[Skip to Navigation]
Sign In
Figure 1.  Association of Socioeconomic Indicators With Having at Least 1 Elective or Nonelective Hospitalization Among 1566 Participants
Association of Socioeconomic Indicators With Having at Least 1 Elective or Nonelective Hospitalization Among 1566 Participants

Calendar year was used as the time scale, with survivors having a censoring date of December 31, 2017, for elective hospitalization (8189 person-years of follow-up) and nonelective hospitalization (8282 person-years of follow-up). All estimates were adjusted for age, age squared, country of birth, and marital status. SEP indicates socioeconomic position; SHR, subhazard ratio.

Figure 2.  Association of Socioeconomic Indicators With Number of Elective and Nonelective Hospitalizations
Association of Socioeconomic Indicators With Number of Elective and Nonelective Hospitalizations

Overall, the total number of elective hospitalizations was 2871 and nonelective hospitalization was 5102. All estimates were adjusted for age, age squared, country of birth, and marital status. RR indicates rate ratio; SEP, socioeconomic position.

Figure 3.  Association of Socioeconomic Indicators With Length of Stay of First Elective and Nonelective Hospitalization
Association of Socioeconomic Indicators With Length of Stay of First Elective and Nonelective Hospitalization

All estimates were adjusted for age, age squared, country of birth, and marital status. LOS indicates length of stay; RR, rate ratio; and SEP, socioeconomic position.

Figure 4.  Association of Socioeconomic Indicators With Cumulative Length of Stay of Elective and Nonelective Hospitalizations
Association of Socioeconomic Indicators With Cumulative Length of Stay of Elective and Nonelective Hospitalizations

All estimates were adjusted for age, age squared, country of birth, and marital status. LOS indicates length of stay; RR, rate ratio; and SEP, socioeconomic position.

