adjust_hr_with_binary

Adjust an observed hazard ratio with a binary confounder

The strength of evidence provided by epidemiological and
observational studies is inherently limited by the potential for
unmeasured confounding.  We focus on three key quantities: the
observed bound of the confidence interval closest to the null, the
relationship between an unmeasured confounder and the outcome, for
example a plausible residual effect size for an unmeasured continuous
or binary confounder, and the relationship between an unmeasured
confounder and the exposure, for example a realistic mean difference
or prevalence difference for this hypothetical confounder between
exposure groups. Building on the methods put forth by Cornfield et al.
(1959), Bross (1966), Schlesselman (1978), Rosenbaum & Rubin (1983),
Lin et al. (1998), Lash et al. (2009), Rosenbaum (1986), Cinelli &
Hazlett (2020), VanderWeele & Ding (2017), and Ding & VanderWeele
(2016), we can use these quantities to assess how an unmeasured
confounder may tip our result to insignificance.

Lucy D'Agostino McGowan

tipr

Tipping Point Analyses

adjust_hr_with_binary function

<dl><dt>effect_observed</dt>
<dd>Numeric positive value. Observed exposure - outcome hazard ratio.
This can be the point estimate, lower confidence bound, or upper
confidence bound.</dd>
<dt>exposed_confounder_prev</dt>
<dd>Numeric between 0 and 1. Estimated prevalence of the
unmeasured confounder in the exposed population</dd>
<dt>unexposed_confounder_prev</dt>
<dd>Numeric between 0 and 1. Estimated prevalence of the
unmeasured confounder in the unexposed population</dd>
<dt>confounder_outcome_effect</dt>
<dd>Numeric positive value. Estimated relationship
between the unmeasured confounder and the outcome</dd>
<dt>verbose</dt>
<dd>Logical. Indicates whether to print informative message.
Default: <code>TRUE</code></dd>
<dt>hr_correction</dt>
<dd>Logical. Indicates whether to use a correction factor.
The methods used for this function are based on risk ratios. For rare
outcomes, a hazard ratio approximates a risk ratio. For common outcomes,
a correction factor is needed. If you have a common outcome (&gt;15%),
set this to <code>TRUE</code>. Default: <code>FALSE</code>.</dd></dl>

Arguments

Adjust an observed hazard ratio with a binary confounder — adjust_hr_with_binary

<dl>

<dt>effect_observed</dt>
<dd>Numeric positive value. Observed exposure - outcome hazard ratio.
This can be the point estimate, lower confidence bound, or upper
confidence bound.</dd>


<dt>exposed_confounder_prev</dt>
<dd>Numeric between 0 and 1. Estimated prevalence of the
unmeasured confounder in the exposed population</dd>


<dt>unexposed_confounder_prev</dt>
<dd>Numeric between 0 and 1. Estimated prevalence of the
unmeasured confounder in the unexposed population</dd>


<dt>confounder_outcome_effect</dt>
<dd>Numeric positive value. Estimated relationship
between the unmeasured confounder and the outcome</dd>


<dt>verbose</dt>
<dd>Logical. Indicates whether to print informative message.
Default: <code>TRUE</code></dd>


<dt>hr_correction</dt>
<dd>Logical. Indicates whether to use a correction factor.
The methods used for this function are based on risk ratios. For rare
outcomes, a hazard ratio approximates a risk ratio. For common outcomes,
a correction factor is needed. If you have a common outcome (&gt;15%),
set this to <code>TRUE</code>. Default: <code>FALSE</code>.</dd>

</dl>

adjust_hr_with_binary: Adjust an observed hazard ratio with a binary confounder

Description

Usage

Value

Arguments

Examples