Skip to main content

Table 4 Univariable age- and sex-adjusted logistic regression analysis for the association between mortality and exposure variables

From: Scottish Index of Multiple Deprivation (SIMD) indicators as predictors of mortality among patients hospitalised with COVID-19 disease in the Lothian Region, Scotland during the first wave: a cohort study

Exposure Variable (SI units)

Range

Univariable Age and Sex-adjusted

Odds Ratio (95% CI) and P value

Performance Status, (WHO standard categories)

0

-

1

2.24 (1.08–4.78) P = 0.033

2

3.75 (1.78–8.24) P =  < 0.001

3

3.74 (1.81–8.09) P = 0.001

4

7.11 (2.85–18.30) P < 0.001

Heart rate on admission (beats per minute)

60–99

-

< 60

1.29 (0.44–3.67) P = 0.628

> 99

1.92 (1.28–2.88) P = 0.002

Haemoglobin concentration (grams/Litre)

> 129

-

< 100

2.92 (1.50–5.80) P = 0.002

100–129

0.82 (0.54–1.27) P = 0.331

Neutrophil count (cells × 105)

2–7.5

-

< 2

2.26 (1.10–4.59) P = 0.025

> 7.5

1.61 (1.07–2.43) P = 0.022

Lymphocyte count (cells × 105)

> 1.4

-

< 0.5

1.77 (0.96–3.30) P = 0.071

0.5–1.4

0.83 (0.50–1.39) P = 0.476

Creatinine (milligrams/decilitre)

< 125

-

> 125

2.46 (1.63–3.70) P < 0.001

SIMD quintile

5 (least deprived)

 

4

1.35 (0.78–2.33) P = 0.284

3

1.91 (1.06–3.45) P = 0.032

2

1.24 (0.75–2.05) P = 0.393

1 (most deprived)

1.61 (0.91–2.85) P = 0.102

Income deprivation rate per datazone (categorised by quartile)

Q1 (least deprived)

-

Q2

2.00 (1.17–3.46) P = 0.018

Q3

1.70 (0.99–2.96) P = 0.056

Q4 (most deprived)

2.05 (1.22–3.51) P = 0.007

Employment rate (categorised by quartile)

Q1 (least deprived)

-

Q2

0.91 (0.51–1.61) P = 0.756

Q3

1.41 (0.87–2.29) P = 0.168

Q4 (most deprived)

1.26 (0.80–2.00) P = 0.315

Comparative Illness Factor (Standardised Ratio)

< expected

-

> expected

1.24 (0.86–1.78) P = 0.250

Hospital admissions per datazone related to alcohol use (Standardised Ratio)

< expected

-

> expected

1.68 (1.12–2.48) P = 0.009

Hospital admissions per datazone related to drug use (Standardised Ratio)

< expected

-

> expected

1.33 (0.92–1.93) P = 0.126

Standardised mortality ratio (Standardised Ratio)

< expected

-

> expected

1.09 (0.77–1.57) P = 0.616

Proportion of population being prescribed drugs for anxiety, depression or psychosis per datazone (categorised into quartiles)

Q1 (least deprived)

-

Q2

0.97 (0.57–1.67) P = 0.908

Q3

1.47 (0.86–2.55) P = 0.162

Q4 (most deprived)

0.93 (0.54–1.61) P = 0.793

Standardised ratio of emergency stays in hospital (Standardised Ratio)

< expected

-

> expected

1.29 (0.89–1.87) P = 0.182

Working age people with no qualifications (Standardised Ratio)

< expected

-

> expected

0.97 (0.67–1.40) P = 0.886

Average drive time to a General Practitioner (GP) surgery in minutes (categorised into quartiles)

Q1 (least deprived)

-

Q2

0.87 (0.53–1.41) P = 0.567

Q3

1.10 (0.68–1.80) P = 0.692

Q4 (most deprived)

1.08 (0.66–1.76) P = 0.753

Average public transport travel time to a General Practitioner (GP) surgery in minutes (categorised into quartiles)

Q1 (least deprived)

-

Q2

1.15 (0.71–1.87) P = 0.576

Q3

1.18 (0.73–1.90) P = 0.493

Q4 (most deprived)

1.27 (0.74–2.17) P = 0.391

Percentage of people in households that are overcrowded (categorised into quartiles)

Q1 (least deprived)

-

Q2

0.73 (0.44–1.20) P = 0.217

Q3

0.89 (0.55–1.43) P = 0.619

Q4 (most deprived)

1.31 (0.79–2.19) P = 0.296

  1. Odds ratios, 95% confidence intervals and p-values from univariable, age and sex adjusted, logistic regression analysis for the association between mortality and clinical admission variables, SIMD quintile and 12 selected indicators within the SIMD. For the clinical variables: variables were categorised according to standard reference ranges, with normal values used as the reference. For SIMD quintiles: the least deprived quintile was the reference variable. For the 12 selected SIMD indicators: 1) standardised ratios in the SIMD were transformed into binary variables. Ratios represented observed occurrences divided by the predicted occurrences per datazone, where the reference value was 100, which is the Scotland average for a population with the same age and sex profile. Values above 100 were classed as “ > expected” and values below 100 were “ < expected”. 2) SIMD indicators that were continuous variables (percentages, proportions, or time in minutes) were categorised into quartiles with the least deprived quartile as the reference