Key results and quality assessments for studies investigating household income (n=13)
First author (Year) | Key results | Risk of Bias¶ | ||||||
---|---|---|---|---|---|---|---|---|
Household income | Association between SDoH and MM? | Value (95% CI, p value) | Comparator | Adjusted for… | Selection | Information (Exposure) | Information (Outcome) | Confounding |
Agborsangaya (2012)61 | Yes | OR 2.39**(1.72–3.33) | Annual household income <$30 k vs ≥$100 k CAD | Age, sex, education, living with children | H | M | M | L |
Agborsangaya (2013)62 | Yes | OR 2.9 (2.2–3.7) | Annual household income <$30 k vs ≥$100 k CAD | Age, sex, education, obesity | H | H | M | L |
Chung (2015)65 | Yes | OR 1.52 (1.39–1.66, p<0.001) | Monthly income <4 k vs >40 k HKD | Age, gender, education, housing, employment | H | M | M | L |
Hayek (2017)68 | Yes | PRR 1.7 (1.2–2.5, p=0.005) | Monthly income ≤$2 k vs >$4 k USD | Unclear | U | H | H | U |
Johnson-Lawrence (2017)69 | Yes | OR 1.45 (1.38–1.53) | Lowest income tertile vs highest | Age, gender, ethnicity, education, interview year, region, marital status, last doctor visit, employment, home ownership | U | M | H | L |
Katikireddi(2017)39 | Yes | OR 1.53 (1.25–1.87, p<0.05) | Lowest income† tertile vs highest | Age, age2, age3, sex, cohort, prior multimorbidity, time between waves and sex*cohort interaction | M | M | M | L |
Ki(2017)40 | Yes | OR 3.48* (3.20–3.78) | “Poor” (less than half the median annual household income†) vs “non-poor” | No adjustment | U | H | M | H |
Laires(2018)41 | Yes | OR 2.16* (1.95–2.40) | Lowest income† quintile vs highest | No adjustment | L | H | M | H |
Lebenbaum(2018)66 | Yes | OR 0.57 (0.52–0.62, p<0.001) | Highest income† quintile vs lowest | Age, age,2 sex, marital status, immigration status, education, rurality, homeownership, smoking, alcohol use | L | M | H | L |
Lujic (2017)43 | Yes | OR 0.58‡ (95% CI 0.52 to 0.66) | Income >$70 k vs <$20 k CAD | Age, sex | H | M | M | M |
Neilsen (2017)48 | Yes | OR 1.44 (1.32–1.59, p<0.05) | Lowest income tertile vs highest | Age, sex, education | U | H | M | L |
Prazeres (2015)52 | No | OR 0.8§ (0.5–1.1, p=0.182) | ‘Some monthly income left over’ vs ‘Not enough monthly income to make ends meet’ | Age, sex, marital status, education, professional status, residence area, living arrangement | H | M | L | L |
Roberts (2015)53 | Yes | OR 4.4 (3.6–5.5) | Lowest income quintiles vs highest | Age, sex, household education, Aboriginal status, activity level smoking, stress, blood pressure, obesity | H | M | H | M |
Schäfer(2012)60 | Yes | −0.27 conditions(−0.47 to −0.08, p=0.005) | Change per unit on income† scale (one unit=one of steps: €400 to €1100 to €3000 to €8100 net income per month) | Age, gender, marital status, job autonomy, household composition, income | H | M | L | U |
Verest(2019)59 | Yes | OR 5.36* , †† (4.88–5.88) | “Lots of problems” managing money vs “no problems” | No adjustment | H | H | M | H |
*OR calculated from data reported in paper.
†Income equivalised to account for number and/or age of residents in household.
‡Based on self-reported health data. Findings consistent across hospital and medication health data.
§Multimorbidity defined as ≥2 chronic conditions.
¶H, High; M, Medium; L, Low; U, Unclear.
**Inequalities greater for ages 25–44.
††Inequalities greater for women and similar by ethnicity group.