A systematic review method [25] was used in order to make a thorough critical appraisal of the quality of available evidence. The key characteristics of a systematic review are: 1) a clearly stated set of objectives with an explicit reproducible methodology; 2) a systematic search that attempts to identify all studies that meet the eligibility criteria; 3) an assessment of the validity of the findings of the included studies; and 4) a systematic presentation and synthesis of the characteristics and findings of the included studies ([26] , p. 55).

Two electronic databases, EMBASE and Ovid MEDLINE(R), were used to access information and answer the review question. In February, June and November 2014 we searched for articles using the following search words: capacity building, Nordic countries (Norway, Sweden, Denmark, Finland, Iceland and Scandinavia), nursing research, nursing role, nursing care, leadership, nursing attitude and nursing practice both alone and in combination. The searches were made by a specialized librarian and restricted to articles published between January 2004 and November 2014. The first author (K.L.) discussed the search words and outcomes with the specialized librarian and the co-authors. After the first search the search words were narrowed and changed to: capacity building, nursing research and nursing practice for the second and third search. The search words were used both alone and in combination.

The authors are nurse researchers working in different hospital contexts representing three disciplines (nursing, caring and public health) and three Nordic countries; Finland, Denmark and Norway. Their specialities are nursing leadership, mental health, coping with chronic disease, perioperative nursing, elder care and depression. They are familiar with various designs including implementation science, as well as a range of methods and the use of theoretical frameworks. The process was inspired by PRISMA [26] [27] .

The articles were selected if they fulfilled all of the following inclusion criteria: the subjects of the study were nurses; it described research capacity in nursing care or practice; and was written in the English language. The studies included in this review were quantitative research studies (correlation, quasi-experimental, survey). The exclusion criteria were: duplicates, no information on research capacity building in clinical nursing practice, a context other than hospitals, reviews, collaborative projects between developed and under-developed countries, qualitative and theoretical study designs, education, and published before 2004. The state of knowledge was developed in three steps (Figure 1).

The second author (E.E.S.) invited all the authors to participate in two meetings in order to plan and structure the entire literature review, formulate and discuss the review questions, search for literature, i.e., data collection, and agree on the inclusion and exclusion criteria, in addition to criteria for critically appraising and analysing the findings. References in the selected studies were also scrutinised and discussions held with the co-authors regarding the key search outcomes and references [28] . The five authors reviewed and assessed the titles and abstracts of the articles for inclusion based on the inclusion and exclusion criterion. A total of 4748 titles were identified in the initial search, of which duplicates (n = 217), articles not written in English (n = 4), published before 2004 (n = 21), with a main focus other than capacity building in nursing (n = 62), a context other than hospitals (n = 22), collaborative projects with developing countries (n = 9), reviews (n = 7), or with a qualitative theoretical study design (n = 29) were excluded. Of the remaining 4377 articles, three were included. The second electronic search revealed a total of 2774 titles of which 18 articles were reviewed, 15 excluded and three included. Finally, the third search revealed 2763 titles of which 97 were reviewed, 95 excluded and two included.

The references of the eight included articles were also reviewed, but no additional articles were found. One of the author (E.S.) checked and validated the three search histories in order to achieve an overview and make sure that no article had been missed.

The characteristics of the included articles are reported in Table 1. The following data components were examined in each of the articles: name of author(s), country, aim and research questions, subjects and study setting, theoretical framework or model, measurement instruments, reliability, validity and analysis.