Table.  Characteristics of Participants by Indicators of Socioeconomic Position at Baseline
Characteristics of Participants by Indicators of Socioeconomic Position at Baseline
1.
Beard  JR, Officer  A, de Carvalho  IA,  et al.  The world report on ageing and health: a policy framework for healthy ageing.   Lancet. 2016;387(10033):2145-2154. doi:10.1016/S0140-6736(15)00516-4PubMedGoogle ScholarCrossref
2.
He  W, Goodkind  D, Kowal  P. An Aging World. US Census Bureau; 2016.
3.
Braveman  P.  Health disparities and health equity: concepts and measurement.   Annu Rev Public Health. 2006;27:167-194. doi:10.1146/annurev.publhealth.27.021405.102103PubMedGoogle ScholarCrossref
4.
Australian Institute of Health and Welfare. Health System Overview. July 23, 2020. Accessed December 10, 2021. https://www.aihw.gov.au/reports/australias-health/health-system-overview
5.
Australian Institute of Health and Welfare.  Admitted patient care 2017–18: Australian hospital statistics. AIHW; 2019.
6.
Furler  JS, Harris  E, Chondros  P, Powell Davies  PG, Harris  MF, Young  DYL.  The inverse care law revisited: impact of disadvantaged location on accessing longer GP consultation times.   Med J Aust. 2002;177(2):80-83. doi:10.5694/j.1326-5377.2002.tb04673.xPubMedGoogle ScholarCrossref
7.
van Doorslaer  E, Masseria  C, Koolman  X; OECD Health Equity Research Group.  Inequalities in access to medical care by income in developed countries.   CMAJ. 2006;174(2):177-183. doi:10.1503/cmaj.050584PubMedGoogle ScholarCrossref
8.
Andersen  R, Newman  JF.  Societal and individual determinants of medical care utilization in the United States.   Milbank Q. 2005;83(4):10.1111/j.1468-0009.2005.00428.x. doi:10.1111/j.1468-0009.2005.00428.xGoogle Scholar
9.
Amegbor  PM, Plumb  KB, Rosenberg  MW.  Determinants of overnight stay in health centres and length of admission: a study of Canadian seniors.   Can J Aging. 2020;39(4):533-544. doi:10.1017/S0714980819000771PubMedGoogle ScholarCrossref
10.
Antonelli-Incalzi  R, Ancona  C, Forastiere  F, Belleudi  V, Corsonello  A, Perucci  CA.  Socioeconomic status and hospitalization in the very old: a retrospective study.   BMC Public Health. 2007;7(1):227. doi:10.1186/1471-2458-7-227PubMedGoogle ScholarCrossref
11.
Wachelder  JJH, van Drunen  I, Stassen  PM,  et al.  Association of socioeconomic status with outcomes in older adult community-dwelling patients after visiting the emergency department: a retrospective cohort study.   BMJ Open. 2017;7(12):e019318-e019318. doi:10.1136/bmjopen-2017-019318PubMedGoogle ScholarCrossref
12.
Hart  CL, McCartney  G, Watt  GCM.  Occupational class differences in later life hospital use by women who survived to age 80: the Renfrew and Paisley prospective cohort study.   Age Ageing. 2015;44(3):515-519. doi:10.1093/ageing/afu184PubMedGoogle ScholarCrossref
13.
Payne  RA, Abel  GA, Guthrie  B, Mercer  SW.  The effect of physical multimorbidity, mental health conditions and socioeconomic deprivation on unplanned admissions to hospital: a retrospective cohort study.   CMAJ. 2013;185(5):E221-E228. doi:10.1503/cmaj.121349PubMedGoogle ScholarCrossref
14.
Cournane  S, Byrne  D, Conway  R, O’Riordan  D, Coveney  S, Silke  B.  Social deprivation and hospital admission rates, length of stay and readmissions in emergency medical admissions.   Eur J Intern Med. 2015;26(10):766-771. doi:10.1016/j.ejim.2015.09.019PubMedGoogle ScholarCrossref
15.
McCartney  G, Hart  C, Watt  G.  How can socioeconomic inequalities in hospital admissions be explained? a cohort study.   BMJ Open. 2013;3(8):e002433. doi:10.1136/bmjopen-2012-002433PubMedGoogle Scholar
16.
Australian Institute of Health and Welfare. Australia’s health 2020: in brief. July 23, 2020. Accessed December 10, 2021. https://www.aihw.gov.au/getmedia/2aa9f51b-dbd6-4d56-8dd4-06a10ba7cae8/aihw-aus-232.pdf.aspx?inline=true
17.
Cumming  RG, Handelsman  D, Seibel  MJ,  et al.  Cohort profile: the Concord Health and Ageing in Men Project (CHAMP).   Int J Epidemiol. 2009;38(2):374-378. doi:10.1093/ije/dyn071PubMedGoogle ScholarCrossref
18.
World Medical Association.  World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects.   JAMA. 2013;310(20):2191-2194. doi:10.1001/jama.2013.281053Google ScholarCrossref
19.
Trewin D, Pink B. ANZSCO: Australian and New Zealand standard classification of occupations. Australian Bureau of Statistics, Statistics New Zealand. September 11, 2006. Accessed March 7, 2022. http://www.jobmarkets.com.au/doc/ANZSCO_2006_pdf.pdf
20.
Khalatbari-Soltani  S, Blyth  FM, Naganathan  V,  et al.  Cohort Profile update: the Concord Health and Ageing in Men Project (CHAMP).   Int J Epidemiol. 2022;51(1):31-32h. doi:10.1093/ije/dyab171PubMedGoogle ScholarCrossref
21.
Centre for Health Record Linkage. Accessed February 8, 2022. https://www.cherel.org.au/
22.
The International Statistical Classification of Diseases and Related Health Problems.  Tenth Revision, Australian Modification (ICD-10-AM). 7th ed. National Centre for Classification in Health; 2010.
23.
World Health Organization. International Statistical Classification of Diseases and Related Health Problems 10th Revision; 2019. Accessed February 8, 2022. https://icd.who.int/browse10/2019/en
24.
Hernán  MA, Hernández-Díaz  S, Werler  MM, Mitchell  AA.  Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology.   Am J Epidemiol. 2002;155(2):176-184. doi:10.1093/aje/155.2.176PubMedGoogle ScholarCrossref
25.
Gray  B. cmprsk: Subdistribution analysis of competing risks. 2020. Accessed February 8, 2022. https://cran.r-project.org/package=cmprsk
26.
Therneau  TM. A package for survival analysis in R. 2021. Accessed February 8, 2022. https://cran.r-project.org/package=survival
27.
Venables  WN, Ripley  BD, Ripley  BD.  Modern Applied Statistics with S. 4th ed. Springer; 2003.
28.
Fine  JP, Gray  RJ.  A proportional hazards model for the subdistribution of a competing risk.   J Am Stat Assoc. 1999;94(446):496-509. doi:10.1080/01621459.1999.10474144Google ScholarCrossref
29.
Schoenfeld  D.  Chi-squared goodness-of-fit tests for the proportional hazards regression model.   Biometrika. 1980;67(1):145-153. doi:10.1093/biomet/67.1.145Google ScholarCrossref
30.
Tung  H-H, Lu  T-M, Chen  L-K, Liang  S-Y, Wu  S-F, Chu  K-H.  Health literacy impact on elderly patients with heart failure in Taiwan.   J Clin Gerontol Geriatr. 2014;5(3):72-76. doi:10.1016/j.jcgg.2014.01.005Google ScholarCrossref
31.
Carlucio da Silva  FB, Filipini  R, Guarnieri  AP,  et al.  Sociodemographic profile and health condition of elderly patients attended at a community primary health center.   J Clin Gerontol Geriatr. 2016;7(3):93-98. doi:10.1016/j.jcgg.2016.03.003Google ScholarCrossref
32.
