,120 | 8856 (40.0) | 13,264 (60.0) | 1 | 1 | |
| Yes | 5490 | 1879 (34.2) | 3611 (65.8) | 1.10 (1.06–1.14) | 1.03 (0.99–1.07) | |
| Made a will | ||||||
| No | 3923 | 2717 (69.3) | 1206 (30.7) | 1 | 1 | |
| Yes | 24,703 | 8451 (34.2) | 16,252 (65.8) | 2.14 (2.02–2.27) | 1.76 (1.65–1.89) | |
Note: all factors were included in the multiple Poisson regression model to estimate the adjusted PRs. Bold indicated significantly higher than the reference group; bold and italics indicates significantly lower than the reference group.
| Factors | n | Non-ACD group, n (%) | ACD group, n (%) | Crude PR (95%CI) | Adj. PR (95%CI) | |
|---|---|---|---|---|---|---|
| Age at follow-up (years) | ||||||
| 49–59 | 7965 | 7456 (93.6) | 509 (6.4) | 1 | 1 | |
| 60–74 | 14,935 | 13,053 (87.4) | 1882 (12.6) | 1.97 (1.79–2.18) | 1.52 (1.35–1.71) | |
| 75–84 | 4616 | 3595 (77.9) | 1021 (22.1) | 3.46 (3.11–3.85) | 2.44 (2.11–2.82) | |
| 85+ | 1110 | 778 (70.1) | 332 (29.9) | 4.68 (4.07–5.37) | 3.12 (2.56–3.79) | |
| Sex | ||||||
| Male | 13,974 | 12,366 (88.5) | 1608 (11.5) | 1 | 1 | |
| Female | 14,652 | 12,516 (85.4) | 2136 (14.6) | 1.27 (1.19–1.35) | 1.19 (1.09–1.28) | |
| Household income (AS) | ||||||
| <$20,000 | 3151 | 2658 (84.4) | 493 (15.6) | 1 | 1 | |
| $20,000–39,999 | 5643 | 4810 (85.2) | 833 (14.8) | 0.94 (0.84–1.06) | 0.96 (0.84–1.10) | |
| $40,000–69,999 | 5862 | 5154 (87.9) | 708 (12.1) | 0.77 (0.69–0.87) | 0.98 (0.85–1.13) | |
| $70,000 or more | 9063 | 8130 (89.7) | 933 (10.3) | 0.66 (0.59–0.73) | 1.10 (0.94–1.28) | |
| Won’t disclose | 4868 | 4100 (84.2) | 768 (15.8) | 1.01 (0.90–1.13) | 1.12 (0.97–1.29) | |
| Highest education | ||||||
| No school certificate or other qualification | 2046 | 1738 (84.9) | 308 (15.1) | 1 | 1 | |
| School or intermediate certificate | 5417 | 4714 (87.0) | 703 (13.0) | 0.86 (0.75–0.99) | 0.90 (0.76–1.08) | |
| Higher school or leaving certificate | 2761 | 2428 (87.9) | 333 (12.1) | 0.80 (0.69–.94) | 0.96 (0.79–1.17) | |
| Trade or apprenticeship | 2926 | 2608 (89.1) | 318 (10.9) | 0.72 (0.62–0.84) | 0.89 (0.73–1.09) | |
| Certificate or diploma | 6599 | 5760 (87.3) | 839 (12.7) | 0.84 (0.74–0.96) | 1.03 (0.87–1.22) | |
| University degree or higher | 8674 | 7462 (86.0) | 1212 (14.0) | 0.93 (0.82–1.05) | 1.27 (1.07–1.51) | |
| Private health insurance | ||||||
| No | 8689 | 7618 (87.7) | 1071 (12.3) | 1 | 1 | |
| Yes | 19,937 | 17,264 (86.6) | 2673 (13.4) | 1.09 (1.01–1.17) | 1.19 (1.09–1.31) | |
| Health care concession card (Health Care Card) | ||||||
| No | 20,860 | 18,363 (88.0) | 2497 (12.0) | 1 | 1 | |
| Yes | 7766 | 6519 (83.9) | 1247 (16.1) | 1.34 (1.25–1.44) | 1.08 (0.98–1.18) | |
| Born in Australia | ||||||
| No | 6115 | 5328 (87.1) | 787 (12.9) | 1 | 1 | |
| Yes | 22,367 | 19,436 (86.9) | 2931 (13.1) | 1.02 (0.94–1.10) | 1.03 (0.93–1.13) | |
| Speaks language other than English at home | ||||||
| No | 26,882 | 23,347 (86.8) | 3535 (13.2) | 1 | 1 | |
| Yes | 1744 | 1535 (88.0) | 209 (12.0) | 0.91 (0.79–1.04) | 1.18 (0.99–1.39) | |
| Employment status | ||||||
| Not working | 16,199 | 13,493 (83.3) | 2706 (16.7) | 1 | 1 | |
