Let’s say you have an interest in measuring the connection between kind 2 diabetes mellitus (T2DM) and despair. In lots of circumstances, one would use digital well being information knowledge and conduct a logistic regression with despair because the dependent variable and T2DM (doubtlessly together with demographics and different comorbidities) because the impartial variables. Nevertheless, using EHR is doubtlessly problematic. As famous in Goldstein et al. (2016), there’s a risked of “knowledgeable presence” because the pattern of sufferers in EHR possible differs from these in most of the people since people solely seem once they have a medical encounter.
Particularly, Goldstein and co-authors notice that extra frequent visits enhance the possibility of being recognized with a illness:
Quan et al. assessed sensitivities primarily based on Worldwide Classification of Illnesses, Ninth Revision, codes throughout 32 widespread circumstances. They discovered that sensitivities for prevalence of a situation ranged from 9.3% (weight reduction) to 83.1% (metastatic most cancers). Diabetes with problems, for instance, has a sensitivity of 63.6%. Subsequently, the extra medical encounters somebody has, the extra possible that the presence of diabetes might be detected.
On the similar time, whereas extra encounters cut back the chance of a false detrimental, in addition they enhance the chance of a false optimistic on account of rule-out diagnoses.
Since phenotype algorithms are usually designed to detect the prevalence of a situation through ever/by no means algorithms (you both have the situation otherwise you don’t), the extra health-care encounters somebody has the upper the likelihood of a false-positive analysis.
Two sorts of bias might come up:
- Bias to variety of doctor visits. Determine 1A from this paper reveals that the variety of encounters could also be a confounding issue. It’s not proof, nevertheless, whether or not there’s potential for M bias, bias from conditioning on a collider. A collider is a variable that’s an consequence of two different variables.
- Bias on account of common sickness. The authors notice {that a} common sickness could also be the reason for each diabetes and despair. As an illustration, maybe somebody sustained an harm which make them train much less and eat much less wholesome (inflicting T2DM) and the harm itself additionally elevated despair. Whereas my instance offers a selected harm, the “common sickness” within the Goldstein paper might or will not be absolutely captured or identified. Thus, the authors declare that the variety of encounters could possibly function a proxy for common sickness.
In brief, the authors argue that controlling for variety of visits might be helpful for (i) controlling for the truth that analysis is correlated with variety of encounters and (ii) variety of encounters could also be a proxy for common sickness.
The authors then conduct a simulation train utilizing EHR knowledge from the Duke College Well being System. The authors conduct 4 analyses analyzing the connection between consequence and publicity controlling for: (i) demographics solely, (ii) medical encounters, (iii) Charlson Comorbidity Index (CCI), and (iv) medical encounters and CCI.
The authors summarize their findings as follows:
If the presence of a medical situation will not be captured with excessive likelihood (i.e., excessive sensitivity), there’s the potential for inflation of the impact estimate for affiliation with one other such situation. This potential for bias is exacerbated when the medical situation additionally results in extra affected person encounters…Principle suggests, and our simulations affirm, that conditioning on the variety of health-care encounters can take away this bias. The affect of conditioning is biggest for diagnoses captured with low sensitivity.
The authors notice that whereas there’s some concern of M bias–because the variety of encounters is probably going a collider–M bias is probably going considerably much less problematic than confounding bias most often. Others research (Liu et al. 2012) have confirmed that M-bias is usually smaller than confounder bias.
An apart: Berkson’s Bias
The issue of sicker sufferers showing in EHR knowledge causes a manifestation of Berkson’s bias:
As a result of samples are taken from a hospital in-patient inhabitants, reasonably than from most of the people, this may end up in a spurious detrimental affiliation between the illness and the chance issue. For instance, if the chance issue is diabetes and the illness is cholecystitis, a hospital affected person with out diabetes is extra prone to have cholecystitis than a member of the overall inhabitants, for the reason that affected person will need to have had some non-diabetes (probably cholecystitis-causing) purpose to enter the hospital within the first place. That end result might be obtained no matter whether or not there’s any affiliation between diabetes and cholecystitis within the common inhabitants.
