Chronic diseases, such as heart disease, cancer and diabetes, are responsible for about 60% of all deaths globally and 70% of deaths in the United States [1,2] with morbidity and mortality projected to increase both nationally and internationally over the next several decades [1,3]. Physical inactivity, poor diet, tobacco use, alcohol consumption, and other modifiable behavioral risk factors account for a substantial number of these deaths [4,5], allowing ample intervention opportunities through public health programs and policies.

Calls for the use of evidence-based decision making (EBDM) processes to develop chronic disease control and prevention programs come from both academia and practice, including major health organizations such as the World Health Organization and the Centers for Disease Control and Prevention [6-10]. The concept of EBDM in public health has evolved over the past decade and can be summarized as a process that utilizes the best available scientific evidence regarding the effectiveness of various programs or policies and translates that evidence to real world practice by incorporating community-level data, resources, and priorities [11,12].

There is a well-recognized gap between the production of scientific evidence and the use of that evidence in "real world" settings [13-16] (e.g., policy making bodies, health departments). Closing the translation gap is a complicated process, and increasing amounts of literature address this topic, often referred to as "knowledge transfer" or "dissemination and implementation research" [13-16]. The use of EBDM in public health agencies depends on many factors, including the training and experience of the workforce, organizational resources and climate (e.g., funding, buy-in from leadership and elected officials), and the availability, applicability, and dissemination of evidence on a given topic [16-21].

Research is needed to understand the determinants and approaches that will enhance the uptake of EBDM processes in public health agencies. We conducted a two phase research project that aimed to increase the use of chronic disease evidence-based interventions (EBIs) in public health agency settings. In the first phase, 447 state-level chronic disease practitioners across the U.S. completed a survey that assessed the importance, availability, and use of various components of EBDM in chronic disease. Through quantitative and qualitative methods, we examined practitioner's barriers and solutions to improving the use of EBIs in state health departments and assessed gaps in the importance and availability of core chronic disease competencies [22-24].

In phase 2, we conducted in-depth projects in two U.S. states: Kansas and Mississippi. Under the U.S. constitutional doctrine of reserved powers, the states retain enormous authority to protect the public's health [25]. The states shoulder their broad public health responsibilities through work carried out by state and local health agencies. These interventions are primarily focused on chronic disease prevention and control (primary and secondary prevention), not on management of chronic disease. Non-governmental organizations (NGOs) and other community partners also play critical roles in public health, providing health services and implementing interventions and policy changes in a variety of capacities.

There are large variations in the populations these agencies serve, their types of governance, the services they provide, and the education and job functions of their staff [26-30]. Due to this heterogeneity in public health agencies, it is important to have tools to collect localized data that reflect the unique nature of an agency's workforce and community partners [31]. We began phase 2 by developing a brief survey tool to assess baseline capacity for EBDM, seeking to identify specific targets for increasing the dissemination of EBIs in these two states. This article presents methods and findings from the initial phase 2 survey assessment with the goal of encouraging other public health agencies, in the U.S. and across the globe, to assess EBDM in their own workforce.

The Kansas survey was delivered to 391 valid email addresses and received 190 responses, yielding a 49% response rate. Survey responses were nearly evenly split among SHD practitioners (36%), LHD practitioners (33%) and community partners (31%) (Table 1). Over half (55%) had more than 10 years of experience in public health, and 49% indicated that they held a master's or doctoral degree.

The Mississippi survey had a 75% response rate with 72 surveys completed out of the 96 delivered. State practitioners represented 56% of the responses, and the remaining 44% were from district health offices. The majority of respondents (72%) had more than 10 years of public health experience, and over half (56%) held a master's or doctoral degree.

Nearly 80% of respondents were female in both Kansas and Mississippi. Also in both surveys, practitioners at the state level were significantly more likely to hold master's or doctoral degrees than those at the local or district level (Kansas p = 0.03, Mississippi p < 0.01). In Kansas, the largest job categories represented were program managers, administrators or coordinators (48%) and health educators (15%). In Mississippi, over a third of the respondents were nurses (35%) while this group represented less than 7% of Kansas' responses.