| Part-time | 5364 | 4848 (90.4) | 516 (9.6) | 0.58 (0.52–0.63) | 0.82 (0.73–0.91) | |
| Full-time | 6753 | 6274 (92.9) | 479 (7.1) | 0.42 (0.38–0.47) | 0.72 (0.63–0.82) | |
| Current partner (married/de facto) | ||||||
| No | 7391 | 6081 (82.3) | 1310 (17.7) | 1 | 1 | |
| Yes | 21,235 | 18,801 (88.5) | 2434 (11.5) | 0.65 (0.60–0.69) | 0.76 (0.69–0.82) | |
| Number of self-reported conditions | ||||||
| None | 6255 | 5615 (89.8) | 640 (10.2) | 1 | 1 | |
| One | 9205 | 8113 (88.1) | 1092 (11.9) | 1.16 (1.05–1.28) | 1.06 (0.95–1.19) | |
| Two | 7073 | 6084 (86.0) | 989 (14.0) | 1.37 (1.24–1.51) | 1.12 (0.99–1.26) | |
| Three | 3809 | 3241 (85.1) | 568 (14.9) | 1.46 (1.30–1.63) | 1.09 (0.95–1.24) | |
| Four or more | 2284 | 1829 (80.1) | 455 (19.9) | 1.95 (1.73–2.19) | 1.25 (1.07–1.46) | |
| Self-reported good quality of life | ||||||
| No | 25,782 | 22,506 (87.3) | 3276 (12.7) | 1 | 1 | |
| Yes | 2112 | 1751 (82.9) | 361 (17.1) | 1.35 (1.20–1.50) | 0.97 (0.83–1.13) | |
| Needing help with daily activity | ||||||
| No | 26,432 | 23,153 (87.6) | 3279 (12.4) | 1 | 1 | |
| Yes | 1861 | 1450 (77.9) | 411 (22.1) | 1.78 (1.60–1.97) | 1.23 (1.06–1.43) | |
| Psychological distress | ||||||
| Low | 25,271 | 22,016 (87.1) | 3255 (12.9) | 1 | 1 | |
| High | 1545 | 1346 (87.1) | 199 (12.9) | 1.00 (0.86–1.15) | 1.12 (0.94–1.32) | |
| Physical functioning | ||||||
| No limitations | 6507 | 5862 (90.1) | 645 (9.9) | 1 | 1 | |
| Minor limitations | 9073 | 8034 (88.5) | 1039 (11.5) | 1.16 (1.05–1.28) | 0.93 (0.84–1.03) | |
| Moderate limitations | 7547 | 6473 (85.8) | 1074 (14.2) | 1.44 (1.30–1.58) | 0.96 (0.86–1.08) | |
| Severe limitations | 3200 | 2609 (81.5) | 591 (18.5) | 1.86 (1.67–2.08) | 0.98 (0.84–1.14) | |
| Carer status | ||||||
| No | 24,813 | 21,672 (87.3) | 3141 (12.7) | 1 | 1 | |
| Yes | 3362 | 2846 (84.7) | 516 (15.3) | 1.21 (1.10–1.33) | 1.17 (1.05–1.30) | |
| Fall in the past 12 months | ||||||
| No | 22,120 | 19,438 (87.9) | 2682 (12.1) | 1 | 1 | |
| Yes | 5490 | 4616 (84.1) | 874 (15.9) | 1.31 (1.22–1.42) | 1.11 (1.01–1.21) | |
| Made a will | ||||||
| No | 3923 | 3863 (98.5) | 60 (1.5) | 1 | 1 | |
| Yes | 24,703 | 21,019 (85.1) | 3684 (14.9) | 9.75 (7.63–12.72) | 8.74 (6.39–12.40) | |
Note: all factors were included in the multiple Poisson regression model to estimate the adjusted PRs. Bold indicated significantly higher than the reference group; bold and italics indicates significantly lower than the reference group
Factors such as having a partner, being a carer, and having a will were also assessed. Individuals with a current partner (married or de facto) had higher ACP completion rates (61.5% vs 59.5%; adj. PR: 1.07; 95%CI: 1.03–1.12). For ACD, those with partners had lower completion rates (11.5% vs 17.7%; adj. PR: 0.76; 95%CI: 0.69–0.82). Those with carer status had higher ACD completion rates (15.3% vs 12.7%; Adj. PR: 1.17; 95%CI: 1.05–1.30). Those with wills had higher ACP completion rates (65.8% vs 30.7%; adj. PR: 1.76; 95%CI: 1.65–1.89) and ACD completion rates (14.9% vs 1.5%; adj. PR: 8.74; 95%CI: 6.39–12.40) than those without wills.