Sarwar  MR, Iftikhar  S, Sarfraz  M.  Influence of education level of older patients on polypharmacy, potentially inappropriate medications listed in Beer’s Criteria, and unplanned hospitalization: a cross-sectional study in Lahore, Pakistan.   Medicina (Kaunas). 2018;54(4):57. doi:10.3390/medicina54040057PubMedGoogle ScholarCrossref
33.
Pujol  J, Godoy  P, Soldevila  N,  et al; CIBERESP Cases and Controls in Pandemic Influenza Working Group.  Social class based on occupation is associated with hospitalization for A(H1N1)pdm09 infection. Comparison between hospitalized and ambulatory cases.   Epidemiol Infect. 2016;144(4):732-740. doi:10.1017/S0950268815001892PubMedGoogle ScholarCrossref
34.
Heliövaara  M.  Occupation and risk of herniated lumbar intervertebral disc or sciatica leading to hospitalization.   J Chronic Dis. 1987;40(3):259-264. doi:10.1016/0021-9681(87)90162-7PubMedGoogle ScholarCrossref
35.
Tertulien  T, Chen  Y, Althouse  AD, Essien  UR, Johnson  A, Magnani  JW.  Association of income and educational attainment in hospitalization events in atrial fibrillation.   Am J Prev Cardiol. 2021;7:100201. doi:10.1016/j.ajpc.2021.100201PubMedGoogle Scholar
36.
Prescott  E, Lange  P, Vestbo  J.  Socioeconomic status, lung function and admission to hospital for COPD: results from the Copenhagen City Heart Study.   Eur Respir J. 1999;13(5):1109-1114. doi:10.1034/j.1399-3003.1999.13e28.xPubMedGoogle ScholarCrossref
37.
Eagan  TML, Gulsvik  A, Eide  GE, Bakke  PS.  The effect of educational level on the incidence of asthma and respiratory symptoms.   Respir Med. 2004;98(8):730-736. doi:10.1016/j.rmed.2004.02.008PubMedGoogle ScholarCrossref
38.
Bosdriesz  JR, Willemsen  MC, Stronks  K, Kunst  AE.  Socioeconomic inequalities in smoking cessation in 11 European countries from 1987 to 2012.   J Epidemiol Community Health. 2015;69(9):886-892. doi:10.1136/jech-2014-205171PubMedGoogle ScholarCrossref
39.
Andersen  KK, Steding-Jessen  M, Dalton  SO, Olsen  TS.  Socioeconomic position and incidence of ischemic stroke in Denmark 2003-2012: a nationwide hospital-based study.   J Am Heart Assoc. 2014;3(4):e000762. doi:10.1161/JAHA.113.000762PubMedGoogle Scholar
40.
Vart  P, Coresh  J, Kwak  L, Ballew  SH, Heiss  G, Matsushita  K.  Socioeconomic status and incidence of hospitalization with lower-extremity peripheral artery disease: atherosclerosis risk in communities study.   J Am Heart Assoc. 2017;6(8):e004995. doi:10.1161/JAHA.116.004995PubMedGoogle Scholar
41.
Havranek  EP, Mujahid  MS, Barr  DA,  et al; American Heart Association Council on Quality of Care and Outcomes Research, Council on Epidemiology and Prevention, Council on Cardiovascular and Stroke Nursing, Council on Lifestyle and Cardiometabolic Health, and Stroke Council.  Social determinants of risk and outcomes for cardiovascular disease: a scientific statement from the American Heart Association.   Circulation. 2015;132(9):873-898. doi:10.1161/CIR.0000000000000228PubMedGoogle ScholarCrossref
42.
Turrell  G, Mathers  CD.  Socioeconomic status and health in Australia.   Med J Aust. 2000;172(9):434-438. doi:10.5694/j.1326-5377.2000.tb124041.xPubMedGoogle ScholarCrossref
43.
Dannefer  D.  Cumulative advantage/disadvantage and the life course: cross-fertilizing age and social science theory.   J Gerontol B Psychol Sci Soc Sci. 2003;58(6):S327-S337. doi:10.1093/geronb/58.6.S327PubMedGoogle ScholarCrossref
44.
Barnett  K, Mercer  SW, Norbury  M, Watt  G, Wyke  S, Guthrie  B.  Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study.   Lancet. 2012;380(9836):37-43. doi:10.1016/S0140-6736(12)60240-2PubMedGoogle ScholarCrossref
45.
Fors  S, Lennartsson  C, Lundberg  O.  Health inequalities among older adults in Sweden 1991-2002.   Eur J Public Health. 2008;18(2):138-143. doi:10.1093/eurpub/ckm097PubMedGoogle ScholarCrossref
46.
Gondek  D, Ploubidis  GB, Hossin  MZ, Gao  M, Bann  D, Koupil  I.  Inequality in hospitalization due to non-communicable diseases in Sweden: age-cohort analysis of the Uppsala Birth Cohort Multigenerational Study.   SSM Popul Heal. 2021;13:100741. doi:10.1016/j.ssmph.2021.100741PubMedGoogle Scholar
47.
Krause  N, Borawski-Clark  E.  Social class differences in social support among older adults.   Gerontologist. 1995;35(4):498-508. doi:10.1093/geront/35.4.498PubMedGoogle ScholarCrossref
48.
Khalatbari-Soltani  S, Stanaway  F, Cvejic  E,  et al.  Contribution of psychosocial factors to socioeconomic inequalities in mortality among older Australian men: a population-based cohort study.   Int J Equity Health. 2020;19(1):177. doi:10.1186/s12939-020-01277-2PubMedGoogle ScholarCrossref
49.
Khalatbari-Soltani  S, Blyth  FM, Naganathan  V,  et al.  Socioeconomic status, health-related behaviours, and death among older people: the Concord health and aging in men project prospective cohort study.   BMC Geriatr. 2020;20(1):261. doi:10.1186/s12877-020-01648-yPubMedGoogle ScholarCrossref
50.
Khalatbari-Soltani  S, Cumming  RG, Chomik  R,  et al.  The association between home ownership and the health of older men: Cross-sectional analysis of the Australian Concord Health and Ageing in Men Project.   Australas J Ageing. 2021;40(3)e199-e206. doi:10.1111/ajag.12896Google Scholar
51.
Brameld  KJ, Holman  CDJ.  The use of end-quintile comparisons to identify under-servicing of the poor and over-servicing of the rich: a longitudinal study describing the effect of socioeconomic status on healthcare.   BMC Health Serv Res. 2005;5:61. doi:10.1186/1472-6963-5-61PubMedGoogle ScholarCrossref
52.
Yusuf  F, Leeder  SR.  Can’t escape it: the out-of-pocket cost of health care in Australia.   Med J Aust. 2013;199(7):475-478. doi:10.5694/mja12.11638PubMedGoogle ScholarCrossref
53.
Yusuf  F, Leeder  S.  Recent estimates of the out-of-pocket expenditure on health care in Australia.   Aust Health Rev. 2020;44(3):340-346. doi:10.1071/AH18191PubMedGoogle ScholarCrossref
54.
Adler  NE, Newman  K.  Socioeconomic disparities in health: pathways and policies.   Health Aff (Millwood). 2002;21(2):60-76. doi:10.1377/hlthaff.21.2.60PubMedGoogle ScholarCrossref
55.
Kim  S, Symons  M, Popkin  BM.  Contrasting socioeconomic profiles related to healthier lifestyles in China and the United States.   Am J Epidemiol. 2004;159(2):184-191. doi:10.1093/aje/kwh006PubMedGoogle ScholarCrossref
56.
Kim  TJ, Vonneilich  N, Lüdecke  D, von dem Knesebeck  O.  Income, financial barriers to health care and public health expenditure: a multilevel analysis of 28 countries.   Soc Sci Med. 2017;176:158-165. doi:10.1016/j.socscimed.2017.01.044PubMedGoogle ScholarCrossref
57.
Allin  S, Masseria  C, Mossialos  E.  Measuring socioeconomic differences in use of health care services by wealth versus by income.   Am J Public Health. 2009;99(10):1849-1855. doi:10.2105/AJPH.2008.141499PubMedGoogle ScholarCrossref
Original Investigation
Public Health
April 7, 2022

