The Odds Ratio Summary presents the highest to lowest odds ratios (left to right) across all age groups. The p-value is visualized using a heatmap, where lighter colors indicate less statistical significance. You can hover over each point in the graph for detailed information.
The Odds Ratio (OR) is calculated using a contingency table with the following formula:
OR = (a × d) / (b × c)
where:
- a = Number of patients with the primary disease and comorbidity
- b = Number of patients with the primary disease but without comorbidity
- c = Number of patients without the primary disease but with comorbidity
- d = Number of patients without the primary disease and without comorbidity
The odds ratio is a valuable metric for understanding the interaction between diseases. It provides insights beyond mere prevalence, helping decision-makers identify significant associations between conditions. Conducting a city-wide odds ratio analysis allows us to capture the unique epidemiological patterns within the population of Semarang City. This localized approach can inform public health strategies, resource allocation, and targeted interventions to address the specific health challenges faced by the community.
Groups:
- Male Age Groups:
- M0, M10, ... M90 or L0, L10, ... L90 represent males aged 0-9, 10-19, ..., 90+ years.
- Female Age Groups:
- F0, F10, ... F90 or P0, P10, ... P90 represent females aged 0-9, 10-19, ..., 90+ years.
Variables:
- ICD10 Code 1 and ICD10 Code 2: ICD-10 codes of diseases included in the odds ratio calculation. Only diseases with 1,000 or more patients were considered, resulting in 203 diseases. The odds ratio is calculated for each pair of these diseases.
- OR (Odds Ratio): Indicates the association between the primary disease and the comorbid condition. A higher OR suggests a stronger association.
- ci_lo (Confidence Interval Lower Bound): The lower limit of the 95% confidence interval.
- ci_hi (Confidence Interval Upper Bound): The upper limit of the 95% confidence interval.
- P-value: Measures statistical significance. Smaller values indicate stronger significance. Very small values are expressed in scientific notation (e.g., 5e-3 = 0.005).
Interpretation of OR, 95% CI, and P-value:
- OR > 1, 95% CI does not include 1, P-value < 0.05 → Significantly positive association
- OR < 1, 95% CI does not include 1, P-value < 0.05 → Significantly negative association
- OR = 1, 95% CI includes 1, P-value > 0.05 → No significant association
- OR ≠ 1, 95% CI does not include 1, P-value > 0.05 → Potentially significant but not statistically strong, possibly due to limited sample size