Individuals with private health insurance had higher ACP (63.4% vs 55.4%; adj. PR: 1.13; 95%CI: 1.08–1.17) and ACD completion rates (13.4% vs 12.3%; adj. PR: 1.19; 95%CI: 1.09–1.31) than those without private health insurance. Similarly, individuals with a health care concession card (Health Care Card) had higher ACP completion rates than those without (65.9% vs 59.2%; adj. PR: 1.04; 95%CI: 1.00–1.09).
Rates of ACP completion were affected by the number of medications, health conditions, and experience of falls. Those with one or more self-reported medical conditions had higher rates of ACP and ACD completion than those with no self-reported conditions. For ACP there was a linear association, with completion rates of 53.3% for those with no self-reported conditions and 70.6% for those with four or more self-reported conditions (adj. PR: 1.13; 95%CI: 1.05–1.21). Rates of ACD completion also increased for those with three or more self-reported conditions. Individuals who had a fall in the past 12 months had a 15.9% ACD completion rate compared with 12.1% for those without falls (adj. PR: 1.11; 95%CI: 1.01–1.21).
In a sensitivity analysis removing the question, ‘Have you legally nominated a person to make health decisions if you lose capacity’, from the ACP definition, a difference was found in the rates of ACP and the factors associated with these rates. The overall rate for ACP changed from 61% to 49% (Supplementary Table S2). In relation to the factors associated with ACP, the only difference found was that severe limitations in physical function (adj. PR: 1.09; 95%CI: 1.01–1.18) and carer status (adj. PR: 1.06; 95%CI: 1.00–1.12) were now significantly related to ACP rate (Supplementary Table S3).
Discussion
Our study assessed the self-reported sociodemographic and clinical factors influencing ACP and ACD completion rates in Australian individuals over 45 years of age who participated in the 45 and Up Study. Participants in this study who were asked the questions in 2014/2015 had discussed their wishes with someone else (47.9%), nominated a person to make health decisions for them if they lost capacity (39.9%), completed a formal ACD (13.1%), or completed any of these actions (61%). The rate of those who had only discussed their wishes with someone else and completed a formal ACD was 49%.
The strongest predictors of ACP and ACD completion were having a will and increasing age. The strongest predictor of ACP and especially ACD completion was having a will, reflecting previous studies.21 The completion of wills in Australia is much higher than for ACDs, with 60% of adults and 93% of those above 70 years of age having a will.31 This disparity could reflect that wills are financially focused, while ACP and ACD involve addressing themes of incapacity, serious illness and death.2
Increasing age was marginally associated with a higher likelihood of engaging in both ACP and formal ACD planning, consistent with the existing literature.2,6,22,24,32–35 This is hypothesised to result from individuals feeling that they are approaching the end of life, reflecting a sense of urgency in prioritising autonomy of care.36 This perception is associated with increased levels of formal engagement in ACP for those with a subjective rating of remaining life expectancy of less than 25 years.36
Being female was slightly associated with higher rates of ACP and ACD, consistent with other Australian studies.22,24 Being in a married or de facto relationship increased ACP completion rates but decreased formal ACD documentation, contrasting with studies that found higher ACD completion rates among single, divorced or separated participants.21,22 Sellars et al. also found a negative association between ACD and having discussed wishes with a partner.24 Our results may reflect a preference for verbal communication as opposed to documentation, compounded by barriers to ACD completion such as differing legislation, terminologies and local policies in Australia.37 Participants may find comfort in verbal communication, viewing it as sufficient for end-of-life care, especially when communicated to partners who may also be the person nominated to make healthcare decisions on behalf of the patient if they lose the capacity.21,24
There was no clear association between household income and ACP or ACD completion in our study. In contrast, other studies found that older adults with low income were approximately 33% less likely to participate in ACP than those with higher income.38,39
This may be due to patients from lower socioeconomic status having more difficulties understanding and participating in ACP, related to lower levels of health literacy and general education, which are known barriers.40 The contrast between our study and others may be due to the average age of participants being probably beyond retirement age. Thus, household wealth may be a more accurate representation of socioeconomic status compared with income, which was not captured in our survey.
