Team composition and productivity: evidence from nursing teams in the English National Health Service

We study the impact of team composition on productivity in a setting where team production is particularly important: nursing teams in the English National Health Service (NHS). Composition is measured both in terms of the quantity of staff and the quality of staff, as measured by qualifications, rank and experience. We use a panel dataset that links daily staffing rotas with inpatient mortality records for a single NHS Trust that includes 3 large hospitals and 52 wards. Our results show that the probability of a patient death is lower for teams with a greater number of qualified and senior nursing staff, but find no statistically significant impacts of increased numbers of support staff or agency workers. There are returns to experience for qualified staff, with a lower probability of a patient death in teams where nurses have more experience in the Trust, and returns to both team- and ward (physical location)-specific experience, with a lower probability of death for patients treated by staff who regularly work together and on the ward in question. Our results also provide evidence of the value of bosses, with higher mortality rates when there is an unexpected absence of a senior nurse who leads the team.