Socioeconomic Inequalities in Elective and Nonelective Hospitalizations in Older Men

Author Affiliations
  • 1Faculty of Medicine and Health, The University of Sydney School of Public Health, Sydney, New South Wales, Australia
  • 2ARC Centre of Excellence in Population Aging Research (CEPAR), University of Sydney, Sydney, New South Wales, Australia
  • 3Concord Clinical School, Faculty of Medicine and Health, University of Sydney, Sydney, New South Wales, Australia
  • 4Centre for Education and Research on Ageing, Faculty of Medicine and Health, University of Sydney, Sydney, New South Wales, Australia
  • 5Ageing and Alzheimer’s Institute, Concord Repatriation and General Hospital, Sydney Local Health District, Concord, New South Wales, Australia
  • 6ANZAC Research Institute, University of Sydney and Concord Hospital, Sydney, New South Wales, Australia
JAMA Netw Open. 2022;5(4):e226398. doi:10.1001/jamanetworkopen.2022.6398
Key Points

Question  Is socioeconomic position (SEP) associated with all-cause and cause-specific elective and nonelective hospitalizations among older men?

Findings  In this cohort study of 1566 men in Australia aged 70 years or older, low SEP as assessed by 3 individual SEP indicators and a cumulative score was associated with more all-cause nonelective hospitalizations and longer cumulative time in hospital for nonelective hospitalizations. No associations were found between SEP and elective hospitalizations, even though disadvantaged older people are likely to have a higher level of need.

Meaning  These findings point to the existence of socioeconomic inequalities in health care use, despite the Australian universal coverage public health care system.

Abstract

Importance  Among older adults, there is limited and inconsistent evidence on the association between socioeconomic position (SEP) and elective and nonelective hospitalization.

Objective  To evaluate the association between SEP and all-cause and cause-specific elective and nonelective hospitalization and hospital length of stay among older men.

Design, Setting, and Participants  This population-based, prospective cohort study used data from the Concord Health and Aging in Men Project (CHAMP). CHAMP recruited 1705 men aged 70 years or older between January 28, 2005, and June 4, 2007, in Sydney, Australia. Data were analyzed from February 1 to September 30, 2021.

Exposures  Indicators of SEP, including education (university degree certificate, diploma or no postschool qualifications), occupation (professionals and managers; small employers and self-employed; or lower clerical, service, sales workers, skilled, and unskilled workers), and source of income (other sources of income than government pension, reliance on government pensions and other sources of income, or reliant solely on a government pension), and a cumulative SEP score (tertiles) as SEP indicators; 3-level variables present high, intermediate, and low SEP.

Main Outcomes and Measures  All-cause and cause-specific elective and nonelective hospitalizations, number of hospitalizations, and length of stay were the study outcomes, ascertained through data linkage. Associations were quantified using competing-risks survival regression and negative binomial regression.

Results  A total of 1566 men (mean [SD] age, 76.8 [5.4] years) were included. During a mean (SD) 9.07 (3.53) years of follow-up, 1067 men had at least 1 elective hospitalization, and 1255 men had at least 1 nonelective hospitalization. No associations were found between SEP and elective hospitalizations. Being in the lowest tertile for educational level (subhazard ratio [SHR], 1.32; 95% CI, 1.11-1.58), occupational position (SHR, 1.30; 95% CI, 1.12-1.50), sources of income (SHR, 1.33; 95% CI, 1.17-1.52), and cumulative SEP tertile groups (SHR, 1.45; 95% CI, 1.24-1.68) were all associated with having at least 1 nonelective hospitalization compared with those in the highest tertiles. Significant associations were found between being in the lowest SEP groups and increased numbers and longer length of stay of nonelective hospitalizations.

Conclusions and Relevance  In this prospective cohort study, low SEP was inversely associated with nonelective hospitalizations but not elective hospitalization in older men in Australia. These findings point to the existence of socioeconomic inequalities in health care use, indicative of a need to take action to reduce these inequalities.