We observed a positive relationship between ACP completion rates and people with a university degree or higher. Nouri et al. found that low health literacy was independently associated with poor ACP knowledge, hindering completion.41 Other studies also reported that individuals with higher levels of education were more likely to engage in ACP.42,43 Future policy and procedure goals could be targeted to improve education and increase access to patient resources (such as easy to comprehend and access resources), given that this has been shown to improve uptake of ACP in Australia in a variety of inpatient, outpatient and aged care settings.44
Individuals with poorer health were more likely to have an ACP or ACD in place. In our study, this included patients with more self-reported health conditions and those who had fallen in the past 12 months, possibly reflecting increased contact with healthcare professionals, or their health condition prompting ACP discussions.24 Similarly, Australian studies found higher ACP rates in residential aged care settings, where individuals were older and more unwell.6,22,45 ACDs were also more common in those with functional impairment, cognitive decline, cancer, or in palliative care.6,22,46 However, other studies found that ACP conversations rarely occur in palliative care, compounded by healthcare professionals avoiding discussions and lacking robust documentation procedures.2,47 Clinician education is important to improve confidence, and successful interventions can include educational workshops with both theoretical and practical components.44 A systematic review identified that ACP training programs for healthcare professionals improved knowledge, attitude and skills, which included decision aids and instructional sessions with role play.48 Other interventions can involve prompts such as on electronic medical records to remind clinicians to have ACP discussions, which may be particularly helpful in an outpatient setting.49
Our study demonstrates that individuals born in Australia did not have higher ACP completion rates than those born overseas. This finding was inconsistent with previous studies which reflect ethnic, cultural, and language barriers in ACP completion, considering the diversity of Australia’s population.23,50,51 A requirement for inclusion in the study was the ability to speak and read English. Interest and uptake in ACP have been increasing internationally including in non-western countries, and a recent study found that low uptake among Chinese Australians was more related to language barriers and lack of knowledge than to cultural barriers.52 Despite culturally and linguistically diverse patients feeling comfortable discussing ACP with their primary care physician, many did not feel equipped with information for end-of-life discussions, which further emphasises the need for effective education. This can include appropriate language resources, trained interpreters, and education to promote ACP for different language groups.53
Strengths and limitations
A major strength of our study was the utilisation of a large cohort of older individuals from a community setting with no restrictions on formal engagement in healthcare services or settings, providing a broad representation of the Australian public. Limitations include reliance on accurate self-reporting by participants; non-disclosure of household income by a substantial subset of individuals; and the possible reporting of a slightly higher socioeconomic status than the general population by those who did disclose their income. Further studies could assess the impact of the socioeconomic gradient of participants by assessing their residence. Additionally, we cannot accurately ascertain whether higher rates of ACP and ACD completion are due to current clinical or demographic factors, of whether they are influenced by witnessing adverse health outcomes, such as in family members.21 Notably, we found that carers had a higher level of ACD completion than non-carers.
Conclusion
The results of this study highlight sociodemographic and clinical predictors of ACP and ACD completion in Australia. The strongest predictor for completion of both was having a will, followed by advancing age. Other positive predictors included being female, poorer general health, higher levels of education, having private health insurance and not being employed. Those who were married or in a de facto relationship had higher ACP but lower ACD completion rates, suggesting a preference for informal discussions. Our results suggest the need to promote formal ACD completion, such as through government senior newsletters.54 Our results also suggest that tailored ACP discussions for those with lower levels of education may be beneficial.
Data availability
The data that support this study will be shared upon reasonable request to the corresponding author.
Declaration of funding
The data resource that allowed access to the 45 and Up Study was jointly funded by the Sydney Local Health District, the South Eastern Sydney Local Health District and the Central and Eastern Sydney Primary Health Network, which MB is responsible for.