The quality of the eight included articles on nurses’ research capacity was reviewed, evaluated and reflected on in accordance with the Quality Assessment and Validity Tool for Correlation Studies [29] [30] . The quality assessment covered design, sample, measurement and statistical analysis (Table 2). The eight studies were rated as cross-sectional in nature. Six studies used non-probability sampling [31] -[36] , one employed a Web-based questionnaire Survey Monkey [37] and a population of all RNs working in a teaching hospital [38] . Corchon et al. [31] used a control group and a quasi-experimental design. Four studies reported a response rate > 60% [33] - [35] [38] . Three studies employed correlational analysis and had internal consistency > 70 [33] [34] [36] [38] . Seven studies reported using a theoretical or model framework for guidance. Five studies contained demographic characteristics [33] [34] [36] -[38] . Two studies had a sample drawn from more than one site [32] [36] . All studies justified sample size, while two [35] [38] reported that anonymity was assured. However, ethical approval can be considered to include anonymity, even if this is not explicitly stated in the other six studies. All studies reported ethical approval. In addition, the papers were analysed to identify the characteristics of nurses’ research capacity. A thematic analysis was performed by two of the authors (E.S., A.L.H.) by exploring the key findings from each of the studies and coding them under various themes, which were discussed with the co- authors until consensus was achieved.

NGT = Nominal group technique; TFC = Total function count; KAP = Knowledge Attitudes and Practice.

^**All studies were approved by an institutional review board or ethics committee.

In the paper by Brown et al. [36] capacity building was achieved by investigating barriers to the utilisation of research among hospital nurses. Several papers report barriers; nurses were unaware of research [38] ; failure to apply research findings or being hindered from implementing evidence in practice [32] ; lack of resources such as research teams [32] , research training, time allocated to research activities [33] [34] , and the knowledge required to utilize research [33] . Thus, implementing research findings can be hindered due to individual, sociopolitical and knowledge barriers.

In one of the studies it was found that the subjects perceived barriers to research utilisation to a “little extent” and a “moderate extent” ([36] , p. 1947). The main barriers were from the organization subscale, including “the nurse does not have authority to change patient care” and “the nurse is unaware of research”. The lowest ranked barriers were “research conclusions not justified” and “research has methodological inadequacies”. The respondents’ age, years of experience as an RN, Master degree, Doctoral degree, position of nurse manager, clinical nurse supervisor and nurse educators had statistically significant positive correlations with one or more of the subscales. Baccalaureate degrees and staff nurse position showed statistically significant negative correlations [36] . There was a significant positive correlation between knowledge to identify clinical problems and age and years of clinical experience [38] . These results indicate that for most hospital nurses, the barriers to research utilisation measured by the BARRIERS scale have a minimal influence on the implementation of evidence- based practice [32] . The role of nurse researcher was described, where the research facilitators’ responsibilities are to assist clinical staff in the research process from initiation of a project to the final dissemination of the findings [37] . Two papers described aspects of both barriers and facilitators [31] [33] , as well as providing insight into aspects of developing practice based on the best available evidence. Despite the fact that few nurses were engaged in research-based activities, a positive attitude towards research was reported [33] . The nurses were interested in developing knowledge of research skills that enhance patient safety and contribute to continuous quality improvement [33] . Research barriers were related to factors perceived by the nurses as a hindrance to implementing research findings in practice. The nurses employed at a university hospital emphasised lack of designated time and lack of interest as the major research barriers. In addition, lack of knowledge, research supervision and support were perceived as barriers in the clinical setting [33] .

One study by Jamerson and Vermeersch [37] divided the nurse research facilitator’s responsibility into three major areas; the conduct of research (own, and/or others), the development of research capacity, and the building of a research culture within the organization. Learning from experience is one important way of developing research capacity.