Introduction

The rapid growth of aging populations globally has raised concerns about the health and health care needs of older adults.1 Hospitalization rates, the number of hospitalizations, and hospital expenditure increase with age.2 Equity of access to health care should be within reach of all to achieve best health in older people.3

The Australian public health care system is universally accessible under Medicare, which is funded by the Australian government through taxation revenue and offers fee-free treatment for public patients in a public hospital.4 Private health insurance is not compulsory in Australia, and approximately half of the older population has private insurance. Insurance fully or partly covers the costs of being admitted to a private hospital and/or the costs of other ancillary health services that are not subsidized by Medicare.4

Although the Australian public health care system has a high overall performance, it is not free of inequities in access to health services. In Australia, although public hospitals account for most nonelective (ie, unplanned) admissions, a large proportion of admissions to private hospitals are elective, which is mainly funded by private health insurance. Waiting times for elective surgery are usually shorter in a private hospital than in a public hospital, where the median waiting time was 39 days in 2019 to 2020.5 In other words, those with private insurance can generally afford to use private hospitals for more timely treatment for elective surgery. Inequalities in access to health services could also be due to gaps in services covered and out-of-pocket costs for services that are not covered by Medicare, or when the fee charged by service providers is higher than the rebate paid by Medicare.6,7 Socioeconomic position (SEP) has been proposed as an indicator of inequities in access to health services.8

Among older adults, in few studies, low SEP has been shown to be associated with a higher rate and risk of any hospitalization (elective and nonelective).9-12 Moreover, there may be differences in the influence of SEP on elective compared to nonelective hospitalizations. Low SEP was associated with a higher risk of nonelective hospitalizations among adults and older adults in Scotland13 and Ireland,14 but another study reported no association between social class and elective hospitalizations among adults in the UK aged 45 to 65 years.15

Given the increased rate of hospitalization among older Australians in the past decade (3.3% per year), high hospitalization expenditure among Australian men,16 and the limited evidence about inequitable access to hospital care based on SEP, we aimed to examine the associations of education, occupation, income, and cumulative SEP with all-cause and cause-specific elective and nonelective hospitalization among older men in Australia who lived within a defined metropolitan geographical area. We also examined the association between SEP indicators and the number of elective and nonelective hospitalizations and hospital length of stay (LOS).

Methods
Study Population

The Concord Health and Aging in Men Project (CHAMP) is a population-based cohort study among 1705 men aged 70 years or older living in the Local Government Areas of Burwood, Canada Bay, and Strathfield in Sydney, Australia (participation rate of 54%).17 CHAMP began with baseline data collection in 2005 to 2007, with 4 subsequent follow-up assessments. The only baseline exclusion criterion was living in an aged care facility. The CHAMP study complied with the World Medical Association Declaration of Helsinki18 and was approved by the Sydney South West Area Health Service Human Research Ethics Committee. Written informed consent was obtained from all participants involved in the study. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

Assessment of Socioeconomic Indicators at Baseline

We used 3 self-reported SEP indicators from early adult life to older age (education, occupation, and source of income). Education was assessed based on the highest qualification attained and classified as high (university degree), intermediate (trade, apprenticeship, certificate, or diploma), and low (no postschool qualification). The longest occupation performed during working life and the Australian and New Zealand Standard Classification of Occupations were used to create 3 main categories: high (higher professionals and managers, lower professionals and managers, and higher clerical service), intermediate (small employers and self-employed, farmers, lower supervisors, and technicians), and low (lower clerical, service, sales workers, and skilled and unskilled workers).19 Australia’s retirement income system comprises 3 pillars: a means-tested age pension, mandatory occupational superannuation, and voluntary long-term savings. Participants reported their sources of income by selecting from the following 6 response options (more than 1 response allowed): age pension; repatriation pension or veteran’s pension; superannuation or other private income; own business farm, or partnership; wage or salary; and other (please specify). Subsequently, sources of income were classified as high (sources of income do not include any government pension), intermediate (reliance on government pensions and other sources of income), and low (reliant solely on a government pension).

To assess the cumulative consequences of being exposed to low SEP across the life course, a cumulative SEP score was computed by summing the 3 abovementioned SEP indicators coded as 0 (high), 1 (intermediate), and 2 (low). The cumulative SEP score ranged between 0 and 6 and was divided into tertiles: high (score 0-2), intermediate (score 3-4), and low (score 5-6); a higher score indicates a greater disadvantaged SEP.

Outcome Ascertainment

Consenting participants (n = 1639, 96%) were successfully followed up for any hospital admission using the New South Wales Admitted Patient Data Collection for all public and private hospitals in New South Wales, Australia, from January 1, 2005, to December 31, 2017.20 Records were linked to CHAMP baseline data by the Centre for Health Record Linkage 21 using probabilistic record linkage methodology and ChoiceMaker software (ChoiceMaker LLC). Death data were also obtained by the Centre for Health Record Linkage from the New South Wales Registry of Births, Deaths, and Marriages.

The primary outcomes were any first elective and nonelective hospitalizations following recruitment to CHAMP, defined as at least an overnight hospital stay; a given participant could be included for both elective and nonelective analysis. Elective and nonelective hospitalizations were differentiated by coding of urgency of admission. Cause-specific first elective and nonelective hospitalizations were coded according to the International Classification of Diseases and Related Health Problems, Tenth Revision, Australian Modification (ICD-10-AM).22 We focused on 14 major categories and chapters based on the organ system involved: infectious, neoplasm, hematologic, endocrine, psychiatric, neurologic, circulatory, respiratory, digestive, skin-related diseases, musculoskeletal, genitourinary, injury, and abnormal symptoms (signs, abnormal results of clinical or other research procedures, and ill-defined conditions that are not classified elsewhere)23 (eTable 1 in the Supplement). We report associations for the top 5 most frequent causes. Numbers of recorded elective and nonelective hospitalizations were also examined. The length of stay (the number of days between hospital admission and hospital discharge) of first all-cause elective and nonelective and the cumulative LOS of all elective and nonelective hospitalizations during follow-up were secondary outcomes.

Assessment of Covariates

We included age, country of birth (Australian-born or other), and marital status (single; married or de facto; and widowed, separated, or divorced) as potential confounders. A priori, we decided not to adjust for most established hospitalization risk factors, including frailty and chronic conditions, as these factors are likely to be intermediate variables of the association between SEP and hospitalizations.24

Statistical Analysis

Statistical analyses were performed using R, version 1.3.1056 (R Foundation for Statistical Computing); R packages include cmprsk,25 survival,26 and MASS.27 The associations between SEP and participants having at least 1 all-cause and cause-specific elective and nonelective hospitalizations were examined in separate models using Fine and Gray competing-risks survival regression.28 Adjusted subhazard ratios (SHRs) and 95% CIs are reported. Time at risk was calculated from date of baseline data collection to death, date of official withdrawal from the study, or December 31, 2017. All Fine and Gray competing-risks survival regression models were adjusted for age, age squared (to consider accelerated hospitalization rate by age), country of birth, and marital status. The proportional subhazards assumption was assessed using Schoenfeld residuals, and in all models this assumption was satisfied.29 Linear trends were assessed by using orthogonal polynomial contrasts. Rate ratios (RRs) for number of hospitalizations (elective and nonelective) were estimated using negative binomial regression analysis. We also used negative binomial regression to estimate the RRs and 95% CIs of the association of SEP with first elective and nonelective hospital LOS and cumulative LOS (total count of overnight hospital stays during follow-up); zero values were included for those who had no hospitalization. Negative binomial regression considers the skewness of the outcome and allows for overdispersion. For numbers of hospitalizations and cumulative LOS, we used a logarithm function of follow-up duration as the offset variable. We examined the interaction between SEP and country of birth for all analyses using the Wald test. Statistical significance was considered for a 2-sided test with a P < .05 for interaction.

