deltapif The logo of the deltapif method showing an observed and a counterfactual population with coloured squares. A line partitions them in the middle. The image reads 'deltapif'.

CRAN status Lifecycle: experimental R-CMD-check Codecov test coverage

The deltapif R package calculates Potential Impact Fractions (PIF) and Population Attributable Fractions (PAF) from aggregated data. It uses the delta method to derive confidence intervals, providing a robust approach for quantifying the burden of disease attributable to risk factors and the potential impact of interventions.

Installation

You can install the development version of deltapif from GitHub with:

remotes::install_github("RodrigoZepeda/deltapif")

Overview

The package provides two core functions:

Both functions require:

The pif() function additionally requires:

Note A key assumption of the delta method implementation is that the relative risk and exposure prevalence estimates are independent (i.e., derived from different studies or populations).

Usage

Population Attributable Fraction (PAF)

Lee et al. (2022) estimated the fraction of dementia cases attributable to smoking in the US. They reported:

The point estimate of the PAF can be calculated using Levin’s formula:

library(deltapif)

paf(p = 0.085, beta = log(1.59), quiet = TRUE)
#> 
#> ── Population Attributable Fraction: [deltapif-121642715990569] ──
#> 
#> PAF = 4.776% [95% CI: 4.776% to 4.776%]
#> standard_deviation(paf %) = 0.000

Incorporating Uncertainty

To calculate confidence intervals, we need the variance of the log-relative risk. The variance can be derived from the confidence interval following the Cochrane Handbook:

var_log_rr <- ((log(2.20) - log(1.15)) / (2 * 1.96))^2
var_log_rr
#> [1] 0.0273848

We then provide the log-relative risk (log(1.59)) and its variance to paf(), specifying the rr_link as exp to convert the coefficient to a relative risk by exponentiating the log. Since the prevalence variance was not reported, we assume var_p = 0.

paf_dementia <- paf(
  p         = 0.085, 
  beta      = log(1.59), 
  var_beta  = var_log_rr, 
  var_p     = 0
)
paf_dementia
#> 
#> ── Population Attributable Fraction: [deltapif-0305065688938981] ──
#> 
#> PAF = 4.776% [95% CI: 0.717% to 8.669%]
#> standard_deviation(paf %) = 2.028

The results match those reported by Lee et al.: PAF = 4.9% (95% CI: 1.3–9.3).

Potential Impact Fraction (PIF)

Lee et al. (2022) also considered a scenario reducing smoking prevalence by 15% (from 8.5% to 7.225%). The PIF for this intervention is:

lee_pif <- pif(
  p        = 0.085, 
  p_cft    = 0.085 * (1 - 0.15), # 15% reduction
  beta     = log(1.59), 
  var_beta = var_log_rr, 
  var_p    = 0
)
lee_pif
#> 
#> ── Potential Impact Fraction: [deltapif-0201324401270228] ──
#> 
#> PIF = 0.716% [95% CI: 0.118% to 1.311%]
#> standard_deviation(pif %) = 0.304

This result is consistent with the reported estimate: PIF = 0.7% (95% CI: 0.2–1.4).

Attributable and averted cases

Attributable and averted cases can be calculated with the attributable_cases function. For example Dhana et al estimate the number of people with Alzheimer’s Disease in New York, USA 426.5 (400.2, 452.7) thousand. This implies a variance of ((452.7 - 400.2) / 2*qnorm(0.975))^2 = 2647.005.

The number of cases (in thousands) that would be averted if we reduced smoking by 15% assuming the prevalence of smoking is identical to the rest of the US is given by:

averted_cases(426.5, lee_pif, variance = 2647.005)
#> 
#> ── Averted cases: [deltapif-0201324401270228] ──
#> 
#> Averted cases = 3.055 [95% CI: 0.394 to 5.716]
#> standard_deviation(averted cases) = 135.779

Attributable cases can likewise be estimated using the previous paf as:

attributable_cases(426.5, paf_dementia, variance = 2647.005)
#> 
#> ── Attributable cases: [deltapif-0305065688938981] ──
#> 
#> Attributable cases = 20.368 [95% CI: 2.626 to 38.109]
#> standard_deviation(attributable cases) = 905.195

