Impact of specialist follow-up in outpatients with congestive heart failure

Dear CMAJ,

We thank Dr. Finkelstein for his letter. Dr. Finkelstein is appropriately concerned about the inclusion of post-event outcomes in the regression model. Such an incident results in biased associations – namely, one being unable to determine if the predictive factor resulted in the event or if the event resulted in the predictive factor. This has been called “survivor-treatment selection bias”(1) or, more generically, “time-dependent bias” and is relatively common even in highly cited medical journals. In a recent systematic review, we found that 18.6% (95% CI 15.8 to 21.8%) of studies with a survival analysis contained a time- dependent factor and that 40.9% [32.3 to 50.0%] of these studies were susceptible to time-dependent bias. (1)

However, we strongly disagree that our Cox model is done incorrectly, as it corrects for this bias. As we state in the methods section, we adjust for the appropriate time-dependent variables and have a variable expressing ‘time spent in hospital up to that time’. (3) The results section summarizes the findings. The phrase ‘within 1 year after discharge’ stated in the results section refers to the censoring time that we used for all analyses in the study. We do not use ‘future information’ as stated by Dr. Finkelstein, and our methodology is robust.

We state clearly in our Methods section that we performed a sensitivity analysis using all outpatient visits rather than cardiovascular visits to define our groups; however, this analysis was truncated by the journal due to space limitations. Using the same variables as in Table 3 but using all visits rather than cardiovascular visits, we found similar results: compared to those without any outpatient visits, patients seen by a family physician (OR 0.80 [95%CI 0.64 to 0.96]) or a specialist and family physician (OR 0.48 [95%CI 0.40 to 0.58]) had lower mortality rates. Furthermore, similar results were obtained with the Cox model using all visits instead of cardiovascular visits: seeing a specialist (HR 0.95 [95%CI 0.94 to 0.96]) was associated with lower mortality.

Regards,

Justin A. Ezekowitz, MBBCh MSc Carl Van Walraven, MD MSc Finlay A. McAlister, MD MSc Paul W. Armstrong, MD Padma Kaul, PhD

1. Glesby MJ, Hoover DR. Survivor treatment selection bias in observational studies: examples from the AIDS literature Ann Intern Med 1996;124:999–1005.

2. van Walraven C, Davis D, Forster AJ, Wells GA. Time-dependent bias was common in survival analyses published in leading clinical journals. Journal of Clinical Epidemiology 2004; 57: 672-680.

3. Ezekowitz JA, van Walraven C, McAlister FA, Armstrong PA, Kaul P. Impact of specialist follow-up in outpatients with congestive heart failure. CMAJ 2005;172(2):189-94.

Conflict of Interest:

None declared