Models of change should be broadened so that treatment is seen as a complex system of parts, facilitating a nexus of cognitive, social and behavioural changes, embedded within a broader system of events and processes catalysing change. Such a model helps explain the relative absence of between-treatments outcome differences in UKATT and in the alcohol problems treatment and more general psychotherapy research literatures.

ObjectiveThis paper investigates the experiences of women using a social media Facebook group for breastfeeding support and attempts to explore whether it has aided in extended breastfeeding success. In addition, it aims to explore the value of social cognitive theory (SCT) in explaining these experiences.DesignExploratory, deductive approach.MethodsQualitative, audio‐recorded, semi‐structured interviews were conducted from eight women who were members of a private Facebook group'. Data were analysed using theoretical thematic analysis, and SCT was used as a theoretical lens through which the data were interrogated.ResultsA number of themes were identified from the women's experiences, with analysis uncovering the superordinate theme ‘increased self‐efficacy' which provided an understanding of how the group impacted women's experiences and aided them in breastfeeding success. The sub‐themes of education, accessibility, online community, normalization, and extended goals provided a more detailed understanding of how self‐efficacy was increased through group membership.ConclusionThe symbiotic relationship between members of a social media group facilitates greater breastfeeding success and a longer duration of breastfeeding through the central concept of the SCT: reciprocal determinism. Therefore, it is posited that the SCT is a suitable theory of behaviour change which can potentially be used to develop interventions aiming to increase breastfeeding rates and duration.

Aim: To develop a model of why people seek professional treatment for drinking problems, grounded in what clients say about the process at treatment entry. Participants: Three sets of consecutive entrants to the UK Alcohol Treatment Trial, sets commencing at intervals during trial recruitment (total n598). Location: Statutory and non-statutory alcohol problem treatment agencies in three areas of England and Wales. Data: Open-ended interviews according to a brief interview guide, leading to 400-800-word postinterview reports used for analysis (tape recordings used for auditing the interview and analysis process). Analysis: Reports analysed by a team according to grounded theory principles, involving an iterative process with successive refinement of interviewing and analysis with each successive set of data. Findings: A model of professional treatment entry was developed, refined and ''tested'' with the last set of data. The process of seeking professional treatment was depicted in the model as involving a realization of worsening, accumulating and multiple problems related to drinking, especially in health and family domains; in conjunction with, in most cases (but not all), a trigger event and/or family or professional influence; combined with rejection of the possibility of unaided change or nonprofessional help; leading to the seeking or accepting of professional help. Conclusions: The findings support conclusions already in the literature about the process of seeking professional help for a drinking problem, but provide further refinement of existing ideas: for example regarding the accumulation of drinking-related problems, the ways in which a realization of those

Understanding the educational needs of staff can underpin the development of appropriate educational interventions. These modules require evaluation to determine their impact on health professionals' knowledge and confidence.

SummaryManual transcription of numerical data is prone to error. We quantified the transcription error rate for blood results recorded in a critical care setting by comparing the handwritten and printed laboratory results in 100 consecutive patients in the intensive care unit, Glasgow Royal Infirmary. Nine hundred and fifty-four sets of results with 4664 individual values were analysed. There was complete and accurate transcription in 67.6% (n = 645) of cases, a failure to transcribe in 23.6% (n = 225) and inaccurate transcription in 8.8% (n = 84). Transcription was significantly more accurate in the morning (p = 0.02). This study highlights that our current system of recording blood results is unreliable. These results strengthen the case for computerisation of the patient record in terms of data retrieval and transcription accuracy.