In both Kansas and Mississippi, the three biggest gaps between the importance and the availability of competencies necessary for EBDM in chronic disease were: transmitting evidence-based research to policymakers, making decisions based on economic evaluation, and translating evidence-based interventions to "real world" settings (Table 2). In Kansas, mean importance and availability scores were higher for state respondents compared to local respondents, and gaps were larger at the local level than at the state level. Mississippi surveys showed more mixed results, and gap scores were larger at the state level compared to the district level.

Across all four categories, the percentage of Mississippi respondents who agreed that each expected them to use EBDM was higher for all programs compared to chronic disease programs (Table 3). In both categories, the highest percentage of respondents agreed that health department leaders expect them to use EBDM and the lowest percentage was among co-workers. Mississippi participants indicated that their overall top choices of incentive for using EBDM were: trainings and leaders in their organization placing a high priority on EBDM (Table 4). Those in the state office were more likely to prefer high priority among leadership.

The highest rated skill in both surveys was the ability to find data (Kansas mean 7.4, 95%CI 7.1-7.8; Mississippi mean 8.0, 95%CI 7.4-8.5). The ability to use data for public health programming, grant writing or community assessment followed (Kansas mean 7.0, 95%CI 6.7-7.4; Mississippi mean 7.3, 95%CI 6.7-7.9) with developing evidence-based chronic disease programs as the lowest rated skill (Kansas mean 6.3, 95%CI 6.0-6.6; Mississippi mean 6.6, 95%CI 6.2-7.1).

Estimates of the percentage of evidence-based programs among all respondents from health departments were similar between Kansas and Mississippi. Kansas health department employees' mean estimate of the percentage of evidence-based programs in their agency was 65% (95%CI 61-70%). Mississippi survey respondents' overall mean estimate was 67% (95%CI 60-73%). Median estimates for both Kansas and Mississippi were 75%.

Despite increasing calls internationally for the inclusion of EBDM processes in public health programming, policymaking, and strategic planning [6-10,36-41], there is relatively sparse research to assess the workforce's current capacity at the local level. One notable exception is a needs assessments of population health staff conducted in New South Wales, Australia [42,43]. Their studies identified needs for technical support, training, and skills development, particularly among practitioners without master's degrees. The majority (55%) of practitioners recognized the need to increase their own capacity for EBDM. Practitioners indicated that their managers had more positive views than their own on the current promotion of evidence-based practice in population health while colleagues' views were less positive than their own. Based on this needs assessment, a working group identified evidence-based practice competencies [44]. Additionally, recent U.S. public health systems research seeks to better understand the variability in the quality and availability of public health services and to identify approaches that will improve service delivery, including the increased use of EBDM in agency settings [28-30].

As part of our study to increase EBDM capacity in U.S. public health agencies, we developed tailored survey tools to assess baseline EBDM capacity in two U.S. states. The core of our surveys was adopted from a previously testing instrument [22-24] and focused on practitioners' assessments of competencies previously determined to be critical to EBDM in chronic disease [33]. The practice of EBDM requires a broad skill set that includes the analysis and synthesis of evidence, quantitative and qualitative community assessments, and the use of program-planning frameworks [6]. The public health workforce is transdisciplinary by nature, and many who work in the field have no formal training in public health [45-47]. State-level practitioners in phase 1 of our study indicated that a lack of training necessary to conduct EBDM existed among both staff and managers [22]. Continued workforce training and capacity building is necessary, and the use of competencies to guide those efforts is critical for defining educational goals and outcomes [47,48].