The establishment of clinical and academic collaborative networks was reported [33] . A research culture in practice and effective interprofessional collaboration are necessary to improve clinical nurses’ research capacity building. Multidisciplinary collaboration was highlighted as increasing professional communication and leadership [35] . The study by Corchon et al. [31] identified the need to enhance knowledge and research skills by means of integrating research and development into professional practice. A study from Cuba reported several research interventions aimed at increasing nursing research capacity on individual and institutional levels for health systems and services research [32] . The result classified nursing research into three major areas: Patient perceptions, beliefs, knowledge and values; Prevalence and incidence of health problems, indicators of population health status; and evaluation of nursing competencies [32] . Most of the nurses reported that it is necessary to develop strong links on the organizational level including the “D” in R&D (Research & Development). In order to achieve this there are several regional support mechanisms such as; the Clinical Research Support Centre and Nursing Programme Manager [35] , which highlight the need to increase opportunities for mulidisciplinary work for the purpose of R&D [32] [35] . A majority of the nurses wanted to collaborate in areas related to management and organisation of research activities. The main preference appeared to be collaboration with a research supervisor or a research team. Building community partnership as a basis for research and evaluation and conducting literature searches were also perceived as areas requiring improvement. In addition, some of the nurses wanted to strengthen their research skills related to evaluation of implemented research results and qualitative research design. Increased knowledge of the development of clinical guidelines appeared to be important for clinical nurses [33] . A series of small group workshops on research design, implementation, analysis and publication of scientific papers seemed to be an important skill development strategy. Individualized assistance and research colloquia/seminars were emphasised as a means of improving knowledge and skills. The four highest ranked areas of research capacity were considering ethical issues, conducting interviews, knowledge of how to access relevant literature, and keeping a systematic record of the published literature accessed [34] .

Nearly half of the nursing and midwifery group needed funding to develop research capacity at doctoral level as well as postdoctoral training to become researchers. They emphasised the necessity of increasing postdoctoral training opportunities [35] .

In conclusion, the findings of McCance et al. [35] provided insight into the development of expertise that will enable the profession to deliver research programmes. Among the top 12 research activities and development priority areas were the need for a vision, the need to enhance strategic, strong and visible leadership, to identify mechanisms by which research and development activity is integrated in practice, in addition to opportunities for multidisciplinary work. Other skill development strategies included how to incorporate research into practice, promote willingness to engage in research, additional training to conduct research [38] , and how to identify research mentors [31] [33] as well as research capability (research skills, general research knowledge and attitudes towards research) [31] . How to identify a research area was prioritised by more that half of the nurses, while a third of them wanted to design a research project and receive help to obtain research funding [33] .

There is little doubt about the relationship between clinical nurses’ research capacity building and development of evidence-based practice [39] . However, there is a lack of a shared understanding of the meaning of research capacity building [40] and despite the fact that research capacity building and evidence-based practice are closely linked, few studies have proven this relationship. Evidence based practice is related to patient safety and requires willingness to integrate the best available scientific evidence in daily nursing care [41] [42] . Clinical nurses should be interested in developing knowledge that can enhance patient safety and quality of care, thus research capacity building in clinical nursing should focus on how to design studies that highlight clinical questions and the development better care for patients. Cooke [43] suggested that action oriented and user involvement approaches generate useful research that is useful and close to practice. Several studies have pointed out barriers to research capacity building in clinical nursing and defined the factors that must be improved.

Two of the studies assessed these barriers in an intervention to develop clinical nurses’ research capacity [31] [32] . In addition, in one study a consensus process was organised to develop a regional research strategy for research in clinical nursing and midwifery [35] . These are examples of an important first step in the process of research capacity building. However, there remains a paucity of studies that link research capacity building in clinical nursing to patient outcome and quality of care. Segrott’s [2] review from 2006, summarizing studies from 1999 to 2004 on the challenges and strategies involved in developing research capacity building in clinical nursing, concluded that there is a consensus on the need for building research capacity [2] . This is in accordance with McCance et al. [35] , who established that the basis for capacity building is engagement in research and development activities to generate knowledge. According to Cooke [43] , capacity building takes place on four structural levels; individual (by participation), team (multi- and interprofessional involvement), organizational (infrastructure and support), and research network level (developing structures between and outside health organizations). In nursing practice, clinical nurses’ research capacity building is a question of developing abili- ties, i.e., research competence [22] and knowledge implementation [10] [11] impacting on patient outcomes [18] . Outcome measures of research and knowledge implementation could, according to Jamerson & Veermeersch [37] , be policy or procedural changes, cost savings, improvement in satisfaction, decreases in nursing turnover, as well as effects on specific quality indicators, and nursing effectiveness measures.In addition to developing research competence including academic leadership of research by ensuring skills and confidence (coordination of research programmes), research capacity building is enhanced by intellectual and social capital by means of tracking linkages and collaborations [11] . Besides traditional measurement outcomes (a publication in peer reviewed journals and conference presentations) Cooke [11] , highlights the need to disseminate the social impact of research (impact on the lives of patients, for communities, and quality of services). There is also a need for dissemination of economic outcomes related to product outputs and health gains and/or the cost ef- fectiveness of interventions. Another principle is that research is dependent on infrastructure including structure, processes, funding and fellowship. As a professional outcome, sustainable research capacity building requires the necessary competence to undertake research and apply skills to practice and patient outcomes [11] .