Acknowledgements
This research was completed using data collected through the 45 and Up Study (www.saxinstitute.org.au). The 45 and Up Study is managed by the Sax Institute in collaboration with major partner Cancer Council NSW, and partners the Heart Foundation and the NSW Ministry of Health. We thank the many thousands of people participating in the 45 and Up Study. The authors also thank Katherine Meikle, Research Assistant at ICFHS, UNSW, for editorial assistance and Li Yang, Statistician at ICFHS, UNSW, for conducting the sensitivity analysis on the data.
References
1 Johnson CE, McVey P, Rhee JJ, et al. General practice palliative care: patient and carer expectations, advance care plans and place of death – a systematic review. BMJ Support Palliat Care 14: 2018; 001549.
| Crossref | Google Scholar | PubMed |
2 Scott IA, Mitchell GK, Reymond EJ, et al. Difficult but necessary conversations: the case for advance care planning. Med J Aust 2013; 199: 662-666.
| Crossref | Google Scholar | PubMed |
3 Sudore RL, Lum HD, You JJ, et al. Defining Advance Care Planning for Adults: A Consensus Definition From a Multidisciplinary Delphi Panel. J Pain Symptom Manage 2017; 53: 821-832.
| Crossref | Google Scholar | PubMed |
4 McLennan VEJ, Boddy JHM, Daly MG, et al. Relinquishing or taking control? Community perspectives on barriers and opportunities in advance care planning. Aust Health Rev 2015; 39: 528-532.
| Crossref | Google Scholar | PubMed |
5 Advance Care Planning Australia. Advance Care Planning Explained. 2021. Available at https://www.advancecareplanning.org.au/understand-advance-care-planning/advance-care-planning-explained
6 Detering KM, Buck K, Ruseckaite R, et al. Prevalence and correlates of advance care directives among older Australians accessing health and residential aged care services: multicentre audit study. BMJ Open 2019; 9: e025255.
| Crossref | Google Scholar | PubMed |
8 Detering KM, Hancock AD, Reade MC, et al. The impact of advance care planning on end of life care in elderly patients: randomised controlled trial. BMJ 2010; 340: c1345.
| Crossref | Google Scholar | PubMed |
9 Caplan GA, Meller A, Squires B, et al. Advance care planning and hospital in the nursing home. Age Ageing 2006; 35: 581-585.
| Crossref | Google Scholar | PubMed |
10 Molloy DW, Guyatt GH, Russo R, et al. Systematic implementation of an advance directive program in nursing homes: a randomized controlled trial. JAMA 2000; 283: 1437-1444.
| Crossref | Google Scholar | PubMed |
11 Brinkman-Stoppelenburg A, Rietjens JA, van der Heide A. The effects of advance care planning on end-of-life care: a systematic review. Palliat Med 2014; 28: 1000-1025.
| Crossref | Google Scholar | PubMed |
12 Houben CHM, Spruit MA, Groenen MTJ, et al. Efficacy of advance care planning: a systematic review and meta-analysis. J Am Med Dir Assoc 2014; 15: 477-489.
| Crossref | Google Scholar | PubMed |
13 Levy C, Morris M, Kramer A. Improving end-of-life outcomes in nursing homes by targeting residents at high-risk of mortality for palliative care: program description and evaluation. J Palliat Med 2008; 11: 217-225.
| Crossref | Google Scholar | PubMed |
14 Johnson SB, Butow PN, Bell ML, et al. A randomised controlled trial of an advance care planning intervention for patients with incurable cancer. Br J Cancer 2018; 119: 1182-1190.
| Crossref | Google Scholar | PubMed |
15 Batchelor F, Hwang K, Haralambous B, et al. Facilitators and barriers to advance care planning implementation in Australian aged care settings: A systematic review and thematic analysis. Australas J Ageing 2019; 38: 173-181.
| Crossref | Google Scholar | PubMed |
16 Yadav KN, Gabler NB, Cooney E, et al. Approximately One In Three US Adults Completes Any Type Of Advance Directive For End-Of-Life Care. Health Aff 2017; 36: 1244-1251.
| Crossref | Google Scholar | PubMed |
17 Glaudemans JJ, Moll van Charante E, Wind J, et al. Experiences with approaches to advance care planning with older people: a qualitative study among Dutch general practitioners. BMJ Open 2018; 8: e024762.
| Crossref | Google Scholar | PubMed |
18 Cartwright C, Montgomery J, Rhee J, et al. Medical practitioners’ knowledge and self-reported practices of substitute decision making and implementation of advance care plans. Intern Med J 2014; 44: 234-239.
