Background Patient experience surveys (PESs) often include informative free-text comments, but with no way of systematically, efficiently and usefully analysing and reporting these. The National Cancer Patient Experience Survey (CPES), used to model the approach reported here, generates > 70,000 free-text comments annually. Main aim To improve the use and usefulness of PES free-text comments in driving health service changes that improve the patient experience. Secondary aims (1) To structure CPES free-text comments using rule-based information retrieval (IR) (‘text engineering’), drawing on health-care domain-specific gazetteers of terms, with in-built transferability to other surveys and conditions; (2) to display the results usefully for health-care professionals, in a digital toolkit dashboard display that drills down to the original free text; (3) to explore the usefulness of interdisciplinary mixed stakeholder co-design and consensus-forming approaches in technology development, ensuring that outputs have meaning for all; and (4) to explore the usefulness of Normalisation Process Theory (NPT) in structuring outputs for implementation and sustainability. Design A scoping review, rapid review and surveys with stakeholders in health care (patients, carers, health-care providers, commissioners, policy-makers and charities) explored clinical dashboard design/patient experience themes. The findings informed the rules for the draft rule-based IR [developed using half of the 2013 Wales CPES (WCPES) data set] and prototype toolkit dashboards summarising PES data. These were refined following mixed stakeholder, concept-mapping workshops and interviews, which were structured to enable consensus-forming ‘co-design’ work. IR validation used the second half of the WCPES, with comparison against its manual analysis; transferability was tested using further health-care data sets. A discrete choice experiment (DCE) explored which toolkit features were preferred by health-care professionals, with a simple cost–benefit analysis. Structured walk-throughs with NHS managers in Wessex, London and Leeds explored usability and general implementation into practice. Key outcomes A taxonomy of ranked PES themes, a checklist of key features recommended for digital clinical toolkits, rule-based IR validation and transferability scores, usability, and goal-oriented, cost–benefit and marketability results. The secondary outputs were a survey, scoping and rapid review findings, and concordance and discordance between stakeholders and methods. Results (1) The surveys, rapid review and workshops showed that stakeholders differed in their understandings of the patient experience and priorities for change, but that they reached consensus on a shortlist of 19 themes; six were considered to be core; (2) the scoping review and one survey explored the clinical toolkit design, emphasising that such toolkits should be quick and easy to use, and embedded in workflows; the workshop discussions, the DCE and the walk-throughs confirmed this and foregrounded other features to form the toolkit design checklist; and (3) the rule-based IR, developed using noun and verb phrases and lookup gazetteers, was 86% accurate on the WCPES, but needs modification to improve this and to be accurate with other data sets. The DCE and the walk-through suggest that the toolkit would be well accepted, with a favourable cost–benefit ratio, if implemented into practice with appropriate infrastructure support. Limitations Small participant numbers and sampling bias across component studies. The scoping review studies mostly used top-down approaches and focused on professional dashboards. The rapid review of themes had limited scope, with no second reviewer. The IR needs further refinement, especially for transferability. New governance restrictions further limit immediate use. Conclusions Using a multidisciplinary, mixed stakeholder, use of co-design, proof of concept was shown for an automated display of patient experience free-text comments in a way that could drive health-care improvements in real time. The approach is easily modified for transferable application. Future work Further exploration is needed of implementation into practice, transferable uses and technology development co-design approaches. Funding The National Institute for Health Research Health Services and Delivery Research programme.

ObjectivesTo determine communication strategies associated with smoking cessation in the National Health Service community pharmacy Stop Smoking programme.Setting11 community pharmacies in three inner east London boroughs.Participants9 stop smoking advisers and 16 pairs of smokers who either quit or did not quit at 4 weeks, matched on gender, ethnicity, age and smoking intensity.Method1–3 audio-recorded consultations between an adviser and each pair member over 5–6 weeks were analysed using a mixed-method approach. First a content analysis was based on deductive coding drawn from a theme-oriented discourse analysis approach and the Roter Interaction Analysis System. Core themes were identified through this quantification to explore in detail the qualitative differences and similarities between quitters and non-quitters.ResultsQuantitative analysis revealed advisers used a core set of counselling strategies that privileged the ‘voice of medicine’ and often omitted explicit motivational interviewing. Smokers tended to quit when these core strategies were augmented by supportive talk, clear permission for smokers to seek additional support from the adviser between consultations, encouragement for smokers to use willpower. The thematic analysis highlighted the choices made by advisers as to which strategies to adopt and the impacts on smokers. The first theme ‘Negotiating the smoker–adviser relationship’ referred to adviser judgements about the likelihood the smoker would quit. The second theme, ‘Roles of the adviser and smoker in the quit attempt’, focused on advisers’ counselling strategies, while the third theme, ‘Smoker and adviser misalignment on reasons for smoking, relapsing and quitting’, concerned inconsistencies in the implementation of National Centre for Smoking Cessation and Training recommendations.DiscussionAdvisers in community pharmacies should use the advantages of their familiarity with smokers to ensure appropriate delivery of patient-centred counselling strategies and reflect on the impact on their counselling of early judgements of smoker success.