Developing a research culture is an important part of building research capacity. All the included studies have a clinical setting in non-academic institutions, mostly hospitals. Despite the fact that academic employees in the hospitals are active researchers who produce scientific articles, they normally have their academic position in a university and may thus lack the necessary influence to change the research culture in the clinic. However, clinical researchers’ knowledge of research based practice and their ability to supervise research projects can inspire and motivate clinical nurses who are interested in research, thus influencing the culture in their department. This “bottom-up” approach is important for changing the culture, but works slowly [2] . On the other hand, Cooke [1] argued that it is important to develop a research culture that generates research that is useful for practice on the ability of nurses and teams to conduct research. He further argued that capacity building is conducted within a policy context that can nourish or restrict the progress. In addition, research training improves clinical nurses’ research capability, knowledge and skills [31] .

Akerjordet and Severinsson [44] pointed out that leadership is a key component in the development of a research culture. Leaders in university hospitals have traditionally had an academic education, although mostly in medicine. In the same way as nursing leaders have been shown to be essential for the development of an evidence based practice culture [23] , they also seem to be key actors in the establishment a research culture [20] and therefore require an academic education.

The intervention studies included in this review have a clear aim of enhancing clinical nurses’ capacity building in practice by educational and strategic processes, which are important premises for changing the culture. The concept of capacity building contains processes of adaption to change, where the cultural challenge is to promote and make it possible by reducing barriers for nurses in the clinic to work in a more evidence-based manner, thus ensuring that the research culture in nursing is sustainable and continues to develop. In the included studies the nurses seemed willing to engage in research, but lacked the necessary resources. By establishing a nursing research culture the institution places research on the agenda, but systematic work is needed to enhance it.

In his review ten years ago, Segrott ([2] , p. 648) pointed out that we must move from “problem and solutions to process and outcome”. Some of the included studies still focus on barriers and actions that must be taken. Only two articles outline a process of capacity building in nursing research on an educational and strategic level. One article describes a strategic consensus process on a regional level. Educational and strategic processes resulted in increased research knowledge among nurses that was maintained for over a year [31] . Such processes are important as a first step in research capacity building in clinical nursing. However, to contribute to better quality and safety of care they must be integrated in the organization in the form of continuous and stable activities that ensure the sustainability and further development of clinical research projects.

While, activities such as journal clubs and workshops initiated as a project or by the research department can be a good start for research awareness in clinical nursing, a successful outcome depends on a sound, integrated research strategy and leadership support. In the intervention study by Corchon et al. [31] , journal clubs were seen as an effective strategy for promoting research, becoming familiar with research methods, seeing the value of research for practice and selecting a topic that is important for practice. About 50% of the nurses in the clinic had a positive attitude to research and clearly stated their barriers and needs in terms of conducting research. Key factors were knowledge and leadership that clearly prioritizes research by providing the necessary resources to engage in research projects and to put knowledge into practice [44] .