Table 2 Strength of the evidence for the relationship between exposure to ambient PM2.5 and dementia outcomes analyzed

From: A systematic review with a Burden of Proof meta-analysis of health effects of long-term ambient fine particulate matter (PM2.5) exposure on dementia

Health outcome

RR for PM2.5 level of 15.5 μg m−3 (95% UI without between-study heterogeneity)

RR for PM2.5 level of 15.5 μg m−3 (95% UI with between-study heterogeneity)

BPRF

ROS

Star rating

Publication bias

No. of effect estimates

Selected bias covariates

Risk–outcome pair in GBD 2023

All dementia outcomes combined

1.37 (1.20–1.57)

1.37 (1.13–1.69)

1.14

0.13

☆☆

Yes

49

Covariate for AD (1, AD; 0, other dementia types other than AD)

Yes

Subtype: AD

1.39 (1.35–1.43)

1.39 (1.33–1.44)

1.38

0.32

☆☆☆

Yes

12

None

No

Subtype: VaD

3.15 (0.91–10.65)

3.13 (0.81–13.48)

–

–

–

No

6

Covariate for the selection bias (0, low risk; 1, medium risk; 2, high risk)

No

  1. The reported mean RR and its 95% UI reflect the risk an individual who has been exposed to the mean PM2.5 exposure level across all included studies (15.5 μg m−3) has of developing the outcome of interest relative to that of someone who has not been exposed to the reference PM2.5 level of 2.0 μg m−3. The health outcome ‘Dementia’ includes extracted effect estimates from all included studies, whereas AD and VaD only include effect estimates for the corresponding dementia subtypes. We report the 95% UI when not incorporating between-study heterogeneity and when accounting for between-study heterogeneity. The BPRF is calculated for risk–outcome pairs that were found to have significant relationships at a 0.05 level of significance when not incorporating between-study heterogeneity (the lower bound of the 95% UI without γ does not cross the null RR value of one). The BPRF corresponds to the fifth quantile estimate of RR accounting for between-study heterogeneity closest to the null for each risk–outcome pair, and it reflects the most conservative estimate of excess risk associated with ambient PM2.5 exposure that is consistent with the available data. The ROS is calculated as log(BPRF). Negative ROSs indicate that the evidence of the association is very weak and inconsistent. For ease of interpretation, we have transformed the ROS and BPRF into a star rating (1–5) with a higher rating representing a larger effect with stronger evidence. The potential existence of publication bias, which, if present, would affect the validity of the results, was tested using Egger’s regression. Included studies represent all available relevant data identified through our systematic reviews up to June 2023. The selected bias covariates were chosen for inclusion in the model using algorithm that systematically detects bias covariates that correspond to significant sources of bias in the observations included. If selected, the observations were adjusted to better reflect the gold-standard values of the covariate. The Supplementary Information provides more information about the candidate bias covariates that were selected for in each model. The risk–outcome pairs in this study were not included in GBD 2021 analyses but are planned for inclusion in the next iteration of GBD.
Back to article page