Timothy L. Lash

BACKGROUND: The Charlson comorbidity index is often used to control for confounding in research based on medical databases. There are few studies of the accuracy of the codes obtained from these databases. We examined the positive predictive value (PPV) of the ICD-10 diagnostic coding in the Danish National Registry of Patients (NRP) for the 19 Charlson conditions. METHODS: Among all hospitalizations in Northern Denmark between 1 January 1998 and 31 December 2007 with a first-listed diagnosis of a Charlson condition in the NRP, we selected 50 hospital contacts for each condition. We reviewed discharge summaries and medical records to verify the NRP diagnoses, and computed the PPV as the proportion of confirmed diagnoses. RESULTS: A total of 950 records were reviewed. The overall PPV for the 19 Charlson conditions was 98.0% (95% CI; 96.9, 98.8). The PPVs ranged from 82.0% (95% CI; 68.6%, 91.4%) for diabetes with diabetic complications to 100% (one-sided 97.5% CI; 92.9%, 100%) for congestive heart failure, peripheral vascular disease, chronic pulmonary disease, mild and severe liver disease, hemiplegia, renal disease, leukaemia, lymphoma, metastatic tumour, and AIDS. CONCLUSION: The PPV of NRP coding of the Charlson conditions was consistently high.

Abstract In a cohort study, the numerator and denominator of each disease frequency (incidence proportion, incidence rate, or incidence odds) are measured, which requires enumerating the entire population and keeping it under surveillance. A case‐control study observes the population more efficiently by using a sample of the population, which becomes the control series, in place of complete assessment of the denominators of the disease frequencies. This extra sampling step can make a case‐control study much more efficient than a cohort study of the same population, but it introduces a number of subtleties and avenues for bias that are absent in typical cohort studies. For diseases that are sufficiently rare, cohort studies become impractical, and case‐control studies become the only useful alternative. The cases in a case‐control study should be the same people who would be considered cases in a cohort study of the same population. The definition of the source population for these cases determines the population from which controls should be sampled. Ideally, control selection would involve direct sampling from the source population, with equal probability of selection, although in many circumstances, variations in and substitutes for this ideal sampling design may be considered.

Quantitative bias analysis serves several objectives in epidemiological research. First, it provides a quantitative estimate of the direction, magnitude and uncertainty arising from systematic errors. Second, the acts of identifying sources of systematic error, writing down models to quantify them, assigning values to the bias parameters and interpreting the results combat the human tendency towards overconfidence in research results, syntheses and critiques and the inferences that rest upon them. Finally, by suggesting aspects that dominate uncertainty in a particular research result or topic area, bias analysis can guide efficient allocation of sparse research resources. The fundamental methods of bias analyses have been known for decades, and there have been calls for more widespread use for nearly as long. There was a time when some believed that bias analyses were rarely undertaken because the methods were not widely known and because automated computing tools were not readily available to implement the methods. These shortcomings have been largely resolved. We must, therefore, contemplate other barriers to implementation. One possibility is that practitioners avoid the analyses because they lack confidence in the practice of bias analysis. The purpose of this paper is therefore to describe what we view as good practices for applying quantitative bias analysis to epidemiological data, directed towards those familiar with the methods. We focus on answering questions often posed to those of us who advocate incorporation of bias analysis methods into teaching and research. These include the following. When is bias analysis practical and productive? How does one select the biases that ought to be addressed? How does one select a method to model biases? How does one assign values to the parameters of a bias model? How does one present and interpret a bias analysis?. We hope that our guide to good practices for conducting and presenting bias analyses will encourage more widespread use of bias analysis to estimate the potential magnitude and direction of biases, as well as the uncertainty in estimates potentially influenced by the biases.