We appreciate the thoughtful comments from Dr. Madias. We agree that digital measurements of QRS duration provide greater precision than physician measurements, although the former were unavailable for all but a small minority of participants in our study sample. Participants with shorter QRS durations at the earliest examination were more likely than those with longer QRS durations to experience lengthening of the QRS duration over time; however, the rate of progressing from the lowest (QRS duration <100 ms) or the middle category (QRS duration 100 to <120 ms) to a higher category over the entire follow-up period was 3.7% and 5.5%, respectively. Because rates of progression to a higher QRS category were relatively low, an analysis of its relation to outcomes would have limited power in our sample even if performed over the entire follow-up period. We agree that presence of fascicular block and degree of QRS axis could also be contributors to risk for permanent pacemaker; although coding of these variables was not available in our study sample, this could be the subject of future research. As pointed out, variation in heart rate is a likely precursor to sick sinus rhythm and eventual pacemaker placement. Given the high variability of heart rate as a single clinical measure, we reported the results of secondary analyses that adjusted for PR interval as well as heart rate, whereby PR interval is inversely correlated to heart rate but demonstrates greater temporal stability. We also have previously reported on the independent relation of PR interval with incident pacemaker placement.