TY - JOUR T1 - Measurement of harms in community care: a qualitative study of use of the NHS Safety Thermometer JF - BMJ Quality & Safety JO - BMJ Qual Saf SP - 625 LP - 632 DO - 10.1136/bmjqs-2017-006970 VL - 27 IS - 8 AU - Liz Brewster AU - Carolyn Tarrant AU - Janet Willars AU - Natalie Armstrong Y1 - 2018/08/01 UR - http://qualitysafety.bmj.com/content/27/8/625.abstract N2 - Objectives Measurement is a vital part of improvement work. While it is known that the context of improvement work influences its success, less is known about how context affects measurement of underlying harms. We sought to explore the use of a harm measurement tool, the NHS Safety Thermometer (NHS-ST), designed for use across diverse healthcare settings in the particular context of community care.Methods This is a qualitative study of 19 National Health Service (NHS) organisations, 7 of which had community service provision. We conducted ethnographic observations of practice and interviews with front-line nursing and senior staff. Analysis was based on the constant comparison method.Results Measurement in community settings presents distinct challenges, calling into question the extent to which measures can be easily transferred. The NHS-ST was seen as more appropriate for acute care, not least because community nurses did not have the same access to information. Data collection requirements were in tension with maintaining a relationship of trust with patients. The aim to collect data across care settings acted to undermine perceptions of the representativeness of community data. Although the tool was designed to measure preventable harms, care providers questioned their preventability within a community setting. Different harms were seen as priorities for measurement and improvement within community settings.Conclusions Measurement tools are experienced by healthcare staff as socially situated. In the community setting, there are distinct challenges to improving care quality not experienced in the acute sector. Strategies to measure harms, and use of any resulting data for improvement work, need to be cognisant of the complexity of an environment where healthcare staff often have little opportunity to monitor and influence patients. ER -