Combining fractions from subpopulations

Multiple fractions can be combined into totals and ensembles. For example the fraction among men and women can be combined into an overall fraction by specifying the distribution of the subgroups in the population:

paf_men   <- paf(p = 0.41, beta = 0.31, var_p = 0.001,
                 var_beta = 0.14,
                 label = "Men")
paf_women <- paf(p = 0.37, beta = 0.35, var_p = 0.001, 
                 var_beta = 0.16,
                 label = "Women")

Assuming the distribution is 51% women and 49% men:

paf_total(paf_men, paf_women, weights = c(0.49, 0.51))
#> 
#> ── Population Attributable Fraction: [deltapif-0997466165469617] ──
#> 
#> PAF = 13.201% [95% CI: 10.473% to 15.845%]
#> standard_deviation(paf %) = 11.187
#> ────────────────────────────────── Components: ─────────────────────────────────
#> • 12.968% (sd %: 15.867) --- [Men]
#> • 13.424% (sd %: 15.773) --- [Women]
#> ────────────────────────────────────────────────────────────────────────────────

This is equivalent to calculating:

\[ \textrm{PAF}_{\text{All}} = 0.49 \cdot \text{PAF}_{\text{Men}} + 0.51 \cdot \text{PAF}_{\text{Women}} \]

Combining fractions from multiple risks

Fractions from disjointed risks can be calculated as an ensemble. For example the fraction of exposure to lead and the fraction of exposure to asbestus:

paf_lead  <- paf(p = 0.41, beta = 0.31, var_p = 0.001,
                 var_beta = 0.014,
                 label = "Lead")
paf_absts <- paf(p = 0.61, beta = 0.15, var_p = 0.001, 
                 var_beta = 0.001,
                 label = "Asbestus")

A fraction of environmental exposure considering both can be calculated by multiplying the inverse of the fractions, assuming a commonality correction (say of c(0.1, 0.2)):

paf_ensemble(paf_lead, paf_absts, weights = c(0.1, 0.2))
#> 
#> ── Population Attributable Fraction: [deltapif-132925867320251] ──
#> 
#> PAF = 3.070% [95% CI: 3.033% to 3.108%]
#> standard_deviation(paf %) = 0.625
#> ────────────────────────────────── Components: ─────────────────────────────────
#> • 12.968% (sd %: 5.085) --- [Lead]
#> • 8.985% (sd %: 1.904) --- [Asbestus]
#> ────────────────────────────────────────────────────────────────────────────────

where this quantity estimates:

\[ \textrm{PAF}_{\text{Ensemble}} = 1 - (1 - 0.1 \cdot \textrm{PAF}_{\text{Lead}}) \cdot (1 - 0.2 \cdot \textrm{PAF}_{\text{Asbestus}}) \]

Adjusting fractions for commonality

Adjuting for commonality is usually performed when different risks can be concurrent. In the previous example, exposure to lead and to asbestus can happen at the same time. Mukadan et al propose the individual weighted (adjusted) fractions based on commonality weights. These weights represent the proportion of the variance shared among risk factors. To calculate the adjusted fractions one needs to estimate:

\[ \textrm{PIF}_k^{\text{Adjusted}} = \dfrac{\text{PIF}_k}{\sum_k \text{PIF}_k} \cdot \text{PIF}_{\text{Overall}} \] where

\[ \textrm{PIF}^{\text{Overall}} = 1 - \prod\limits_k (1 - w_k \text{PIF}_k) \] with

\[ w_k = 1 - \text{commonality}_k \]

The adjusted fractions can be calculated with the weighted_adjusted as:

weighted_adjusted_paf(paf_lead, paf_absts, weights = c(0.2, 0.3))
#> $Lead
#> 
#> ── Population Attributable Fraction: [Lead_adj] ──
#> 
#> PAF = 3.083% [95% CI: 0.964% to 5.202%]
#> standard_deviation(paf %) = 1.081
#> 
#> $Asbestus
#> 
#> ── Population Attributable Fraction: [Asbestus_adj] ──
#> 
#> PAF = 2.136% [95% CI: 1.294% to 2.978%]
#> standard_deviation(paf %) = 0.429

which returns a named list of the adjusted fractions.

Learn More

Contributing

Contributions are welcome! Please file issues and pull requests on GitHub.