The purpose of this study was to analyze and compare the recorded content of two experimental nursing interventions used in a hypertensive-education program. One intervention was characterized as deliberative or high indirect, the other as task-centered or low indirect. A total of 4,149 nurse-patient interaction units was derived from 62 typed interview transcripts, equally representative of both approaches. Two trained coders categorized the interactions, based upon recurring themes of the teaching content; 24 distinct topical categories were identified. Intercoder agreement above chance for all 4,149 interaction units was k = .68, with confidence limits of 99% +/- .03, p less than .0001. A space-time unit for quantifying the relative amounts of time spent on the various categories also was developed, and analyses were conducted for space-time units as well as interaction units. The main findings showed no definable differences in the frequency of categories or the amounts of time spent in their discussion in the two approaches. These results were contrasted to results obtained when the validity of the nursing approaches was tested previously, and differences between the two interventions in attaining social and health goals were interpreted in light of the findings of the content analysis.