| Crossref | Google Scholar | PubMed |
19 Boddy J, Chenoweth L, McLennan V, et al. It’s just too hard! Australian health care practitioner perspectives on barriers to advance care planning. Aust J Prim Health 2013; 19: 38-45.
| Crossref | Google Scholar | PubMed |
20 Jimenez G, Tan WS, Virk AK, et al. Overview of Systematic Reviews of Advance Care Planning: Summary of Evidence and Global Lessons. J Pain Symptom Manage 2018; 56: 436-459.
| Crossref | Google Scholar |
21 White B, Tilse C, Wilson J, et al. Prevalence and predictors of advance directives in Australia. Intern Med J 2014; 44: 975-980.
| Crossref | Google Scholar | PubMed |
22 Buck K, Nolte L, Sellars M, et al. Advance care directive prevalence among older Australians and associations with person-level predictors and quality indicators. Health Expect 2021; 24: 1312-1325.
| Crossref | Google Scholar | PubMed |
23 Sinclair C, Sellars M, Buck K, et al. Association Between Region of Birth and Advance Care Planning Documentation Among Older Australian Migrant Communities: A Multicenter Audit Study. J Gerontol B Psychol Sci Soc Sci 2021; 76: 109-120.
| Crossref | Google Scholar | PubMed |
24 Sellars M, Detering KM, Sinclair C, et al. Personal and Interpersonal Factors and Their Associations With Advance Care Planning Documentation: A Cross-sectional Survey of Older Adults in Australia. J Pain Symptom Manage 2020; 59: 1212-1222.
| Crossref | Google Scholar |
25 Risk J, Mohammadi L, Rhee J, et al. Barriers, enablers and initiatives for uptake of advance care planning in general practice: a systematic review and critical interpretive synthesis. BMJ Open 2019; 9: e030275.
| Crossref | Google Scholar | PubMed |
26 Comino E, Harris E, Page J, et al. The 45 and Up Study: a tool for local population health and health service planning to improve integration of healthcare. Public Health Res Pract 2016; 26: 2631629.
| Crossref | Google Scholar | PubMed |
27 Bleicher K, Summerhayes R, Baynes S, et al. Cohort Profile Update: The 45 and Up Study. Int J Epidemiol 2023; 52: e92-e101.
| Crossref | Google Scholar | PubMed |
28 Kabir A, Tran A, Ansari S, et al. Impact of multimorbidity and complex multimorbidity on mortality among older Australians aged 45 years and over: A large population-based record linkage study. BMJ Open 2022; 12: e060001.
| Crossref | Google Scholar | PubMed |
29 Kessler R, Barker P, Colpe L, et al. Screening for serious mental illness in the general population. Arch Gen Psychiatry 2003; 60: 184-189.
| Crossref | Google Scholar | PubMed |
30 Ware JJ, Sherbourne C. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care 1992; 30(6): 473-483.
| Google Scholar | PubMed |
32 McCarthy EP, Pencina MJ, Kelly-Hayes M, et al. Advance care planning and health care preferences of community-dwelling elders: The Framingham Heart Study. J Gerontol A Biol Sci Med Sci 2008; 63: 951-959.
| Crossref | Google Scholar | PubMed |
33 Chu LW, Luk JK, Hui E, et al. Advance directive and end-of-life care preferences among Chinese nursing home residents in Hong Kong. J Am Med Dir Assoc 2011; 12: 143-152.
| Crossref | Google Scholar | PubMed |
34 Boerner K, Carr D, Moorman S. Family relationships and advance care planning: do supportive and critical relations encourage or hinder planning? J Gerontol B Psychol Sci Soc Sci 2013; 68: 246-256.
| Crossref | Google Scholar | PubMed |
35 Musa I, Seymour J, Narayanasamy MJ, et al. A survey of older peoples’ attitudes towards advance care planning. Age and Ageing 2015; 44: 371-376.
| Crossref | Google Scholar | PubMed |
36 Fleuren N, Depla MFIA, Pasman HRW, et al. Association Between Subjective Remaining Life Expectancy and Advance Care Planning in Older Adults: A Cross-Sectional Study. J Pain Symptom Manage 2021; 62: 757-767.
| Crossref | Google Scholar | PubMed |
37 Rhee JJ, Zwar NA, Kemp LA. Uptake and implementation of Advance Care Planning in Australia: findings of key informant interviews. Aust Health Rev 2012; 36: 98-104.