Results

The study sample included 1566 participants (mean [SD] age, 76.8 [5.4] years); the demographic characteristics of participants by SEP are shown in the Table. Of 1705 participants at baseline, we excluded 139 participants who did not agree to data linkage and those who had missing data for SEP or covariates (eFigure 1 in the Supplement). Excluded participants tended to be older, born outside Australia, and have lower levels of education, occupation, and sources of income compared with included participants (eTable 2 in the Supplement).

Elective Hospitalization
Association of SEP With All-Cause and Cause-Specific Elective Hospitalization

During a mean (SD) 9.07 (3.53) years of follow-up, 1067 of 1566 participants had at least 1 elective hospitalization (eTable 3 in the Supplement), with the median first elective LOS of 3.0 days (IQR, 1.0-8.0 days). There was no statistically significant association between SEP indicators and having at least 1 elective hospitalization (Figure 1). Out of 14 major categories of organ system, neoplasm, circulatory, digestive, musculoskeletal, and genitourinary were the most frequent causes for first elective hospitalization (eFigure 2 in the Supplement). No statistically significant associations were evident between SEP indicators and cause-specific elective hospitalization (eFigure 3 in the Supplement).

Association of SEP With Number of Elective Hospitalizations

Overall, there were 2871 records of elective hospitalizations during follow-up, with the median cumulative LOS of 10.0 days (IQR, 3.0-24.0 days). No association was evident between SEP indicators and the number of elective hospitalizations (Figure 2).

Association of SEP With LOS of Elective Hospitalizations

No associations were evident between SEP indicators and LOS of first elective hospitalization, except for education (compared with the high educational level, relative risks [RRs], 1.29; 95% CI, 1.00-1.67 for intermediate and 1.28; 95% CI, 0.98-1.66 for low) and occupational position (compared with the high occupation level, RRs, 1.17; 95% CI, 0.96-1.43 for intermediate and 1.28; 95% CI, 1.04-1.57 for low) (Figure 3). For cumulative LOS of elective hospitalizations, the only association of an SEP indicator was for education (compared with the high education level, RRs, 1.29; 95% CI, 0.95-1.75 for intermediate and 1.61; 95% CI, 1.18-2.20 for low) (Figure 4).

Nonelective Hospitalization
Association of SEP With All-Cause and Cause-Specific Nonelective Hospitalization

During follow-up, 1255 of 1566 participants had at least 1 nonelective hospitalization (eTable 3 in the Supplement), with a median first nonelective LOS of 5.0 days (IQR, 3.0-11.0 days). There was an inverse association across SEP groups as assessed by educational level, occupational position, sources of income, and cumulative SEP and having at least 1 nonelective hospitalization (P < .001 for trend) (Figure 1). Out of 14 major categories of organ system, circulatory, injury, abnormal symptoms, respiratory, and digestive were the most frequent causes for first nonelective hospitalization (eFigure 2 in the Supplement). The only statistically significant associations were between being in the intermediate (SHR, 1.92; 95% CI, 1.20-3.08) or low (SHR, 1.96; 95% CI, 1.31-2.93) source of income groups and having higher risk of nonelective hospitalization because of respiratory conditions (eFigure 4 in the Supplement). There was a tendency toward a higher risk of nonelective hospitalization due to circulatory conditions among those with low educational level and occupational position, although this finding was not statistically significant.

Association of SEP With Number of Nonelective Hospitalizations

Overall, there were 5102 records of nonelective hospitalizations during follow-up, with the median cumulative LOS of 26.0 days (IQR, 9.0-60.5 days). There were statistically significant associations between being in the lowest educational level (RR, 1.59; 95% CI, 1.32-1.92 vs high educational level), occupational position (RR, 1.30; 95% CI, 1.12-1.51 vs high occupation level), sources of income (RR, 1.40; 95% CI, 1.23-1.59 vs high source of income level), and cumulative SEP (RR, 1.45; 95% CI, 1.24-1.69 vs high cumulative SEP) and having an increased number of nonelective hospitalizations (Figure 2).

Association of SEP With LOS of Nonelective Hospitalizations

Significant associations were evident between being in the lowest SEP groups and having longer LOS of first nonelective hospitalization (RRs, 1.28; 95% CI, 1.02-1.60 for lowest vs the highest educational level; 1.18; 95% CI, 0.99-1.42 for the lowest vs the highest occupation; 1.27; 95% CI, 1.08-1.48 for the lowest vs the highest income; and 1.34; 95% CI, 1.11-1.62 for the lowest vs the highest cumulative SEP tertile) (Figure 3). Moreover, being in low SEP groups was associated with a higher risk of having longer cumulative LOS of nonelective hospitalizations (RRs, 2.01; 95% CI, 1.55-2.62 for lowest vs the highest educational level; 1.45; 95% CI, 1.18-1.79 for the lowest vs the highest occupation; 1.43; 95% CI, 1.19-1.72 for the lowest vs the highest income; and 1.69; 95% CI, 1.36-2.11 for the lowest vs the highest cumulative SEP tertile) (Figure 4).

Discussion

In this cohort study, we found that being in low SEP groups was associated with at least 1 nonelective hospitalization, number of nonelective hospitalizations, and cumulative LOS for nonelective hospitalizations. Of all the most common cause-specific hospitalizations, the only statistically significant finding was an inverse association between source of income and nonelective hospitalization due to respiratory conditions. We found no association of SEP indicators with having at least 1 elective hospitalization, cause-specific elective hospitalization, number of elective hospitalizations, and LOS of elective hospitalizations.

Current Findings in Context of Other Evidence
Association of SEP With All-Cause and Cause-Specific Elective and Nonelective Hospitalization

We found that having at least 1 elective hospitalization was not socially patterned, which agrees with the only published prospective study among UK adults (aged 45 to 65 years) evaluating the association between social class and planned hospitalization.15 We also found no association between SEP indicators and number of elective hospitalizations. Given the greater prevalence of some reasons for elective hospitalization (eg, cardiac procedures, joint replacement) and of risk factors for hospitalizations among those in low SEP groups (eg, comorbidities), these results suggest that access to elective hospitalizations is not in proportion to need—with the least advantaged groups being underserved and the most advantaged groups overusing services.3 To our knowledge, no other studies have focused on socioeconomic inequalities in cause-specific elective hospitalization.

We found associations between low SEP and nonelective hospitalization. Our results are consistent with previous studies showing that socioeconomic deprivation is inversely associated with unplanned or emergency admissions among adults and older adults.13,14 We found that cumulative exposure to socioeconomic disadvantage over the life course was associated with nonelective hospitalization, mainly driven by the effect of education and income; to our knowledge, there have been no previous studies of this association.