Identification of the largest gaps in EBDM competencies within a state or locality provides health department leaders with actionable targets for the improvement of EBDM capacity. The three largest competency gaps in the Kansas and Mississippi surveys were consistent with findings from our national survey of state-level chronic disease practitioners [24] and may translate to other states. Our research shows that practitioners identify important targets for improving EBDM as: 1) communication with policymakers, 2) use of economic evaluation, and 3) translation of research to practice. These are skills that practitioners identify as important, yet unavailable, and such skills can be improved through trainings and technical support [49]. Because all of the competencies included on the survey are considered high or medium priority from previous research [33], agencies may also want to provide trainings for those competencies with low availability scores or may consider the use of incentives or priority-setting to improve competencies with low importance scores. Evidence-based public health trainings, based on these key EBDM competencies, have been found to be effective methods of integrating new knowledge and skills into the public health workforce [49-51]. As part of our research project, EBDM training courses were conducted in both Kansas and Mississippi to address gaps in competencies. While not attempted in our project, the use of knowledge brokers in Canada is another emerging and promising strategy for facilitating the translation of research to practice [52].

Practitioners in our surveys estimated that approximately two-thirds of programs in their agency were evidence-based. Mean estimates from Kansas (65%) and Mississippi (67%) were consistent with the 58% and 65% estimates obtained in follow-up surveys of EBDM training courses offered to public health professionals in Missouri and nationwide [49, unpublished data, Brownson]. Survey respondents were provided with a standard definition of EBDM before answering this question, but the results should still be interpreted with caution given they are self-reported and not objectively validated. In our qualitative results from phase 1 of this study, chronic disease practitioners identified a lack of consensus among practitioners regarding the precise meaning of the term 'evidence-based' as a barrier to the practice of EBDM [22]. The same program may be deemed 'evidence-based' by one practitioner and not another, and more objective measures are needed. A next logical step in this work is to compare self-reported data (e.g., on use of evidence-based interventions) with program reports (e.g., content analysis of grant applications).

New questions on the Mississippi survey provided results worthy of inclusion in subsequent surveys. Although sample sizes were relatively small, the expectation to use EBDM was lower for chronic disease programs compared to all programs, and the expectation from health department leaders nearly doubled that of co-workers. Creating a culture of EBDM in chronic disease control and prevention that encompasses all job types and levels of management will be an important step in increasing the use of EBIs [6]. Practitioners in our nationwide survey identified a lack of incentives for using EBDM as the highest of nine quantitatively measured barriers [23], and the Mississippi survey explored preferences for a range of incentives. Among Mississippi's customized list of incentives, respondents preferred leaders placing a "high priority" on EBDM and the provision of EBDM trainings. Leadership buy-in is a critical first step in order for practitioners to be able to utilize the knowledge and skills gained from EBDM trainings.

This survey's biggest limitation was that data were self-reported. We cannot directly validate our findings against a gold standard. Furthermore, the response rate in both states was low, and non-response bias is possible. Nearly half (51%) of Kansas recipients (a more diverse sample including community partners) and 25% of Mississippi recipients did not complete the survey. People with strong opinions on EBDM, either positive or negative, may have been more likely to respond. Data were not available to compare respondents with non-respondents across demographic characteristics. While this survey was created with ease of replication in mind, agencies with limited funds will not be able to offer gift cards incentives to increase response rates. Incentives for survey completion can take many forms, and agencies should use available resources.

Top competency gaps in Kansas and Mississippi reinforced findings from our previous nationwide survey [24], indicating that, overall, practitioners need more training and tools for transmitting research to policymakers, making decisions based on economic evaluations, and translating EBIs to "real world" settings. Using our survey tool, health departments and NGOs can assess the unique EBDM capacity within their own workforce and use the localized survey findings to identify specific action points that will strengthen their EBDM capacity. These can include training programs focused on specific EBDM skills or can focus on incentives and policies that could affect the organizational culture and climate in a workplace [53]. EBDM is being advocated in many countries and by many health organizations. Our survey methods should be useful across numerous parts of the globe for assessing EBDM capacity and identifying approaches that will enhance the EBDM processes in public health programming and policymaking.

The authors declare that they have no competing interests.

Study concept and design: EAB, ADD, RCB. Development of survey tool and data interpretation: all authors. Data collection: JAJ, SW. Manuscript drafting: JAJ. Study supervision: RCB. All authors read and approved the final manuscript.

The pre-publication history for this paper can be accessed here:

http://www.biomedcentral.com/1472-6963/12/57/prepub