| Crossref | Google Scholar | PubMed |
38 Inoue M. The Influence of Sociodemographic and Psychosocial Factors on Advance Care Planning. J Gerontol Soc Work 2016; 59: 401-422.
| Crossref | Google Scholar | PubMed |
39 Nouri S, Lyles CR, Rubinsky AD, et al. Evaluation of Neighborhood Socioeconomic Characteristics and Advance Care Planning Among Older Adults. JAMA Netw Open 2020; 3: e2029063.
| Crossref | Google Scholar | PubMed |
40 Silies K, Huckle T, Schnakenberg R, et al. Contextual factors influencing advance care planning in home care: process evaluation of the cluster-randomised controlled trial STADPLAN. BMC Geriatr 2022; 22: 345.
| Crossref | Google Scholar | PubMed |
41 Nouri SS, Barnes DE, Volow AM, et al. Health Literacy Matters More Than Experience for Advance Care Planning Knowledge Among Older Adults. J Am Geriatr Soc 2019; 67: 2151-2156.
| Crossref | Google Scholar | PubMed |
42 Kermel-Schiffman I, Werner P. Knowledge regarding advance care planning: A systematic review. Arch Gerontol Geriatr 2017; 73: 133-142.
| Crossref | Google Scholar | PubMed |
43 De Vleminck A, Batteauw D, Demeyere T, et al. Do non-terminally ill adults want to discuss the end of life with their family physician? An explorative mixed-method study on patients’ preferences and family physicians’ views in Belgium. Fam Pract 2018; 35: 495-502.
| Crossref | Google Scholar | PubMed |
44 Liang Q, Eng D, Tate J. Improving uptake of advance care planning through education and training in a public hospital geriatric clinic. Australas J Ageing 2023; 42: 577-583.
| Crossref | Google Scholar | PubMed |
45 Detering KM, Sinclair C, Buck K, et al. Organisational and advance care planning program characteristics associated with advance care directive completion: a prospective multicentre cross-sectional audit among health and residential aged care services caring for older Australians. BMC Health Serv Res 2021; 21: 700.
| Crossref | Google Scholar | PubMed |
46 Brown M, Jarrad S. Putting “the powers” in place: barriers for people with memory loss in planning for the future. J Law Med 2008; 15: 530-537.
| Google Scholar | PubMed |
47 Voss H, Vogel A, Wagemans AMA, et al. Advance Care Planning in Palliative Care for People With Intellectual Disabilities: A Systematic Review. J Pain Symptom Manage 2017; 54: 938-960.
| Crossref | Google Scholar | PubMed |
48 Chan CWH, Ng NHY, Chan HYL, et al. A systematic review of the effects of advance care planning facilitators training programs. BMC Health Serv Res 2019; 19: 362.
| Crossref | Google Scholar | PubMed |
49 Patel N, De Lima B, Neukam S, et al. Prioritizing advance care planning in primary care clinics. J Am Geriatr Soc 2024; 72: 613-615.
| Crossref | Google Scholar | PubMed |
50 Ohr S, Jeong S, Saul P. Cultural and religious beliefs and values, and their impact on preferences for end-of-life care among four ethnic groups of community-dwelling older persons. J Clin Nurs 2017; 26: 1681-1689.
| Crossref | Google Scholar | PubMed |
51 Sinclair C, Smith J, Toussaint Y, et al. Discussing dying in the diaspora: attitudes towards advance care planning among first generation Dutch and Italian migrants in rural Australia. Soc Sci Med 2014; 101: 86-93.
| Crossref | Google Scholar | PubMed |
52 Yap SS, Chen K, Detering KM, et al. Exploring the knowledge, attitudes and needs of advance care planning in older Chinese Australians. J Clin Nurs 2018; 27: 3298-3306.
| Crossref | Google Scholar | PubMed |
53 Detering K, Sutton E, Fraser S, et al. Feasibility and acceptability of advance care planning in elderly Italian and Greek speaking patients as compared to English-speaking patients: an Australian cross-sectional study. BMJ Open 2015; 5: e008800.
| Crossref | Google Scholar | PubMed |
54 NSW Government. Seniors Card EXTRA e-newsletter. 2024. Available at https://www.nsw.gov.au/community-services/seniors-card/extra-newsletter [accessed September 2024].