Considering each SEP indicator, limited evidence exists for nonelective hospitalizations; our study adds to this body of evidence by examining and comparing patterns for both elective and nonelective hospitalizations. Similar to the previously published studies on any hospitalizations,30-32 we found a significant association between having a low educational level and higher risk of nonelective hospitalization. To date, there is no study examining the association between occupational position and nonelective hospitalization among older adults. However, the higher risk of nonelective hospitalization among those with low occupational position in our study is consistent with previous studies on any hospitalizations among young adults in Spain33 and Finland.34 In addition, our study found higher risk of nonelective hospitalizations for men reliant on a government pension solely (a relatively low level of income), which is in line with a retrospective cohort study among patients 65 years or older in the Netherlands11 and a study among 339 US adults with a mean age of 72.3 years focusing on any hospitalization.35

Regarding cause-specific nonelective hospitalization, we found an association between sources of income and nonelective hospitalization due to respiratory conditions, which is in agreement with previous studies on any cause-specific hospitalizations in Europe.10,36 We also found a tendency toward a higher risk of nonelective hospitalization because of respiratory diseases among those with a low educational level which agrees with results from other studies on any hospitalizations.36,37 These associations could be explained potentially in part by higher smoking rates among lower socioeconomic groups.38

Our results showed a tendency (not statistically significant) toward higher risk of nonelective hospitalization due to circulatory conditions among those with low educational level and occupational position, which agrees with previous studies in Europe, UK, and the US.10,15,39,40 This finding is consistent with the well-established association between low SEP and increased risk of cardiovascular disease.41

Association of SEP With LOS

Research on the association between SEP and longer LOS for nonelective hospitalizations among community-dwelling older adults is limited. Our study shows that having a low educational level and occupational position, having a limited source of income, and being in the lowest tertile of cumulative SEP were significantly associated with a longer cumulative LOS for nonelective hospitalizations (ie, spending more time in hospital). These results are consistent with previous studies among older adults in Canada,9 Rome,10 and the Netherlands.11

Possible Mechanisms

Overall, the inverse association between SEP and risk of hospitalization has several potential explanations. Socioeconomically disadvantaged individuals are more likely to be exposed to unhealthy diets, tobacco consumption, and low physical activity levels and therefore accumulate risk factors of morbidity42,43 that are of greater complexity and severity.15,44,45 Moreover, higher levels of nonelective hospitalization among those with low SEP may also be related to lack of health literacy, health care seeking behavior, and having low levels of social support.46-48 Previous CHAMP studies have shown that health-related behaviors, multimorbidity, and social support are patterned by socioeconomic indicators.48-50 Moreover, more people with a low SEP than a high SEP lack private health insurance. This is an important issue, as having private health insurance enhances access to timely elective health care, which might reduce the number of nonelective hospitalizations.51 Persons without private health insurance would have to pay the complete cost of private hospital admissions (other than the Medicare rebate for a clinician’s services) or wait to have their elective hospitalization in a public hospital. High out-of-pocket cost of health care for privately performed elective procedures in Australia could be another explanation.52 Increasing copayments lead to people forgoing medical care, mainly among the least advantaged groups.53 Further studies are needed to examine the mediating role of multimorbidity, poor health-related behaviors, and health insurance status on SEP inequalities in hospitalizations.

Indicators of SEP can influence hospitalizations differently. Occupational status could have an important effect on the health of older adults.45 Low occupation level might cause physical (eg, osteoarthritis) and psychosocial risks and subsequently lead to a higher risk of hospitalization (eg, elective joint replacement surgery).54 Finally, income directly affects the availability of health resources that can reduce the risk and number of hospitalizations.55 Moreover, forgoing planned health care owing to financial reasons (ie, out-of-pocket payments) is highly prevalent worldwide and in Australia.56

Strengths and Limitations

To our knowledge, this is the first study evaluating the association of 3 individual SEP indicators as well as a cumulative SEP score from early adult life to older age with all-cause and cause-specific elective and nonelective hospitalization. The study had the benefit of a long follow-up period and high-quality record linkage. Moreover, men recruited to CHAMP are similar to the general population of older men in Australia.17 We considered death as a competing risk factor for any and cause-specific elective and nonelective hospitalization.

This study also has some limitations. First, CHAMP only included men 70 years or older, so the results may not be generalizable to women or people younger than 70 years. These results may also not be generalizable to rural and regional populations. Second, although we included 3 individual indicators of SEP and 1 cumulative SEP score, we lack data on wealth, which is a better SEP indicator at older age to capture variations in financial security.57 Moreover, we do not have data on working conditions, such as job stability, working hours, and shift work, which may more comprehensively reflect the occupational position of an individual.49 We also do not have data on private insurance; having private insurance will allow people, regardless of SEP, to enhance access to special health services and timely access to health care.51 Third, hospitalizations were defined as at least 1 overnight hospital stay; thus, the results may not be generalizable to day-only admissions. Fourth, owing to the small number of cause-specific hospitalization, the power of our study was low for some outcomes.

Conclusions

In this cohort study, low SEP as assessed by 3 individual SEP indicators and a cumulative score was associated with more all-cause nonelective hospitalizations and longer cumulative time in hospital for nonelective hospitalizations among older men in Australia. No associations were observed for elective hospitalization, even though disadvantaged older people are likely to have a higher level of need. These findings point to the existence of socioeconomic inequalities in health care use, indicative of a need to take action to reduce these inequalities.

Back to top
Article Information

Accepted for Publication: February 20, 2022.

Published: April 7, 2022. doi:10.1001/jamanetworkopen.2022.6398

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Xu P et al. JAMA Network Open.

Corresponding Author: Saman Khalatbari-Soltani, PhD, Faculty of Medicine and Health, The University of Sydney School of Public Health, New South Wales, Australia (saman.khalatbarisoltani@sydney.edu.au).

Author Contributions: Dr Khalatbari-Soltani had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Blyth, Naganathan, Cumming, Handelsman, Seibel, Le Couteur, Waite, Khalatbari-Soltani.

Acquisition, analysis, or interpretation of data: Xu, Blyth, Naganathan, Cumming, Handelsman, Le Couteur, Waite, Khalatbari-Soltani.

Drafting of the manuscript: Xu, Khalatbari-Soltani.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Xu, Handelsman, Khalatbari-Soltani.

Obtained funding: Naganathan, Seibel, Le Couteur, Waite.

Administrative, technical, or material support: Blyth, Naganathan, Handelsman, Le Couteur, Khalatbari-Soltani.

Supervision: Blyth, Naganathan, Cumming, Khalatbari-Soltani.

Conflict of Interest Disclosures: Prof Seibel reported receiving grants from the National Health and Medical Research Council of Australia during the conduct of the study. Prof Waite reported receiving grants from the National Health and Medical Research Council of Australia during the conduct of the study. No other disclosures were reported.

Funding/Support: The CHAMP study is funded by the National Health and Medical Research Council of Australia (project grant no. 301916), the Ageing and Alzheimer's Institute, and the Sydney Medical School Foundation. Dr Khalatbari-Soltani was supported by the Australian Research Council Centre of Excellence in Population Aging Research (project no. CE170100005).

Role of the Funder/Sponsor: The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Additional Contributions: The authors are grateful to all the participants, and staff of the CHAMP study.

