Healthcare in many countries is facing growing demands from an ageing population, in parallel with decreasing resources. In order to optimize healthcare effectiveness, evidence-based practice has been proposed as a strategy for incorporating new and adequate knowledge into practice [1–3]. The application of the best available evidence in the care of individual patients is in fact crucial for all healthcare organizations to deliver quality care. Evidence-based practice includes consciously applying the best available evidence based on research findings, clinical experience, and patient preferences . Because nurses are the largest group of healthcare practitioners, their contribution to evidence-based practice is pivotal . It is, therefore, disquieting that studies indicate that nurses' use of research in their clinical practice varies considerably [5–7]. Additionally, little is known about the transition from education to practice and how nurses' capacity for making use of research findings in clinical practice develops over time. Thus, this study investigated the extent of Swedish nurses' perceived use of research over the first five years of their professional career.
In order to enhance the readiness for a changing healthcare system, Swedish nursing education has shifted from vocational training to a three-year bachelor's degree program, in parallel with developments in many other countries. This shift involves an academic perspective in both content and educational methods, including courses in research methods, a bachelor's thesis, and a move towards more self-directed learning. Newly graduated nurses are thereby expected to posses the skills underpinning evidence-based practice, namely questioning prevailing practices, searching, critically appraising, and using scientific knowledge in clinical practice . However, research studies and a national audit of higher education show that the academic perspective in education is still deficient and nursing students perceive a gap between the academic and clinical elements of their education [9, 10]. These flaws in undergraduate education may impact newly graduated nurses' capacity to base their practice on research findings.
Research utilization is a prominent facet of the concept of evidence-based practice. A definition was proposed by Estabrooks and colleagues: 'that process by which specific research-based knowledge (science) is implemented in practice' . Research utilization has been conceptualized to comprise instrumental, conceptual, and persuasive use of research. The concepts were developed in social science, e.g., by Larsen, who proposed that knowledge utilization could be classified as instrumental and conceptual . Beyer and Trice added symbolic (persuasive) utilization . These three research utilization concepts were presented to nursing researchers through the work of Stetler [14, 15]. Estabrooks continued this work by developing definitions and measures and assessing the constructs in nursing . Instrumental utilization refers to the concrete application of research to practice in making specific decisions/interventions; conceptual utilization refers to a change in thinking in response to research, but not necessarily in behaviour (an informing and enlightening use); and persuasive utilization refers to the use of specific research findings to convince others.
In a recent systematic review of instruments for measuring research use, Estabrooks' operationalization of instrumental, conceptual, and persuasive use is called 'kinds of research use' . In another systematic review on extent of nurses' use of research, Squires et
al. report five surveys using 'kinds of research use' as a measure . Research use ranged from moderate-low to high depending on the kind of research use, on average these results represented 5 (using research on half of the shifts) on a 7-point frequency response scale. In Estabrooks' index study, a survey including Canadian staff nurses, scores were highest for conceptual use, followed by instrumental and persuasive use . Profetto-McGrath et
al. found the same distribution of results among Canadian nurses in adult surgical and paediatric care . Kenny, who investigated nurses in US Army hospitals, also reported a comparable distribution . Milner and co-workers studied research use among Canadian staff nurses, educators and managers, and reported-in line with the other studies-that conceptual use of research was most frequent, with persuasive being least frequent . In contrast, in two recent Swedish studies, including two national samples of newly graduated nurses (one to three years after graduation), instrumental research use was most common, followed by conceptual and persuasive use [6, 21]. Aside from one of the Swedish studies  that had a one-year longitudinal approach (two measurement waves), we have not been able to identify any study using the 'kinds of research use' measures in a longitudinal study. Looking at studies using other measurement tools for research use, there appear to be a few intervention studies using pre- and post-measurement designs, thus not presenting true longitudinal data having more than two measurement occasions .
As in many other developed countries, Swedish healthcare is challenged by increasing demands on the healthcare system and simultaneously being allocated fewer resources. The number of hospital beds has decreased by 21% during the period from 1999 to 2008. Today, Swedish healthcare has the fewest hospital beds per inhabitant compared to other countries within the OECD (Organization for Economic Co-operation and Development). This has resulted in an estimated number of occupied beds in medical wards of 100% to 105% , i.e., overcrowding due to limited hospital bed capacity. Newly graduated nurses predominantly work in hospital settings , and are thus exposed to a busy work environment including seriously ill patients, often with multiple diseases and short lengths of stay. Also, the job turnover of nurses is high in many settings , creating a situation where the novice nurse all too soon can become 'the most experienced' at the workplace, adding to a demanding work context. It has been suggested that circumstances during early work life, such as work-related stress and a lack of experiential knowledge, hinder the provision of evidence-based practice [25, 26]. The challenge that faces newly graduated nurses in clinical practice has been described as a 'reality shock'  or, more recently, as a 'transition shock' [28, 29]. According to the authors advocating the existence of a transition shock, new nurses are confronted with the hierarchical hospital system, characterized by dominant normative behaviours, described as prescriptive, intellectually oppressive, and cognitively restrictive . Previously adopted school values come into conflict with work life values, and skills such as the critical appraisal of current practices and openness to new knowledge may therefore be difficult to maintain.
In conclusion, newly graduated nurses are faced with a challenging work environment that may affect their ability to apply evidence-based practice. However, little is known about the trajectory of nurses' research use during the first years of clinical practice. Therefore, the aim of this study was to prospectively examine the extent of nurses' use of research during the first five years after undergraduate training and to specifically assess changes over time.