Table 2.

Summary of the analytic process.

Thematic analysis phase	Implementation	Supplementary techniques derived from grounded theory
Familiarization	Initial notes made as transcripts read repeatedly	Listen to audio recordings
Generate initial codes	Using the first 22 transcripts, initial codes and a coding manual were developed. This coding manual was used to analyze subsequent transcripts, and amendments were made iteratively when necessary	Line-by-line open coding on a portion of the data and constant comparison
Searching for themes	As the analysis evolved, codes related to similar manifest or latent concepts were grouped together. These groupings were considered as candidate themes and subthemes	Constant comparison, identify key concepts in the data, and write memos
Reviewing themes	Candidate themes and subthemes were reviewed to ensure that they worked in relation to the coded extracts and the individual interviews and that they captured relevant material from across the data set	Constant comparison, search for deviant cases, generate selected case summaries to capture participant stories, and changes across 3- and 12-month interviews
Defining and naming themes and their interrelations	Themes were refined and explicitly defined to clearly and succinctly capture patterns in the data relevant to the research objectives Cross-tabulations (using NVivo’s matrix query) to compare theme content and relevance between the TCBTa and WCBTb groups and between 3 and 12 months	Constant comparison
Reporting	Selected compelling examples to illustrate themes and subthemes. Final analysis and contextualization in relation to the literature and research objectives	N/Ac

^aTCBT: telephone-based cognitive behavioral theory.

^bWBCT: web-based cognitive behavioral theory.

^cN/A: not applicable.