References
1.
Beard  JR, Officer  A, de Carvalho  IA,  et al.  The world report on ageing and health: a policy framework for healthy ageing.   Lancet. 2016;387(10033):2145-2154. doi:10.1016/S0140-6736(15)00516-4PubMedGoogle ScholarCrossref
2.
He  W, Goodkind  D, Kowal  P. An Aging World. US Census Bureau; 2016.
3.
Braveman  P.  Health disparities and health equity: concepts and measurement.   Annu Rev Public Health. 2006;27:167-194. doi:10.1146/annurev.publhealth.27.021405.102103PubMedGoogle ScholarCrossref
4.
Australian Institute of Health and Welfare. Health System Overview. July 23, 2020. Accessed December 10, 2021. https://www.aihw.gov.au/reports/australias-health/health-system-overview
5.
Australian Institute of Health and Welfare.  Admitted patient care 2017–18: Australian hospital statistics. AIHW; 2019.
6.
Furler  JS, Harris  E, Chondros  P, Powell Davies  PG, Harris  MF, Young  DYL.  The inverse care law revisited: impact of disadvantaged location on accessing longer GP consultation times.   Med J Aust. 2002;177(2):80-83. doi:10.5694/j.1326-5377.2002.tb04673.xPubMedGoogle ScholarCrossref
7.
van Doorslaer  E, Masseria  C, Koolman  X; OECD Health Equity Research Group.  Inequalities in access to medical care by income in developed countries.   CMAJ. 2006;174(2):177-183. doi:10.1503/cmaj.050584PubMedGoogle ScholarCrossref
8.
Andersen  R, Newman  JF.  Societal and individual determinants of medical care utilization in the United States.   Milbank Q. 2005;83(4):10.1111/j.1468-0009.2005.00428.x. doi:10.1111/j.1468-0009.2005.00428.xGoogle Scholar
9.
Amegbor  PM, Plumb  KB, Rosenberg  MW.  Determinants of overnight stay in health centres and length of admission: a study of Canadian seniors.   Can J Aging. 2020;39(4):533-544. doi:10.1017/S0714980819000771PubMedGoogle ScholarCrossref
10.
Antonelli-Incalzi  R, Ancona  C, Forastiere  F, Belleudi  V, Corsonello  A, Perucci  CA.  Socioeconomic status and hospitalization in the very old: a retrospective study.   BMC Public Health. 2007;7(1):227. doi:10.1186/1471-2458-7-227PubMedGoogle ScholarCrossref
11.
Wachelder  JJH, van Drunen  I, Stassen  PM,  et al.  Association of socioeconomic status with outcomes in older adult community-dwelling patients after visiting the emergency department: a retrospective cohort study.   BMJ Open. 2017;7(12):e019318-e019318. doi:10.1136/bmjopen-2017-019318PubMedGoogle ScholarCrossref
12.
Hart  CL, McCartney  G, Watt  GCM.  Occupational class differences in later life hospital use by women who survived to age 80: the Renfrew and Paisley prospective cohort study.   Age Ageing. 2015;44(3):515-519. doi:10.1093/ageing/afu184PubMedGoogle ScholarCrossref
13.
Payne  RA, Abel  GA, Guthrie  B, Mercer  SW.  The effect of physical multimorbidity, mental health conditions and socioeconomic deprivation on unplanned admissions to hospital: a retrospective cohort study.   CMAJ. 2013;185(5):E221-E228. doi:10.1503/cmaj.121349PubMedGoogle ScholarCrossref
14.
Cournane  S, Byrne  D, Conway  R, O’Riordan  D, Coveney  S, Silke  B.  Social deprivation and hospital admission rates, length of stay and readmissions in emergency medical admissions.   Eur J Intern Med. 2015;26(10):766-771. doi:10.1016/j.ejim.2015.09.019PubMedGoogle ScholarCrossref
15.
McCartney  G, Hart  C, Watt  G.  How can socioeconomic inequalities in hospital admissions be explained? a cohort study.   BMJ Open. 2013;3(8):e002433. doi:10.1136/bmjopen-2012-002433PubMedGoogle Scholar
16.
Australian Institute of Health and Welfare. Australia’s health 2020: in brief. July 23, 2020. Accessed December 10, 2021. https://www.aihw.gov.au/getmedia/2aa9f51b-dbd6-4d56-8dd4-06a10ba7cae8/aihw-aus-232.pdf.aspx?inline=true
17.
Cumming  RG, Handelsman  D, Seibel  MJ,  et al.  Cohort profile: the Concord Health and Ageing in Men Project (CHAMP).   Int J Epidemiol. 2009;38(2):374-378. doi:10.1093/ije/dyn071PubMedGoogle ScholarCrossref
18.
World Medical Association.  World Medical Association Declaration of Helsinki: ethical principles for medical research involving human subjects.   JAMA. 2013;310(20):2191-2194. doi:10.1001/jama.2013.281053Google ScholarCrossref
19.
Trewin D, Pink B. ANZSCO: Australian and New Zealand standard classification of occupations. Australian Bureau of Statistics, Statistics New Zealand. September 11, 2006. Accessed March 7, 2022. http://www.jobmarkets.com.au/doc/ANZSCO_2006_pdf.pdf
20.
Khalatbari-Soltani  S, Blyth  FM, Naganathan  V,  et al.  Cohort Profile update: the Concord Health and Ageing in Men Project (CHAMP).   Int J Epidemiol. 2022;51(1):31-32h. doi:10.1093/ije/dyab171PubMedGoogle ScholarCrossref
21.
Centre for Health Record Linkage. Accessed February 8, 2022. https://www.cherel.org.au/
22.
The International Statistical Classification of Diseases and Related Health Problems.  Tenth Revision, Australian Modification (ICD-10-AM). 7th ed. National Centre for Classification in Health; 2010.
23.
World Health Organization. International Statistical Classification of Diseases and Related Health Problems 10th Revision; 2019. Accessed February 8, 2022. https://icd.who.int/browse10/2019/en
24.
Hernán  MA, Hernández-Díaz  S, Werler  MM, Mitchell  AA.  Causal knowledge as a prerequisite for confounding evaluation: an application to birth defects epidemiology.   Am J Epidemiol. 2002;155(2):176-184. doi:10.1093/aje/155.2.176PubMedGoogle ScholarCrossref
25.
Gray  B. cmprsk: Subdistribution analysis of competing risks. 2020. Accessed February 8, 2022. https://cran.r-project.org/package=cmprsk
26.
Therneau  TM. A package for survival analysis in R. 2021. Accessed February 8, 2022. https://cran.r-project.org/package=survival
27.
Venables  WN, Ripley  BD, Ripley  BD.  Modern Applied Statistics with S. 4th ed. Springer; 2003.
28.
Fine  JP, Gray  RJ.  A proportional hazards model for the subdistribution of a competing risk.   J Am Stat Assoc. 1999;94(446):496-509. doi:10.1080/01621459.1999.10474144Google ScholarCrossref
29.
Schoenfeld  D.  Chi-squared goodness-of-fit tests for the proportional hazards regression model.   Biometrika. 1980;67(1):145-153. doi:10.1093/biomet/67.1.145Google ScholarCrossref
30.
Tung  H-H, Lu  T-M, Chen  L-K, Liang  S-Y, Wu  S-F, Chu  K-H.  Health literacy impact on elderly patients with heart failure in Taiwan.   J Clin Gerontol Geriatr. 2014;5(3):72-76. doi:10.1016/j.jcgg.2014.01.005Google ScholarCrossref
31.
Carlucio da Silva  FB, Filipini  R, Guarnieri  AP,  et al.  Sociodemographic profile and health condition of elderly patients attended at a community primary health center.   J Clin Gerontol Geriatr. 2016;7(3):93-98. doi:10.1016/j.jcgg.2016.03.003Google ScholarCrossref
32.
Sarwar  MR, Iftikhar  S, Sarfraz  M.  Influence of education level of older patients on polypharmacy, potentially inappropriate medications listed in Beer’s Criteria, and unplanned hospitalization: a cross-sectional study in Lahore, Pakistan.   Medicina (Kaunas). 2018;54(4):57. doi:10.3390/medicina54040057PubMedGoogle ScholarCrossref
33.
Pujol  J, Godoy  P, Soldevila  N,  et al; CIBERESP Cases and Controls in Pandemic Influenza Working Group.  Social class based on occupation is associated with hospitalization for A(H1N1)pdm09 infection. Comparison between hospitalized and ambulatory cases.   Epidemiol Infect. 2016;144(4):732-740. doi:10.1017/S0950268815001892PubMedGoogle ScholarCrossref
34.
Heliövaara  M.  Occupation and risk of herniated lumbar intervertebral disc or sciatica leading to hospitalization.   J Chronic Dis. 1987;40(3):259-264. doi:10.1016/0021-9681(87)90162-7PubMedGoogle ScholarCrossref
35.
Tertulien  T, Chen  Y, Althouse  AD, Essien  UR, Johnson  A, Magnani  JW.  Association of income and educational attainment in hospitalization events in atrial fibrillation.   Am J Prev Cardiol. 2021;7:100201. doi:10.1016/j.ajpc.2021.100201PubMedGoogle Scholar
36.
Prescott  E, Lange  P, Vestbo  J.  Socioeconomic status, lung function and admission to hospital for COPD: results from the Copenhagen City Heart Study.   Eur Respir J. 1999;13(5):1109-1114. doi:10.1034/j.1399-3003.1999.13e28.xPubMedGoogle ScholarCrossref
37.
Eagan  TML, Gulsvik  A, Eide  GE, Bakke  PS.  The effect of educational level on the incidence of asthma and respiratory symptoms.   Respir Med. 2004;98(8):730-736. doi:10.1016/j.rmed.2004.02.008PubMedGoogle ScholarCrossref
38.
Bosdriesz  JR, Willemsen  MC, Stronks  K, Kunst  AE.  Socioeconomic inequalities in smoking cessation in 11 European countries from 1987 to 2012.   J Epidemiol Community Health. 2015;69(9):886-892. doi:10.1136/jech-2014-205171PubMedGoogle ScholarCrossref
39.
Andersen  KK, Steding-Jessen  M, Dalton  SO, Olsen  TS.  Socioeconomic position and incidence of ischemic stroke in Denmark 2003-2012: a nationwide hospital-based study.   J Am Heart Assoc. 2014;3(4):e000762. doi:10.1161/JAHA.113.000762PubMedGoogle Scholar
40.
Vart  P, Coresh  J, Kwak  L, Ballew  SH, Heiss  G, Matsushita  K.  Socioeconomic status and incidence of hospitalization with lower-extremity peripheral artery disease: atherosclerosis risk in communities study.   J Am Heart Assoc. 2017;6(8):e004995. doi:10.1161/JAHA.116.004995PubMedGoogle Scholar
41.
Havranek  EP, Mujahid  MS, Barr  DA,  et al; American Heart Association Council on Quality of Care and Outcomes Research, Council on Epidemiology and Prevention, Council on Cardiovascular and Stroke Nursing, Council on Lifestyle and Cardiometabolic Health, and Stroke Council.  Social determinants of risk and outcomes for cardiovascular disease: a scientific statement from the American Heart Association.   Circulation. 2015;132(9):873-898. doi:10.1161/CIR.0000000000000228PubMedGoogle ScholarCrossref
42.
Turrell  G, Mathers  CD.  Socioeconomic status and health in Australia.   Med J Aust. 2000;172(9):434-438. doi:10.5694/j.1326-5377.2000.tb124041.xPubMedGoogle ScholarCrossref
43.
Dannefer  D.  Cumulative advantage/disadvantage and the life course: cross-fertilizing age and social science theory.   J Gerontol B Psychol Sci Soc Sci. 2003;58(6):S327-S337. doi:10.1093/geronb/58.6.S327PubMedGoogle ScholarCrossref
44.
Barnett  K, Mercer  SW, Norbury  M, Watt  G, Wyke  S, Guthrie  B.  Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study.   Lancet. 2012;380(9836):37-43. doi:10.1016/S0140-6736(12)60240-2PubMedGoogle ScholarCrossref
45.
Fors  S, Lennartsson  C, Lundberg  O.  Health inequalities among older adults in Sweden 1991-2002.   Eur J Public Health. 2008;18(2):138-143. doi:10.1093/eurpub/ckm097PubMedGoogle ScholarCrossref
46.
Gondek  D, Ploubidis  GB, Hossin  MZ, Gao  M, Bann  D, Koupil  I.  Inequality in hospitalization due to non-communicable diseases in Sweden: age-cohort analysis of the Uppsala Birth Cohort Multigenerational Study.   SSM Popul Heal. 2021;13:100741. doi:10.1016/j.ssmph.2021.100741PubMedGoogle Scholar
47.
Krause  N, Borawski-Clark  E.  Social class differences in social support among older adults.   Gerontologist. 1995;35(4):498-508. doi:10.1093/geront/35.4.498PubMedGoogle ScholarCrossref
48.
Khalatbari-Soltani  S, Stanaway  F, Cvejic  E,  et al.  Contribution of psychosocial factors to socioeconomic inequalities in mortality among older Australian men: a population-based cohort study.   Int J Equity Health. 2020;19(1):177. doi:10.1186/s12939-020-01277-2PubMedGoogle ScholarCrossref
49.
Khalatbari-Soltani  S, Blyth  FM, Naganathan  V,  et al.  Socioeconomic status, health-related behaviours, and death among older people: the Concord health and aging in men project prospective cohort study.   BMC Geriatr. 2020;20(1):261. doi:10.1186/s12877-020-01648-yPubMedGoogle ScholarCrossref
50.
Khalatbari-Soltani  S, Cumming  RG, Chomik  R,  et al.  The association between home ownership and the health of older men: Cross-sectional analysis of the Australian Concord Health and Ageing in Men Project.   Australas J Ageing. 2021;40(3)e199-e206. doi:10.1111/ajag.12896Google Scholar
51.
Brameld  KJ, Holman  CDJ.  The use of end-quintile comparisons to identify under-servicing of the poor and over-servicing of the rich: a longitudinal study describing the effect of socioeconomic status on healthcare.   BMC Health Serv Res. 2005;5:61. doi:10.1186/1472-6963-5-61PubMedGoogle ScholarCrossref
52.
Yusuf  F, Leeder  SR.  Can’t escape it: the out-of-pocket cost of health care in Australia.   Med J Aust. 2013;199(7):475-478. doi:10.5694/mja12.11638PubMedGoogle ScholarCrossref
53.
Yusuf  F, Leeder  S.  Recent estimates of the out-of-pocket expenditure on health care in Australia.   Aust Health Rev. 2020;44(3):340-346. doi:10.1071/AH18191PubMedGoogle ScholarCrossref
54.
Adler  NE, Newman  K.  Socioeconomic disparities in health: pathways and policies.   Health Aff (Millwood). 2002;21(2):60-76. doi:10.1377/hlthaff.21.2.60PubMedGoogle ScholarCrossref
55.
Kim  S, Symons  M, Popkin  BM.  Contrasting socioeconomic profiles related to healthier lifestyles in China and the United States.   Am J Epidemiol. 2004;159(2):184-191. doi:10.1093/aje/kwh006PubMedGoogle ScholarCrossref
56.
Kim  TJ, Vonneilich  N, Lüdecke  D, von dem Knesebeck  O.  Income, financial barriers to health care and public health expenditure: a multilevel analysis of 28 countries.   Soc Sci Med. 2017;176:158-165. doi:10.1016/j.socscimed.2017.01.044PubMedGoogle ScholarCrossref
57.
Allin  S, Masseria  C, Mossialos  E.  Measuring socioeconomic differences in use of health care services by wealth versus by income.   Am J Public Health. 2009;99(10):1849-1855. doi:10.2105/AJPH.2008.141499PubMedGoogle ScholarCrossref
×