What is the evidence for the influence of socioeconomic status on the association between a combination of unhealthy lifestyle factors and adverse health?

Conference:

SAPC ASM 2021 - virtual

Talk Code:

1D.5

Presenter:

Hamish Foster

Twitter:

@hamishfoster

Co-authors:

Mr. Peter Polz, Prof. Jason M.R. Gill, Dr. Carlos Celis-Morales, Prof. Frances S. Mair, and Prof. Catherine A. O’Donnell

Author institutions:

General Practice and Primary Care, University of Glasgow

Problem

Combinations of unhealthy lifestyle factors (LFs) (e.g. smoking with high alcohol intake and poor diet) are associated with cardiovascular disease (CVD), cancer, and mortality. Low socioeconomic status (SES) is independently associated with the same outcomes. Low SES populations may be more vulnerable to the adverse effects of combinations of unhealthy LFs but interactions between combinations of LFs and SES remain poorly understood. This systematic review synthesised evidence for whether SES moderates associations between combinations of unhealthy LFs and adverse health outcomes.

Approach

Three databases (PubMed, EMBASE, CINAHL) were searched from inception– 03/03/2020. Inclusion criteria: 1) prospective cohorts; 2) investigation of combinations of ≥3LFs and SES; 3) data on SES influence on associations between LF combinations and outcomes; 4) outcomes included ≥1 of CVD, cancer, or mortality. Exclusion criteria: 1) ineligible study design; 2) not full-text publication; 3) not in English language. Results were synthesised without meta-analysis and compared across similar health outcomes. Analyses stratified by SES were used to examine hazard ratios (HRs) for those with the unhealthiest combination of LFs, comparing participants in the highest versus lowest SES strata. Results for tests for interactions between combinations of LFs and SES and results of additional analyses were collected.

Findings

Four studies of prospective cohorts were included (n=42,467–328,594; mean age 51.0–56.8 years; 54.6–59.3% women). Each study examined all-cause mortality and two studies also assessed CVD and cancer outcomes (average follow-up 4.3–19.3 years). LF variables, SES measures, and methods varied substantially between studies and precluded meta-analysis. For all-cause mortality, two studies identified multiplicative interactions between combinations of LFs and SES, but in opposing directions. In the remaining two studies, additional, combined non-SES-stratified analyses indicated an additive interaction. In the combined analyses, with the healthiest LF and highest SES group as reference, all-cause mortality HRs (95%CIs) for unhealthiest LF and lowest SES groups ranged from 1.43 (1.11-1.84) to 3.34 (2.79-3.99). In five of six combined analyses, the unhealthiest LF-lowest SES groups had the highest all-cause mortality risk. Results for other health outcomes were similar.

Consequences

The number of available studies is limited and their methodology heterogenous. However, their synthesis suggests that the combined associations of low SES and combinations of unhealthy LFs are likely to be additive for adverse health. Additional prospective analyses that use different datasets, comprehensive sets of LF variables, SES measures, and health outcomes would improve understanding of whether SES moderates the association between combinations of unhealthy LFs and adverse health. Improving the understanding of the relationships between lifestyle, SES, and adverse health would inform intervention and policy development.

Submitted by:

Hamish Foster

Funding acknowledgement:

HF is funded by a Medical Research Council (MRC) Clinical Research Training Fellowship (grant reference number MR/T001585/1). The MRC played no role in any part of this study. FSM, COD and JMRG are supervisors of the MRC Fellowship. CCM received no specific funding for this work. All authors declare no conflicts of interest.